Document [original]

In vitro validation of peptide-T3 conjugates

as a treatment option for MCT8-deficiency

via a “Trojan Horse”-like mechanism

vorgelegt von

Dipl.-Ing.

Sarah Paisdzior

geb. in Berlin

von der Fakultät III – Prozesswissenschaften

der Technischen Universität Berlin

zur Erlangung des akademischen Grades

Doktor der Naturwissenschaften

- Dr. rer. nat. –

genehmigte Dissertation

Promotionsausschuss:

Vorsitzender: Prof. Dr. Juri Rappsilber

Gutachter: Prof. Dr. Roland Lauster

Gutachterin: Prof. Dr. Heike Biebermann

Gutachter: Prof. Dr. Jens Kurreck

Tag der wissenschaftlichen Aussprache: 14.12.2018

Berlin 2019

This project was founded by the Deutsche Forschungsgesellschaft (DFG) in the priority

programme Thyroid Trans Act SPP1629: KR 1710/5

Abstract

Cells use transmembrane transporters to take up substances that are important for

functionality and survival of the organism. One example of these substances are

thyroid hormones (THs), tyrosine-based hormones containing iodine that are released

by the thyroid gland. The most important THs, thyroxine (T4) and triiodothyronine (T3)

are transported in and out of cells by several transporters including the very specific

TH transporter monocarboxylate transporter 8 (MCT8). This transporter is expressed

in the central nervous system (CNS) among other tissues. Unfortunately, several

inactivation mutations of the gene SLC16A2 (coding for MCT8 on the X-chromosome)

have been found in mostly male patients. They suffer from severe psychomotor

retardation, known as Allan-Herndon-Dudley syndrome (AHDS), due to lack of THs

during fetal development. Since MCT8 is the crucial transporter for the influx of THs

into the brain, the patients are rendered hypothyroid in the CNS, but hyperthyroid in

the periphery. Treatment options with TH derivates are currently limited, so

development of a new therapy is urgently needed.

Here, we investigate a novel therapeutic mechanism, circumventing transporters

altogether by using the internalization mechanism of G protein-coupled receptors

(GPCRs). This concept, dubbed the “Trojan Horse”-like mechanism, includes the

conjugation of T3 with a peptide ligand. The peptide part of the conjugate can

specifically activate GPCRs, leading to an internalization of the receptor-ligand

complex. Once the conjugate enters the cell, T3 is liberated from a special linker and

can start gene regulation as if it were transported over a transmembrane transporter.

Another advantage of this concept is the target-specific delivery of the conjugated

hormone. The choice of the peptide-receptor pair directs the hormone to the target

tissues, but spares tissues that do not express the specific GPCR.

For this project, two peptide-T3 conjugates are characterized and validated in vitro.

The first, Glucagon-like peptide 1 (GLP1)-T3 is derived from the idea of using the

“Trojan Horse”-like approach for the treatment of metabolic diseases and is used as a

proof-of-principle compound to establish an in vitro toolbox. It could be shown that

GLP1-T3 activates signaling and internalization of the target receptor GLP1 receptor

(GLP1R) as well as canonical and non-canonical TH signaling in the presence, but not

in the absence of the GLP1R. In conclusion, GLP1-T3 acts over the “Trojan Horse”-

like mechanism and does not show any adverse effects in vitro.

The second conjugate, oxytocin (OT)-T3 was designed to target the hypothyroid brain

areas of AHDS patients. It was characterized with the methods established with GLP1-

T3, but showed biased signaling effects on its two target GPCRs, OT receptor (OTR)

and Vasopressin arginine receptor 1A (V1AR). Unfortunately, the results show that

OT-T3 is not stable in culture conditions and was not able to use the “Trojan Horse”-

like mechanism to enter the cell.

This project gave an insight into the “Trojan Horse”-like mechanism in vitro. The

investigations showed how the efficiency of the conjugate is dependent on the size of

the peptide as well as the type of target receptor. Specifically, the structure of the

binding pocket of the GPCR determines whether the signaling was changed in

comparison to the unconjugated peptide. Additionally, the conjugates stability is crucial

for a successful internalization of the conjugated TH. The investigations with the here

established in vitro toolbox were able to identify adverse effects to avoid unnecessary

in vivo assays.

Zusammenfassung

Zellen verwenden Transmembrantransporter, um Substanzen aufzunehmen, die

wichtig sind für ihre Funktion und das Überleben des Organismus. Ein Beispiel für

diese Substanzen sind Schilddrüsenhormone (SDH), Tyrosin-basierende, Iod-

enthaltene Hormone, die über die Schilddrüse abgegeben werden. Die wichtigsten

SDH, L-Thyroxin (T4) und Triiodthyronin (T3) werden über verschiedene Transporter

in und aus Zellen transportiert. Einer dieser Transporter ist der für SDH sehr

spezifische Monocarboxylat-Transporter 8 (MCT8), der unter anderen im zentralen

Nervensystem (ZNS) exprimiert wird. Unglücklicherweise wurden mehrere

inaktivierende Mutationen im Gen SLC16A2 (kodierende für MCT8 auf dem X-

Chromosom) gefunden. Die hauptsächlich männlichen Patienten leiden an einer

schweren psychomotorischen Retardierung, dem Allan-Herndon-Dudley-Syndrom

(AHDS) genannt. Bei diesem Patienten prägt sich der Phänotyp aufgrund fehlenden

SDH schon während der fetalen Entwicklung aus. Da MCT8 ein essentieller

Transporter für den Einstrom von SDH ins Gehirn ist, haben die Patienten eine

Hypothyreose im ZNS, während ihre Peripherie hyperthyreot ist. Behandlungsoptionen

mit SDH-Derivaten ist derzeit limitiert, weshalb die Entwicklung von neuen Therapien

dringend notwendig ist.

In dieser Arbeit wird eine neue Behandlungsidee untersucht, bei der Transporter

umgangen wird durch die Nutzung des Internalisierungs-Mechanismus von G-Protein

gekoppelten Rezeptoren (GPCRs). Dieses Konzept, „Trojan Horse“-ähnlicher

Mechanismus genannt (zu Deutsch: „trojanisches Pferd“) beinhaltet die Konjugation

von T3 mit einem Peptidliganden. Der Peptidanteil des Konjugates kann einen GPCRs

spezifisch aktivieren und so die Internalisierung des gesamten Rezeptors-Liganden-

Komplexes herbeiführen. Ist das Konjugat einmal in der Zelle, kann T3, was über einen

speziellen Linker an das Peptid gebunden ist, freigesetzt werden und Genregulation

ähnlich wie T3 über einen SDH-Transporter starten. Ein weiterer Vorteil dieses

Konzepts ist die zielgerichtete Lieferung des konjugierten Hormons. Die Wahl des

Peptid-Rezeptor-Paars definiert die Gewebe, die das Konjugat anzielt, während

Gewebe, die den spezifischen GPCR nicht exprimieren, unberührt werden.

In diesem Projekt wurden zwei Peptid-T3-Konjugate in vitro charakterisiert und

validiert. Das erste, Glucagon-like peptide 1 (GLP1)-T3 entstand aus der Idee, den

„Trojan Horse“-ähnlichen Mechanismus für die Behandlung von metabolischen

Krankheiten zu verwenden und wurde hier als Grundsatzbeweis sowie für die

Entwicklung von in vitro-Methoden verwendet. Es konnte gezeigt werden, dass GLP1-

T3 Signalisierung und Internalisierung am GLP1-Rezeptor (GLP1R) aktiviert.

Zusätzlich konnte die kanonische und nicht-kanonische SDH-Signalisierung in der

Anwesenheit aber nicht in der Abwesenheit vom GLP1R gemessen werden.

Zusammenfassend konnte gezeigt werden, dass GLP1-T3 über den „Trojan Horse“-

ähnlichen Mechanismus transportiert wird und keine Veränderungen in der

Signalisierung zwischen GLP1 und GLP1-T3 in vitro festgestellt werden konnten. Das

zweite Konjugate, Oxytocin (OT)-T3 wurde anhand der hypothyreoten Gehirnareale

der AHDS-Patienten entwickelt. Es wurde mit den durch GLP1-T3 etablierten

Methoden charakterisiert, zeigte aber funktionelle Selektivität an seinen zwei

Zielrezeptoren, OT-Rezeptor (OTR) und Vasopressin-Arginin-Rezeptor 1A (V1AR).

Unglücklicherweise wurde zusätzlich festgestellt, dass OT-T3 in Kulturbedingungen

nicht stabil und damit nicht in der Lage war, die Zelle über den „Trojan Horse“-

ähnlichen Mechanismus in die Zelle eindringen kann.

Diese Arbeit gibt einen Einblick in den „Trojan Horse“-ähnlichen Mechanismus in vitro.

Die Untersuchungen zeigten, dass die Effizienz des Konjugates von der Größe des

Peptids als auch von der Art des Zielrezeptors abhängig ist. Besonders die

Beschaffenheit der Bindungstasche des GPCRs bestimmt, ob sich die Signalisierung

im Vergleich zum unkonjugierten Peptid verändert. Zusätzlich ist die Stabilität des

Konjugates maßgeblich für die erfolgreiche Internalisierung des konjugierten SDH. Die

Untersuchungen mit der hier etablierten in vitro Toolbox waren in der Lage,

unerwünschte Effekte zu identifizieren, um unnötige in vivo-Analysen zu vermieden.

Acknowledgements

Firstly, I would like to express my sincere gratitude to my advisor Prof. Dr. Heike

Biebermann, IEPE Charité, for the continuous support of my Ph.D. study and related

research, for her patience, motivation, and immense knowledge. Her guidance helped

me in all the time of research and writing of this thesis. I could not have imagined

having a better advisor and mentor for my Ph.D. study.

Besides my advisor, I would like to thank Prof. Dr. Heiko Krude, head of institute who

provided me the opportunity to join his institute, and who gave access to the laboratory

and research facilities. Without his precious support, it would not be possible to conduct

this research. His scientific advice was also greatly appreciated and helped my

scientific growth immensely.

I also want to express my gratitude to the rest of the thesis committee: Prof. Dr. Roland

Lauster und Prof. Dr. Jens Kurreck for giving me the opportunity to finish my doctoral

degree in the year 2018.

Additionally, many thanks go to the DGE-founded priority programme “Thyroid Trans

Act” (SPP 1629) for giving me the possibility to conduct the research project and joining

the German thyroid field to enforce discussions and cooperation.

I am also very grateful for our cooperation partners at the Helmholtz Zentrum in Munich

and the Indiana University in Bloomington, in particular Dr. Timo Müller, Dr. Brian

Finana and Prof. Richard DIMarchi that were able to synthesize two Conjugates that I

could validate as well as giving me the opportunity to spend a week at the Institute for

Diabetes and Obesity in Garching, Munich.

I thank my fellow labmates in the IEPE, former and current, but especially Dipl.-Ing.

Julia Bräunig for the stimulating discussions, for the support throughout the project,

and for all the fun we have had in the last three years. For technical assistance I am

very grateful for Çiğdem Çetindag and Sabine Jyrch as well as Rita Oeltjen.

Also, I thank my friends in the Institute of Prof. Köhrle, IEE, who always supported me

with knowledge and material throughout my project.

Last but not the least, I would like to thank my parents and my friends for supporting

me spiritually throughout writing this thesis and my life in general. This accomplishment

would not have been possible without them. Thank you.

Table of Contents

1Introduction 1

1.1Thyroid hormones 1

1.1.1Biosynthesis of thyroid hormones 1

1.1.2Regulation of TH synthesis via the hypothalamus-pituitary-thyroid

axis (HPT) 2

1.1.3Metabolism of TH 3

1.1.4Diseases of disturbed TH production: Hyper- and hypothyroidism 3

1.1.5Target tissues of THs 4

1.1.6Molecular effects of TH in the cell 6

1.2Thyroid hormone transporters 9

1.2.1Monocarboxylate transporter 8 11

1.2.2Allan-Herndon-Dudley syndrome 12

1.2.3Transgenic models for MCT8 deficiency 14

1.2.4Treatment options for MCT8 deficiency 15

1.3Peptide-hormone conjugates 15

1.4G protein-coupled receptors (GPCRs) 16

1.4.1Structure of GPCRs 16

1.4.2Signaling of GPCRs 17

1.4.3Internalization of GPCRs 18

1.4.4Glucagon-like peptide 1 receptor (GLP1R) 19

1.4.5Oxytocin receptor and vasopressin arginine 1 A receptor 20

1.5The “Trojan Horse”-like mechanism 21

2Aim of the study 23

3Material and methods 24

3.1Materials 24

3.1.1Technical equipment 24

3.1.2Chemicals and consumables 25

3.1.3Kits 25

3.1.4Buffers, reagents 26

3.1.5Stimulation agents/Blockers 27

3.1.6Enzymes 27

3.1.7Expression vectors 28

3.1.8cDNA of proteins of interest 28

3.1.9Fluorescent dyes, gel electrophoresis ladder 28

3.1.10Cell culture reagents and media 29

3.1.11Bacteria culture reagents and media 30

3.1.12Bacterial strains and eukaryotic cell lines 30

3.1.13Computer software 30

3.2Methods 31

3.2.1Molecular biology methods 31

3.2.2Cell culture methods 37

3.2.3Biochemical methods 39

3.2.4Imaging methods 44

3.2.5Cell viability assay 45

3.2.6Statistical analysis 45

4Results 46

4.1Functional characterization of the conjugates at their respective GPCR 46

4.1.1Activation of GLP1R signaling 46

4.1.2Initiation of GLP1R internalization 47

4.1.3Activation of OTR and V1AR 49

4.1.4Internalization of OTR and V1AR 51

4.1.5Summary of the functional characterization 51

4.2Determination of TH-dependent signaling of the conjugates 52

4.2.1Finding a suitable inhibitor for TH transport in HEK293 cells 53

54

4.2.2Determination of TH-dependent signaling after challenge with

GLP1-T3 54

4.2.3Determination of TH-dependent signaling after challenge with OT-T3 56

4.2.4Summary of the TH signaling assay results 57

5Discussion 59

5.1Functional characterization of the conjugates at their respective target

GPCR 59

5.1.1GLP1-T3 activates GLP1R 60

5.1.2GLP1-T3 initiates GLP1R internalization 61

5.1.3Oxytocin-T3 exhibits biased signaling at both target receptors 61

5.1.4Oxytocin-T3 initiates internalization of OTR, but not V1AR 64

5.2TH-dependent signaling of the conjugates 65

5.2.1No tested blocker could inhibit endogenous TH transporters in

HEK293 cells 65

5.2.2GLP1-T3 activates TH-dependent signaling in the presence of GLP1R 66

5.2.3Oxytocin-T3 does not display increased TH-dependent signaling

in the presence of both target receptors 67

5.3Summary of all findings and conclusions 68

5.3.1Summary for GLP1-T3 68

5.3.2Summary for OT-T3 69

5.4Suitability of small peptides as backbone for “Trojan-Horse”-conjugates 70

5.5Suitability of OT-T3 for treatment of MCT8-deficiency 72

5.6Prospects of AHDS treatment 74

6List of references 76

7Supplements 98

7.1Cloning primers 98

7.2Genes of interest (cDNA) 99

7.3Plasmid maps 101

7.4Establishment of the β-arrestin2 recruitment BRET assay 102

8Abbreviations 103

9List of figures 107

10List of tables 108

1 Introduction

The discovery of thyroid hormones (THs) and their numerous functions was over a

century ago and a large amount of information has been accumulated since then, with

new verdicts about TH action being constantly revealed. Nevertheless, many

questions, especially about the pathophysiology of disturbed TH action, are still

unsolved. One example is the correlation between a TH transporter mutation and the

severe Allan-Herndon-Dudley syndrome, a psychomotor retardation known since

1944. It took another 60 years after the definition of the syndrome to find the link

between the disease and the molecular cause, first published in 2004. To the present

day, therapeutic options for this complex disease and its devastating effect on

development are limited. In this work we investigate a new treatment approach in vitro

to reveal possible adverse effects.

1.1 Thyroid hormones

In 1915, Edward Kendall isolated a crystalline form of an iodine containing substance

from the thyroid gland. He found that the compound exerts physiological effects such

as increased pulse rate and metabolism, and named it thyroxine (Kendall 1915). In the

last 100 years, numerous investigations have shown that the thyroid gland produces

and releases several kinds of iodinated hormones that are important metabolic and

developmental regulators (Boelaert and Franklyn 2005).

Thyroid hormones are chemically based on a tyrosine-backbone and are bound to

three or more iodine atoms. The most important are thyroxine (3,3',5,5'-Tetraiod-L-

thyronin, T4), triiodothyronine (T3) and reverse triiodothyronine (rT3).

1.1.1 Biosynthesis of thyroid hormones

The thyroid gland is a butterfly shaped organ close to the trachea in the neck. It consists

of two lobes, which are connected by the isthmus glandularis. It produces THs and

releases them into the blood stream. The thyroid gland is connected to the arterial

blood stream by the superior and inferior artery (Nobori et al. 1994).

THs are produced by a special cell type, the thyrocytes, which are arranged in follicles

surrounded by epithelia cells, called C cells, that produce the hormone calcitonin

(Mundy and Guise 1999). These small spherical structures vary in size and consist of

one-layered thyrocytes and a lumen, called the colloid, and are surrounded by blood

vessels to ensure iodine and TH exchange (Johnson 1955). The colloid contains

thyroglobulin, a larger protein that contains tyrosine residues and is a precursor for

THs. The iodine is actively transported from the blood stream by the cells using a

sodium-iodine-symporter (NIS), a basolateral located ion channel (Dohan et al. 2003)

and diffuses through the apical side by a second ion channel called pendrin (Yoshida

et al. 2002).

By oxidizing the iodine, thyroperoxidase (TPO) catalyzes the attachment of the iodine

ions to the thyroglobulin, mediated by hydrogen peroxide (H

), produced by thyroid

oxidases (THOX1 and THOX2), two enzymes closely associated with the TPO at the

apical membrane of the thyrocytes (De Deken et al. 2000). The attachment of iodine

to thyroglobulin results in monoiodotyrosine (MIT) and diiodothyrosine (DIT), that are

then connected by TPO to form T4 (DIT coupled with DIT) or T3 (MIT coupled with

DIT) (Taurog, Dorris, and Doerge 1996).

To release THs into the blood stream, hormones containing thyroglobulin are taken up

by endocytosis and convey to lysosomes for final processing inside the thyrocytes

(Bernier-Valentin et al. 1990).

1.1.2 Regulation of TH synthesis via the hypothalamus-pituitary-thyroid axis

(HPT)

Since THs are important regulators in metabolism and development, their biosynthesis

is in tight control to ensure balanced TH concentrations. It underlies a neuroendocrine

system that depends on the hypothalamus, the anterior pituitary and the thyroid gland

(Figure 1). The hypothalamus, as part of the brain, includes TH sensing neurons. At

low concentration of circulating T4 and T3, the hypothalamus releases thyrotropin-

releasing hormone (TRH), which can bind to the TRH receptor (TRHR), a G protein-

coupled receptor (GPCR) in the pituitary. The pituitary responds to TRH by producing

thyroid-stimulating hormone (TSH), which then stimulates the thyroid to produce and

secrete THs into the blood stream by binding to the TSH receptor (TSHR), another

GPCR located in the basolateral membrane of thyrocytes. Higher concentrations of

THs exert a negative feedback loop to the pituitary as well as the hypothalamus, to

suppress TRH and TSH expression and therefore stop TH release (Kopp 2001).

Figure 1: The HPT axis is the control system of TH

synthesis. Sensing neurons in the hypothalamus

respond to low TH concentrations by releasing TRH.

Receptors in the pituitary are activated by TRH and

the pituitary secretes TSH. In the thyroid gland TSH,

bind to the TSH receptor. The thyroid starts the

biosynthesis and release of THs, mainly T4 and T3. To

protect from overproduction, high TH concentration

in the blood stream inhibits the release of TSH by the

pituitary and TRH by the hypothalamus as a negative

feedback loop. THs in the blood stream activate

responses in different target tissues. TH: thyroid

hormone; TRH: thyrotropin-releasing hormone; TSH:

thyroid-stimulating hormone; T3: triiodothyronine;

T4: 3,3',5,5'-Tetraiod-L-thyronin

1.1.3 Metabolism of TH

Inside their target cells, the majority of THs need to be metabolized in order to

administrate an effect. Since only a fraction of free THs is the biologically active T3

(20%), the inactive form T4 needs to be metabolized to T3. Specialized enzymes, the

deiodinases catalyze the activation and deactivation of THs by reductive elimination of

iodine. There are three isoforms coded by the DIO genes. Those deiodinases contain

a non-canonical amino acid, a selenocysteine, which is important for enzyme function.

Type II deiodinase (DIO2) is mainly expressed in the brain, here mostly in glia cells, in

brown adipose tissue (BAT), thyme and placenta and only catalyzes the reaction from

T4 to T3 by deiodination of the phenyl ring (5’-deiodination on the outer ring). Type III

(DIO3) only catalyzes by eliminating the iodine from the phenyl ring (5-deiodination,

inner ring) to turn T4 into rT3. It is expressed in the brain, placenta and fetus during

gestation. Type I deiodinase (DIO1) catalyzes both reaction from T4 to T3 as well as

from T4 to rT3 and is mostly expressed in liver, kidney, pituitary and thyroid gland.

Hepatic DIO1 activity controls the serum T3 concentrations (Kohrle 2002). THs can be

broken down into further partly biological active metabolites by other enzymes like

decarboxylases and monoaminoxidases. One example for those is 3,5-

Diiodothyronine, (3,5-T2), which is known to have metabolic effects such as reduction

of body weight and temperature (Pietzner et al. 2015).

1.1.4 Diseases of disturbed TH production: Hyper- and hypothyroidism

Any kind of deregulations of the HPT axis can lead to constantly too high or constantly

too low concentrations of TH in the blood stream. Hyperthyroidism, an excessive

production of THs (thyrotoxicosis) can be the result of an autoimmune inflammation,

e.g. in the case of Grave’s disease (Weetman 2000). On the other hand,

hypothyroidism, low circulating TH concentrations might be the result of a defective

thyroid anlage or dyshormonogenesis (Grüters and Krude 2012).

Several studies showed that passage of maternal THs during pregnancy are important

for brain development of the fetus. An infant exposed to reduced TH concentrations

during gestation from a hypothyroid mother is likely to show decreased cognitive

functions and increased susceptibility to behavior problems later in life (de Escobar,

Obregón, and del Rey 2004). In untreated congenital hypothyroidism (CH), patients

show a short stature, mental retardation as well as obesity. Treatment with L-T4 should

start within the first three month of life to stabilize TH levels and therefore normalize

growth and development (Krude, Kühnen, and Biebermann 2015).

The cases of CH can be classified into different types. Thyroid dysgenesis includes

missing or hypoplastic thyroid gland as well as defects in hormone biosynthesis,

whereas central hypothyroidism is mostly based on defects of the hypothalamus or

pituitary (Gruters, Krude, and Biebermann 2004). Several causes for CH have been

found including mutations in: the transcription factors PAX8 (Macchia et al. 1998),

NKX2.1 (Devriendt et al. 1998), FOXE1 (Clifton-Bligh et al. 1998) as well as in genes

of TSHR (Duprez et al. 1994), TPO (Abramowicz et al. 1992), thyroglobulin (Ieiri et al.

1991), NIS (Fujiwara et al. 1997), the two thyroid oxidases THOX1 and THOX2

(Moreno et al. 2002), the β-subunit of TSH (TSHβ) (Hayashizaki et al. 1989) and the

guanine nucleotide binding protein α-subunit (GNAS) (Patten, Smallwood, and Eil

1989). Another TH deficiency due to a transporter defect will be introduced in 1.2.

1.1.5 Target tissues of THs

After release of THs by the thyroid gland, the hormones reach target tissues throughout

the body. Within the blood stream, THs are bound to various proteins and only a

fraction is unbound. Proteins binding THs are thyroxine-binding globulin (TBG),

albumin, transthyretin and lipoproteins. In a bound state, THs are inactive, but

stabilized and are hindered from unspecific binding to lipid membranes. It also ensures

a fast exchange between bound (inactive) and free (active) T4/T3 in an unbalanced

situation acting as a TH reserve that reacts quicker than the HPT axis (Schussler 2000;

Pappa, Ferrara, and Refetoff 2015). Single cells take up T3 and T4 by TH transporters

(introduced in detail in 1.2).

THs target a broad number of tissues including brain, heart, bone, liver and fat tissue.

The effects of THs on target tissues are summarized in Table 1.

Table 1: Summarizes effects of THs on their target tissues

Target

tissue

THs promote Hypothyroid state Hyperthyroid state References

Brain Cognition and

motor function

Decreased

neuronal function

and development,

impaired

myelination

(during gestation)

Early neuronal

differentiation and

termination of

neuronal

proliferation

(Wirth and Meyer

2017; Pappa,

Ferrara, and

Refetoff 2015;

Nicholson and

Altman 1972)

Bone Growth and

bone turnover

Growth arrest,

malformations

Accelerated

growth, increased

bone turnover

(Bassett and

Williams 2003;

Williams, Robson,

and Shalet 1998;

Kamegai et al. 2001)

Heart Heart rate,

blood pressure

Decreased heart

rate, low blood

pressure

Increased heart

rate, high blood

pressure

(Moolman 2002;

Vargas-Uricoechea,

Bonelo-Perdomo,

and Sierra-Torres

2014)

Liver Energy

expenditure,

weight

regulation,

thermogenesis

Decreased

metabolism,

weight gain,

decreased

thermogenesis

Increased

metabolism, weight

loss, increased

thermogenesis

(Tam, Lecoultre,

and Ravussin 2012;

Mullur, Liu, and

Brent 2014)

1.1.5.1 Brain

In the brain, the neuronal development is especially dependent on optimal TH supply

of the fetus as well as the infant after birth. Since the brain is segregated from the rest

of the blood circulation by the blood-brain barrier (BBB), neurons are arranged in

special functional units, the tripartite synapse together with astrocytes. These cells are

star-shaped glia cells, that provide the neurons with energy substrates,

neurotransmitters and also hormones like THs from the endothelia cells of the BBB

(Wirth and Meyer 2017; Pappa, Ferrara, and Refetoff 2015). Maternal TH

concentrations influence brain development during gestation, especially before the

onset of fetal thyroid function. Maternal hypothyroidism can impact neuronal

proliferation, migration, differentiation and myelination as well as synaptic plasticity and

neurochemistry resulting in impaired motor function and cognition (Moog et al. 2017).

In rats, hyperthyroidism has been shown to promote early neuronal differentiation as

well as early termination of proliferation of granular layer cells in the cerebellum

(Nicholson and Altman 1972).

1.1.5.2 Bone

Next to developmental regulations in the brain, THs are especially important in growth.

In bone, they influence linear growth and skeletal development in infants and maintain

adult bone mass. Patients with untreated congenital hypothyroidism with thyroid

dysgenesis due to athyrosis or hypoplasia, resulting in TH deficiency, develop severe

skeletal defects and growth arrest. An excess of THs can lead to accelerated growth

and bone formation in infants and increased bone remodeling in adults (Bassett and

Williams 2003).

The mechanism for TH action on bone is the direct effect of T3 on the bone resorption.

Both, osteoclasts and osteoblasts have been shown to respond to THs with bone

turnover (Williams, Robson, and Shalet 1998). In addition to the influence on the

pituitary as a negative feedback signal for their own synthesis, THs regulate growth

hormone (GH) synthesis and secretion by influencing the pituitary as well as

hypothalamic functions, shown by hypothyroidism leading to decreased GH secretion

(Kamegai et al. 2001).

1.1.5.3 Heart

For the cardiovascular system, THs regulate the expression of myocardial genes that

are important for calcium administration. These can affect both systolic and diastolic

myocardial function. The hormones have also indirect and direct effects on peripheral

vascular smooth muscle tone and can alter the connection between the left ventricle

and the artery system (Moolman 2002). Hyperthyroidism can lead to an increase in

resting heart rate, blood volume, stroke volume, myocardial contractility, and ejection

fraction, whereas hypothyroidism can cause lower heart rate, weakening of myocardial

contraction and relaxation, with prolonged systolic and early diastolic times (Vargas-

Uricoechea, Bonelo-Perdomo, and Sierra-Torres 2014).

1.1.5.4 Liver and fat

The liver is the main actor in metabolism. Together with fat tissue, it regulates the lipid

homeostasis. THs are known to influence body weight and energy expenditure. In BAT,

containing numerous small lipid droplets and being the main source of non-shivering

thermogenesis (Tam, Lecoultre, and Ravussin 2012), THs induce expression of

uncoupling protein 1 (UCP1) which increases thermogenesis and promotes weight

loss. In liver, THs influence cholesterol and fatty acid metabolism directly and indirectly

(Mullur, Liu, and Brent 2014). Patients with hyperthyroidism show increased resting

energy expenditure, weight loss and decreased cholesterol levels. Reduced TH

concentrations in hypothyroidism show the opposite effect on the metabolism. These

effects are due to T3 action in white adipose tissue (WAT) and BAT. In WAT, a tissue

composing of a single large lipid droplet as storage, T3 induces lipolysis and promotes

weight loss (Mullur, Liu, and Brent 2014).

1.1.6 Molecular effects of TH in the cell

Similar to the depot generation with TH binding proteins in the blood circulation, a

protein inside the cell keeps THs within the cytoplasm. This protein, the µ-crystallin

(CRYM) was identified as a NADPH-dependent cytosolic T3-binding protein in the

kangaroo eye lens (Kim, Gasser, and Wistow 1992; Vie et al. 1997) and has been

shown to be abundant in the human cochlea and vestibular tissue of the inner ear (Abe

et al. 2003).

As the biologically most active TH, T3 is known to have canonical as well as non-

canonical effects within its target cells.

1.1.6.1 Canonical effects of T3

Once transported into the cell and converted into T3, the canonical action of THs is

mediated through TH receptors (TR) belonging to the class of nuclear receptors. TR

are ligand-induced transcription factors, that are activated by binding to T3 and form

homo- or heterodimers, together with other nuclear receptors, especially the retinoid X

receptor (RXR) (Lazar, Berrodin, and Harding 1991), translocate to nucleus and bind

to specific thyroid hormone response elements (TRE) to start gene expression. They

are encoded by the genes THRA on chromosome 17 and THRB on chromosome 3

with either of them having three splicing variants. Most variants, TRα1, TRβ1, TRβ2

and TRβ3, are known to bind to THs (Williams 2000). The splicing variant TRα2 and 3

are not able to bind hormones (Harvey and Williams 2002). The TRα isoform is

important for early development, whereas expression of the TRβ isoform takes place

in later development, being mainly expressed in liver, pituitary, kidney, thyroid, heart,

retina, and several brain areas (Cheng, Leonard, and Davis 2010).

The transcription factors all consist of a single polypeptide chain with a very variable

regulatory A/B domain at the N-terminus as a nuclear localization signal, a DNA-

binding domain in the center and a C-terminal T3-binding domain. The latter two

domains are separated by a hinge region, containing additional localization signals

(Zhang, Roggero, and Allison 2018). The TRβ variants differ in the N-terminal A/B

domain and are very homologous throughout the other two domains (Cheng, Leonard,

and Davis 2010), whereas the TRα variants differ in the length of the C-terminal domain

(Plateroti et al. 2001). Additional to binding hormones, the C-terminal domain is also

involved in interactions with co-regulators, like corepressors or coactivators (Feng et

al. 1998) and heterodimerization with the RXR (Cheng, Leonard, and Davis 2010).

TRE contain two hexameric half-sites with the consensus sequence G/AGGTC/GA and

are arranged in palindromic, inverted palindrome, or direct repeat separated by four

nucleotides (DR4). TRE are located in the promotor region of T3-regulated genes

involved in metabolic, cell proliferation and apoptosis and others (Harvey and Williams

2002). Liganded and free TR compete for these TRE, unliganded TR have even been

shown to act as transcriptional repressors in many studies (Hashimoto et al. 2001;

Chassande 2003; Venero et al. 2005; Wallis et al. 2008; Araki et al. 2009; Kapoor et

al. 2010).

1.1.6.2 TR deficiencies

For both TRα and TRβ, mutations have been known in humans. For TRβ, all known

mutations are located in the T3 binding domain (Sakurai et al. 1989). Patients show a

mostly hyperthyroid state of heart and several brain areas, but can also suffer from

hypothyroid symptoms in the periphery, due to variable tissue expression. The

manifestations of the mutation vary from case to case with the majority having a mild

phenotype (Krude, Kühnen, and Biebermann 2015). Refetoff et al was the first to

describe this TH resistance and the disease was therefore named Refetoff syndrome.

The patient suffered from deaf-mutism and goiter among other symptoms (Refetoff,

DeWind, and DeGroot 1967). Later the group could link these symptoms to an

impairment in the TRβ gene (Refetoff and Dumitrescu 2007).

TRα mutations on the other hand show a more severe outcome for patients. The first

mutation was found very recently in 2012 (Bochukova et al. 2012) and 28 cases have

been identified to date (van Gucht et al. 2017). Like for TRβ, most of the known

mutations are located in the ligand-binding domain. The patients suffer from mild

hypothyroidism with impaired growth, mild to moderate mental retardation, mild

skeletal dysplasia with specific facial features and severe constipations (Tylki-

Szymanska et al. 2015). Table 2 summarizes the symptoms reported for TRα and TRβ

deficiencies.

Table 2: Overview of symptoms of TR deficiencies

Affected thyroid

hormone receptor

Reported symptoms References

TRα Comprise growth retardation,

mild-to-moderate mental

retardation, mild skeletal

dysplasia, severe constipation,

hypothyroidism-like

(Bochukova et al. 2012),

(Schoenmakers et al. 2013),

(Moran et al. 2013), (Moran et

al. 2014), (Tylki-Szymanska et

al. 2015)

TRβ Highly variable phenotype, in

some cases IQ less than 85,

delayed speech development,

attention-deficit hyperactivity

disorder (ADHS), deaf-

mutism, thyroid gland

enhancement or goiter and

short stature

(Refetoff, DeWind, and

DeGroot 1967), (Sakurai et al.

1989), (Brucker-Davis et al.

1995), (Refetoff and

Dumitrescu 2007)

1.1.6.3 Mice studies for TR deficiencies

Several mice studies revealed the function and contributions of the different isoforms

of TRs. Knock-outs (KOs) for the Thra gene displayed behavioral abnormalities due to

reduced GABAergic synaptic density (Guadano-Ferraz et al. 2003) with lower heart

rate and decreased body temperature (Wikström et al. 1998). Mice carrying mutations

found in human patients exhibit impaired growth, heart rate and motor functions, as

well as variable metabolic phenotypes with extremely obese or extremely lean mice

dependent on the mutation (Vennström, Mittag, and Wallis 2008). The KO of the Thrb

gene showed altered maturation of the retina and the inner ear as well as dysfunction

in the HPT axis (Jones et al. 2003), whereas mutations in the gene caused severe

cognitive developmental defects and deafness (Flamant et al. 2006). A double KO

(DKO) of both TR revealed severe developmental defects such as hyperactive HPT-

axis, reduced growth and bone maturation (Göthe et al. 1999). These findings indicate

that the isoforms can substitute each other in some functions, but also retain tissue-

specific action.

1.1.6.4 Non-canonical effects of T3

In addition to the well-known canonical effects, T3 can activate non-canonical

pathways, both dependent and independent of TR. This has been speculated, since

the canonical pathway could not explain the whole phenotype of KO mice models

(Shibusawa et al. 2003) and indicated an admission site close to the plasma membrane

or in the cytoplasm. It has been shown that T3 rapidly regulate membrane potential

and cellular depolarization by modulating plasma membrane ion channels and GPCRs

(Sakaguchi, Cui, and Sen 1996; Sun et al. 2000; Sen, Sakaguchi, and Cui 2002) as

well as mediating glucose uptake (Segal and Ingbar 1990). After that, an involvement

of T3 in the phosphatidylinositol 3-kinase (PI3K)/protein kinase Akt pathway was

discovered, which takes place in the cytoplasm after TH transport into the cell (Davis,

Goglia, and Leonard 2016). In addition, TR can also phosphorylate the Mitogen-

activated protein kinase (MAPK, also called ERK1/2, extracellular signal-regulated

kinase 1/2) and activate the respective pathway (Lin et al. 2003). It is unclear whether

TRα, TRβ or both are capable of this type of non-canonical signaling. There have been

several studies describing the direct inhibition of PI3K by TRβ and even excluded an

involvement of TRα (Storey et al. 2006; Martin et al. 2014). Others claim the opposite

with data that show the activation of the PI3K/Akt pathway by a direct interaction with

TRα, but not TRβ (Hiroi et al. 2006) or the involvement of a membrane-bound

translational variant of TRα (p30), that starts a cascade leading to MAPK and PI3K

activation (Kalyanaraman et al. 2014).

Another TR-independent signaling pathway was revealed when a TH binding domain

was found in a protein of the plasma membrane, the integrin αVβ3 (Bergh et al. 2005).

Binding of the hormones to αVβ3 initiates complex cellular events such as

angiogenesis and cell proliferation transduced by phospholipase C (PLC) and protein

kinase C (PKC) activating MAPK. This kinase then phosphorylates further targets such

as signal transducer and activator of transcription (STAT) 1α, estrogen receptor (ER)-

α and TRβ1 (Davis et al. 2000), starting shedding of coactivators or repressors and

protein trafficking to the cell nucleus (Hammes and Davis 2015). Figure 2 depicts the

possible pathways for TH action within the cell. A recent article proposed a more

precise nomenclature for TH signaling pathways. First, it was suggested that the

terminology is changed from former genomic vs. non-genomic to canonical vs. non-

canonical, due to the word “non-genomic” being misleading. Furthermore, four possible

models of action were considered. Type 1 summarizes the effects known from the

canonical pathway as a TR-dependent signaling with direct binding to DNA, type 2

includes the recent discovery of indirect interaction of THs activated TR with DNA

inducing chromatin remodeling (Grøntved et al. 2015). It is suggested that TR is

interacting with other transcription factors, though further investigation is needed. Type

3 sums up the TR-dependent signaling of THs without DNA binding, such as PI3K and

ERK1/2 activation, and type 4 is suggested to classify TR-independent TR signaling

such as integrin activation and CRYM interactions (Flamant et al. 2017).

1.2 Thyroid hormone transporters

For a long time, it was thought that THs pass the plasma membrane by simple diffusion

due to their small size and lipophilic properties (Hennemann et al. 2001). In the 1970s

however, several studies suggested that membrane transporters facilitate the TH

uptake by amino acid transporter or similar proteins (Stitzer and Jacquez 1975).

Since the first discovery, a number of TH transporters have been identified, cloned and

functionally characterized. They are categorized into different families, the

monocarboxylate transporters (MCTs), the organic anion transporter polypeptides

(OATPs), and the L-type amino acid transporters (LATs). They all differ in their

Figure 2:

Canonical and non-canonical pathways of TH. 1: The canonical pathway of TH action includes

the transport of T3 and T4 via TH transporter. Intracellular deiodinases catalyze the reaction from T4 into

T3 (DIO1 and DIO2). The bioactive T3 binds to TR in the nucleus (here TRβ is exemplified. With its co-

receptor RXR, the TR-T3 complex binds TRE in promotor regions of target genes to activate gene

expression. 2: The TR-T3 complex can also bind to PI3K and the MAPK ERK1/2 to activate the respective

signaling pathway and target gene expression. 3: Additionally, THs are also known to bind to the integrin

αVβ3 and activate PI3K and ERK1/2, which lead to gene expression via the respective signaling cascades.

substrate specificities and tissue distribution. All of them belong to the major facilitator

superfamily (MFS) (Kinne, Schülein, and Krause 2011).

Many OATPs are known to transport different iodothyronines and their derivates, but

also transport several other organic compounds like bile salts and steroid hormones in

a sodium-dependent manner. Some of the members of this superfamily are expressed

only in a specific tissue whereas others are expressed more broadly. The genes coding

for these transporters are grouped into the solute carrier family (SLC) (Hagenbuch and

Meier 2004). The most prominent members concerning TH transport are MCT8 and

OATP1C1; both are known to be expressed in the brain, among other tissues. Notably,

the contribution of the latter in the human brain differs from the murine brain (Mayerl et

al. 2014). In the human brain, OATP1C1 is mainly expressed in astrocytes. Here it is

thought to facilitate the uptake of the inactive T4 to be converted into the active T3 by

the deiodinase 3 (DIO3). In the murine brain, it is highly enriched in the choroid plexus

and cerebral micro vessels, therefore on the blood-brain-barrier (Roberts et al. 2008),

as well as in cortical neurons (Wirth, Roth, Blechschmidt, Hölter, Becker, Racz,

Zimmer, Klopstock, Gailus-Durner, and Fuchs 2009).

Some of the amino acid transporters have been shown to also transport THs due to

tyrosine backbone of iodothyronines. For example, the L-type amino acid transporters

facilitate the uptake of several TH derivates over the cell membrane, such as LAT-1

and LAT-2 (Friesema et al. 2001). They mainly facilitate the exchange of neutral amino

acid between the blood stream and several tissues, such as the small intestine, skeletal

muscle, liver, kidney and brain. They consist of a light and a heavy chain, that are

connected via disulfide bond (Wagner, Lang, and Bröer 2001). The heavy subunit

functions as an escort protein that is required for trafficking to the cell surface, whereas

the light chain facilitates actual substrate transport (Palacín, Errasti-Murugarren, and

Rosell 2016).

The family of MCTs consists of 14 homologous proteins, which are known to transport

monocarboxylates such as pyruvate and lactate. The genes encoding for these

transporters consists of the SLC16 family (Halestrap 2012). MCT8 (SLC16A2) as well

as MCT10 (SLC16A10) have been identified as TH transporters with MCT8 being the

only transporter specifically for TH uptake (Friesema et al. 2003; Friesema et al. 2008).

Despite the structural similarity of both transporters, MCT10 was shown to have a

broader substrate specificity and additionally transports aromatic amino acids

(Halestrap 2012). Therefore, MCT8 is considered a primary TH transporter, whereas

MCT0, some OATPs and the above mentioned LAT1 and LAT2 with a wider substrate

spectrum belong to the secondary TH transporters (Kinne, Schülein, and Krause

2011). Table 3 summarizes the substrate specificity of the mentioned transporters.

Table 3: Substrate specificity of TH transporters, numbers of + are indicating the degree of

affinity

Protein 3,3'-T2 T3 T4 rT3 References

MCT8

+ +++ ++ +

(Friesema et al. 2003;

Friesema et al. 2006)

MCT10 ++ +++ ++ +++

(Friesema et al. 2008)

OATP1C1

+ +++ +++

(Kinne, Schülein, and

Krause 2011;

Zevenbergen 2015)

LAT1/LAT2

+++ ++ +/- +/-

(Friesema et al. 2001;

Kinne et al. 2015)

1.2.1 Monocarboxylate transporter 8

The gene encoding for MCT8 (SLC16A2) is located on the X chromosome position

q13.2 and consist of six exons. When it was first cloned, it was annotated after its N-

terminal domain consisting mainly of proline (P), glutamate (E), serine (S) and

threonine (T) repeats, XPCT for X-linked PEST-containing transporter (Lafrenière,

Carrel, and Willard 1994). After the identification as a specific TH transporter, further

investigations tried to elucidate the gene and protein structure as well as the transport

mechanism.

In contrast to other investigated species with only one defined translational start, the

human MCT8 gene reveals two possible translational start sites, giving two hMCT8

isoforms. The long hMCT8L codes for 613 amino acids, whereas the short isoform

hMCT8S is 74 amino acids shorter (539 amino acids). Since human liver transcripts

have shown the presence of both mRNA species (Friesema, Visser, and Visser 2010),

the purpose of the additional N-terminal domain has been investigated. It has been

shown to enhance proteasomal degradation and is a target of ubiquitin conjugation

(Zwanziger et al. 2016). General MCT8 expression have been shown in several

tissues, such as brain, liver, kidney, heart, adrenal and thyroid gland (Visser, Friesema,

and Visser 2011). The transporter protein consists of 12 transmembrane helices with

both, N- and C-terminus positioned intracellularly and connecting intra- and

extracellular loops (Kinne et al. 2010). MCT8 is known to form homodimers or even

higher order oligomers (Biebermann et al. 2005; Visser, Philp, et al. 2009), which have

been found crucial for the function as substrate transporter (Fischer et al. 2013).

Proteasomal degradation dependent on the expressed isoform can influence the

amount of transporter and subsequently affect the capacity for oligomerization

(Zwanziger et al. 2016).

The same study that showed high affinity of MCT8 to THs (KM = 2-5 µM) could also

prove specific transport over the cell membrane. In contrast to the other known

members of the MCT family, MCT8 does not facilitate the uptake of any other aromatic

amino acid (Friesema et al. 2003). The transport seemed to be sodium- and proton-

independent and is assumed to follow the “rocker-switch” model developed for MFS

transporters (Kleinau et al. 2011). This model theorizes that substrate binding

facilitates conformational changes within the protein. The transporter changes from a

“open to outside” into a “open to inside” conformation and therefore transports the

substrate over the plasma membrane (Schweizer et al. 2014; Sun et al. 2000). The

distinctive recognition of THs by the transporter has been investigated by mutation

studies and comparison to the highly homologous MCT10 based on the homology

models. The following amino acid residues have been shown to change substrate

specificity and are compatible with the homology models: H192 (Braun et al. 2013),

R445 and N498 (Groeneweg et al. 2013), A224 in the substrate channel, S310 in the

substrate pocket and F287 (Johannes, Braun, et al. 2016). A study combining

functional and structural information about MCT8 pointed out two amino residues,

R301 and H415 worth future experimental investigation (Kleinau et al. 2011).

Besides structural investigations, studies have been performed due to the identification

of MCT8 as a major TH transporter in the human brain and several loss-of-function

mutations linked to a severe psychomotor retardation, the Allan-Herndon-Dudley

syndrome (Dumitrescu et al. 2004; Schwartz et al. 2005).

1.2.2 Allan-Herndon-Dudley syndrome

The group of Theo Visser was the first to identify a specific transporter for thyroid

hormones, especially for T3 (Friesema et al. 2003). Later, it was speculated that in

patients with elevated T3 concentration this transporter might be disrupted (Friesema

et al. 2004). The Allan-Herndon-Dudley syndrome (AHDS) has been described as

early as 1944 as an X-linked mental retardation with early onset hypotonia, athetoid

limb movements, motor speech impairments and muscle hypoplasia. Allan, Herndon

and Dudley examined a large family of 7 generations, with 29 affected males all

showing the same symptoms of the disorder (Allan, Herndon, and Dudley 1944). In

1990, Bialer et al found a second family with ADHS, with similar facial features and

severe mental retardation in the United States (Bialer et al. 1992). In the following

years, unrelated families have been identified in Brazil (Passos‐Bueno et al. 1993;

Zorick et al. 2004), Germany (Dumitrescu et al. 2004; Friesema et al. 2004), the

Netherlands (Friesema et al. 2004) and Israel (Gika et al. 2010).

The symptoms could now be linked to single-nucleotide point mutations in the MCT8

gene for investigated subjects. Later other groups identified additional variants in the

MCT8 gene (Dumitrescu et al. 2004). Shortly after, more AHDS patients showed

similar TH concentrations and additional MCT8 mutations could be identified

(Papadimitriou et al. 2008). Since then, several different mutations have been found in

the MCT8 gene, leading to functional characterization in vitro. These gene mutations

include 11 gross deletions, 10 small deletions, 13 small insertions, two splice-site

mutations, 39 missense or nonsense mutations and one complex rearrangement

(Schwartz et al. 2005; Jansen et al. 2008; Visser, Jansen, et al. 2009; Visser et al.

2013) with differences in the phenotype dependent on the cell type (Kinne et al. 2009)

and localization of the mutated protein. Milder phenotypes can be due to an only

slightly lower expression pattern compared to wild type and therefore residual activity

of the transporter (Jansen et al. 2008).

Since a MCT8 deficiency has a devastating effect on global development, especially

neuronal development, the transporter is likely to be important for TH supply in the

brain and only in a lesser extent for other organs, probably due to the presence of

additional transporters that are able to compensate for the loss. Neuronal development

starts within the third gestational week (Stiles and Jernigan 2010) and for the first

trimester the embryo is provided with maternal TH (de Escobar et al. 2008). In

comparison to congenital hypothyroidism, where the fetus had maternal TH supply until

birth, MCT8 deficient patients are lacking TH from the beginning of gestation due to

inefficient uptake by the compromised transporter. In 2014, López-Espíndola et al.

published the first post-mortem brain histopathology of a MCT8-deficient male fetus

(week 30 of gestation) and an 11-year old male patient. Both brains showed

histological signs of immaturity, deficient myelination, and altered expression of TH-

dependent neuronal proteins like the light neurofilament subunit (NEFL) and the Ca

binding proteins parvalbumin (PVALB) and calbindin-D28k (CALB) in the frontal cortex

and cerebellum. They could also show increased DIO2 and decreased DIO3 activities

(López-Espíndola et al. 2014).

Due to the inefficient uptake into the brain, T3 exceeds in the periphery. Other

transporters than MCT8 on the other hand can facilitate the transport of T4 over the

BBB, so serum T4 concentrations are low. This results in a peripheral hyperthyroid

state with elevated T3 concentrations that leads to an overactive HPT axis, meaning

reduced TH secretion, as well as an increase in deiodination. The liver is in a

hyperthyroid state, as well as skeletal muscle cells resulting in raised energy

expenditure. Therefore, patients exhibit a hypermetabolic state with low body weight,

increased perspiration, tachycardia, muscle wasting which can lead to a life-

Figure 3: Overview of altered thyroid state of different tissues in AHDS patients. Inactivating

mutations of the MCT8 lead to a hypothyroid state of brain tissue and a hyperthyroid state in the

periphery. An insufficient amount of THs are transported into neurons, so TR signaling is

decreased and neuronal development is impaired. In peripheral tissues, other TH transporters

support the influx of THs leading to slightly elevated TSH generation of the pituitary and

therefore an overactive HPT axis, increased TR signaling in muscle cells inducing cachexia,

increased metabolism in liver cells and decreased retention of T4 in the kidney.

threatening cachexia (Groeneweg et al. 2016). TH retention through the kidney is

reduced due to MCT8 deficiency leading to a T4 accumulation within the kidney tissue

(Bernal, Guadaño-Ferraz, and Morte 2015). Figure 3 visualizes the altered thyroid

state and the resulting symptoms of AHDS patients.

1.2.3 Transgenic models for MCT8 deficiency

To investigate MCT8 and its functions further, a MCT8 KO mouse model was

generated. Though it replicated the thyroid state and serum TH concentrations of the

affected patients (Dumitrescu et al. 2006), the mice lacked most of the behavioral

impairments (Wirth, Roth, Blechschmidt, Hölter, Becker, Racz, Zimmer, Klopstock,

Gailus-Durner, and Fuchs 2009), though the uptake of T3 into the brain has been

shown to be compromised (Trajkovic et al. 2007). Although the KO model delivered an

insight in the mechanism of MCT8 deficiency for the thyroid state (Di Cosmo et al.

2010), the neuronal development is thought to be rescued by another TH transporter,

OATP1C1. It has been shown that a Mct8/Oatp1c1 DKO reflect a bit more the state of

human MCT8 deficiency such as the cerebellar development, the reduced myelination,

and locomotor abnormalities of MCT8-deficient patients (Mayerl et al. 2014).

Due to the limitations of mice models for MCT8, other organisms have been

investigated, such as zebrafish with the slc16a2 gene sharing 56-57% identity with the

human SLC16A2 (Arjona et al. 2011). This KO model showed a neurological

phenotype with impaired embryonic development (Vatine et al. 2013), altered

behavioral performance and hypomyelination (Zada et al. 2014). With a phenotype

closer to the human state of MCT8 deficiency than mice and the transparency of

zebrafish larvae, life-imaging studies can be performed easily.

Since the transgenic KO of a whole gene does usually not mimic the pathogenic

condition in the patients, a new approach was described in 2017. Here, fibroblasts from

two affected AHDS patients and several control samples were taken and

reprogrammed into induced pluripotent stem cells (iPSCs). Additionally, the gene-

editing tool CRISPR/Cas9 (clustered regularly interspaced short palindromic repeats)

was used to either induce or correct the MCT8 mutation. The iPSCs were then used to

generate two different models: a neural culture to investigate TH transport, neural

differentiation and maturation, as well as induced brain microvascular endothelia cells

(iBMECs) to establish a new BBB-model when seeded into a transwell system and

examine TH transport over this cell layer. From the study it was concluded that

although MCT8-deficient neurons had a significantly lower TH transport, T3-dependent

gene expression was not impaired. It has to be noted that T3 concentration exposed

on cells did not emulate in vivo conditions. On the other hand, the BBB-model has

been shown to impair transport and efflux rates for MCT8-deficient cells. Therefore,

the authors concluded, that the BBB is the critical barrier for TH transport and an

affected MCT8 can lead to under-supply of the brain (Vatine et al. 2017).

1.2.4 Treatment options for MCT8 deficiency

Due to the early onset of TH deprivation during gestation, therapeutic options for MCT8

deficiency are currently limited and the developmental brain impairments are almost

certainly irreversible. The most promising treatments so far are based on T3 analogs

that could enter the cell independently of MCT8. A potential treatment approach is

using 3,5-diiodothyropropionic acid (DITPA), a T3 analog, that is able to activate TRs

(Pennock et al. 1992), but maintains normal metabolic activity (Moreno et al. 2008). A

treatment of Mct8 KO mice resulted in an improvement of the peripheral hyperthyroid

state, but was not able to act on the hypothyroid brain (Ferrara et al. 2015).

Another TH analog, 3,5,3’-triiodothyroacetic acid (Triac), a metabolite present in the

normal organism in small concentrations, has also been shown to be transported

MCT8-independently, though the mechanism is unknown. In contrast to DITPA, Triac

was able to restore neural differentiation in Mct8/oatp1c1 KO mice (Kersseboom et al.

2014), though a recent study used Mct8 KO only and observed normalized TH serum

levels, but no measurable Triac concentration in the brain as well as almost no effect

on brain development. The authors deductively warned that the decreased

concentration of free T4 would limit the amount of THs delivered to the brain and would

aggravate the hypothyroid state even further (Bárez-López et al. 2016). It should be

noted that the two mentioned studies above used different mice models. As stated in

the last section, the Mct8 KO mouse model show only a mildly hypothyroid brain and

therefore do not exhibit the abnormalities in neuronal development (Mayerl et al. 2014).

For those reasons, the results from the latter study should be handled with caution.

1.3 Peptide-hormone conjugates

Hormonal treatment as an instrument in medical intervention, such as hormone

replacement therapy with steroids (Scharbo-Dehaan 1996) or TH treatment of CH

(Krude, Kühnen, and Biebermann 2015) have been done for a long time. Though a

central administration can be beneficial in some cases, for several hormones negative

side effects have been observed in certain tissues. The search for a possibility to

exploit positive effects of a treatment, and omit negative, gave rise to the idea of target

specific delivery of hormones. A new approach to deliver hormones specifically into

target tissues is the hybridization with a peptide ligand for a G protein-coupled receptor.

Here, binding of the ligand to receptor and the following internalization of this complex

is utilized to bypass the influx over a transporter. The peptide is acting as guidance to

specific target tissues that express the respective receptor. The conjugate is then

thought to be processed intracellularly and the hormone is released from the peptide.

For example, a peptide-hormone conjugate has been published very recently utilizing

the metabolic effects of T3 under the guidance of glucagon. Here the therapeutic

benefits of both, glucagon and T3, and their synergistic effects have been shown to

correct symptoms of metabolic disease by targeting the liver, a glucagon receptor-

expressing organ. The treatment with the conjugate did not affect the cardiovascular

system and therefore reduced the adverse effects of a systemic T3 application (Finan

et al. 2016).

Before that, in 2012, the first peptide-hormone conjugate was published: Glucagon-

like peptide 1 (GLP1) as guidance for estrogen to ensure target specific delivery only

into GLP1 receptor (GLP1R) expressing tissue (Finan et al. 2012). The steroid

estrogen is known to modulate energy expenditure and feeding behavior (Mauvais-

Jarvis 2011; Xu et al. 2011), but also exhibit undesired gynecological and tumor-

promoting effects (Paruthiyil et al. 2004). Therefore, a central estrogen treatment for

obesity is not a therapeutic option. The synthesized GLP1-estrogen conjugates have

been pharmacologically tested to show the beneficial effects on weight loss in mice

without the unwanted side effects on the uterus (Finan et al. 2012). The efficacy of this

conjugate has been demonstrated in several in vivo and in vitro studies (Tiano et al.

2015; Schwenk et al. 2015; Vogel et al. 2016).

The proposed mechanism behind these findings has been based on the internalization

mechanism of GPCRs. To gain full understanding of this mechanism, the properties of

receptors will be introduced in the following chapters.

1.4 G protein-coupled receptors (GPCRs)

Communication between cells and their neighboring cells or the extracellular space is

important for any biological process. The key players for the transmission of signals

from the outside to the cytoplasm are receptors. The largest family of receptors is the

family of transmembrane GPCRs that are abundant in all living organisms

(Schöneberg et al. 2007). They can act on various types of outside signals such as

light, mechanical impulses, and chemical compounds and conduct these signals to

proteins in the intracellular space (Rosenbaum, Rasmussen, and Kobilka 2009). There

are five classes of GPCRs based on sequence homologies, functional similarities and

a phylogenetic analysis: Class A, the rhodopsin-like GPCRs, class B, the secretin

receptor family, class C, the metabotropic glutamate GPCRs, adhesion and

frizzled/taste GPCRs (Fredriksson et al. 2003).

1.4.1 Structure of GPCRs

All GPCRs have a basic structure in common and consist of seven transmembrane α-

helices (transmembrane helices, TMH1-7) with an extracellular N-terminus and an

intracellular C-terminus, three intracellular (ICLs) and three extracellular loops (ECLs).

The largest class of GPCRs, class A contains 19 subclasses and includes a variety of

receptor types such as neurotransmitter, hormone or light receptors, which are usually

interact with a binding pocket inside of the THMs and ECLs (Rosenbaum, Rasmussen,

and Kobilka 2009). They all share several highly conserved amino acids in their TMHs

(Chelikani et al. 2007). Contrastingly, class B consists of a small group of GPCRs

(around 25 members) with little homology to the class A receptors. They mainly bind

smaller endogenous peptide ligands such as glucagon or secretin with their

comparatively large N-terminal extracellular domain of ~100 to 160 amino acids (Hoare

2005)

1.4.2 Signaling of GPCRs

The eponymous property of GPCRs is the signaling via G proteins within the

cytoplasm. Usually binding to an agonist causes changes in the receptor conformation,

which leads to the coupling of G proteins. They are heterotrimeric proteins consisting

of an α, β and γ subunit, that interact with the intracellular part of the receptor (Dohlman

et al. 1991). The α-subunit exhibits a GTPase activity, hydrolyzing guanine

triphosphate (GTP) to form guanine diphosphate (GDP) upon receptor activation.

Therefore, the GPCR serves as guanine triphosphate exchange factor (GEF), which

leads to the dissociation of the βγ-subunit from the α-subunit (Hurowitz et al. 2000).

There are four distinguished G protein classes, referred to by their α-subunits: Gαs

stimulating the adenylate cyclase (AC), Gαi inhibiting the AC and Gαq/11 activating

phospholipase C (PLC) and G12/13 activating Rho guanine-nucleotide exchange factors

(RhoGEFs). Those downstream factors are characteristic for the respective G protein

and start signaling cascades (Pierce, Premont, and Lefkowitz 2002).

The AC, activated by Gαs and inhibited by Gαi, catalyzes the reaction from adenosine

triphosphate (ATP) from cyclic adenosine monophosphate (cAMP) by

dephosphorylation. cAMP acts as a second messenger by binding to the regulatory

chain of protein kinase A (PKA) and therefore activation the catalytic subunit. The

targets of the PKA are diverse and are dependent on the cell type. Its general function

is the regulation of gene expression. In a feedback loop, an enzyme called the

phosphodiesterase deactivates the PKA by converting cAMP to AMP.

The activation of PLC- by Gαq/11 leads to the hydrolysis of phosphatidylinositol-4,5

bisphosphate (PIP2) to diacylglycerol (DAG) and inositol triphosphate (IP3). The latter

acts as a second messenger and binds to IP3 receptors located in the membrane of

the endoplasmic reticulum (ER). This triggers the release calcium ions (Ca2+) into the

cytoplasm. The calcium immobilization can have various effects such as muscle

contraction, glycogen release or vesicle transport. Ca2+ together with DAG also binds

to protein kinase C (PKC), which has several target proteins such as ERK1/2 (Jeremy

M Berg 2002; New and Wong 2003). The βγ-subunits of G proteins are also capable

of interaction with second messengers, leading to the activation of MAPK pathway

(ERK1/2) (Faure, Voyno-Yasenetskaya, and Bourne 1994).

Since not every GPCR couples to every known G protein, the signaling properties for

every receptor have to be investigated. These properties can differ depending on the

type of ligand activating the receptor as well as the interaction of the GPCR with other

transmembrane or cytosolic proteins. This functional selectivity is also known as

“biased signaling” (Shukla 2014). One example for biased agonism would be the

activation of a receptor by a newly discovered or synthetic agonist that shifts the

signaling preference of the GPCR to different cascades than it is known for its

endogenous ligand. Biased receptors are two distinct GPCRs that prefer divergent

signaling pathways despite being activated by the identical ligand due to structural

differences within the two, for example. The activated signaling cascades can also be

pleiotropic, meaning dependent on the cell environment a GPCR-ligand pair can differ

in different cell system due to differential expression of cofactors (Smith, Lefkowitz,

and Rajagopal 2018). Figure 4 depicts the four most prominent G proteins and their

pathway on the left side.

1.4.3 Internalization of GPCRs

Upon agonist-dependent receptor activation, it has to be ceased at some point in order

to inhibit continuous signaling. This desensitization mechanism is regulated by G

protein-independent signaling involving arrestins. Their family is divided into visual and

non-visual, ubiquitously expressed subtypes and they bind to GPCRs in order to initiate

receptor internalization. For this step, the C-terminal part of the GPCR is

phosphorylated by GPCR kinases (GRK) to initiate the binding of the arrestin molecule.

Clathrin-coated pits mediate the internalization process causing the complex of

receptor and bound ligand to undergo endocytosis. The internalized GPCR is then

sorted to degradation or recycling back to the cell surface (Moore, Milano, and Benovic

2007). As they move from early to late endosomes, the pH decreases, which cause

the receptor to release the bound ligand. Studies on the LDL (low-density lipoprotein)

receptor and the liver-specific asialoglycoprotein receptor suggest that the ligand is

then degraded in the lysosome (Lodish 2000). In addition to the desensitization and

internalization mechanism, arrestins act as signaling scaffold proteins for several

pathways, in particular the MAPK pathway. This signaling cascade controls many

cellular functions, such as cell cycle progression, transcriptional regulation, and

apoptosis (DeWire et al. 2007). As a scaffolding protein, arrestins facilitate kinase-

interaction, ensure signaling specificity, and maintain subcellular distribution of the

proteins (Pierce and Lefkowitz 2001). Some publications suggested that β-arrestins

were actual initiators of these signaling cascades and referred to G protein-

independent, β-arrestin-dependent signaling activation, mainly the ERK1/2

phosphorylation pathway (Eichel, Jullié, and von Zastrow 2016; Shenoy et al. 2006).

Recently, the involvement in actual activation of this pathway by β-arrestins has been

questioned, though the scaffolding role was not dismissed. New gene editing

techniques such as CRISPR/Cas9 made complete KO of G proteins or β-arrestins

possible, rather than a knock down using siRNA. A recent paper was able to show that

β-arrestins at “zero functional G” failed to initiate ERK phosphorylation and

morphological changes, whereas β-arrestin1/2 KO cells had an increase in ERK

phosphorylation after agonist stimulation, hinting towards solely desensitization as the

main role of arrestins and maybe a less pronounced role as a scaffold protein

(Grundmann et al. 2018). The internalization process is depicted in Figure 4 on the

right side.

Recent findings have suggested that the internalization process initiated by β-arrestins

does not necessarily result in an uncoupling of the α-subunit of the G protein and the

rapid termination of signaling (Pavlos and Friedman 2017). Sustained intracellular

(also called endosomal) cAMP signaling has been described for the TSHR (Calebiro

et al. 2009), the luteinizing hormone receptor (LHR) (Lyga et al. 2016), the parathyroid

receptor (PTHR) (Vilardaga et al. 2012), the GLP1R (Kuna et al. 2013), the

vasopressin 2 receptor (V2R) (Feinstein et al. 2013), the β2 adrenergic receptor (β2AR)

(Irannejad et al. 2013) and the dopamine receptors 1, 2 and 3 (D1R, D2R, D3R)

(Kotowski et al. 2011) in physiological related cells or cell lines for most cases. This

process seems to be dependent on the cell type, the receptor family, as well as the

agonist activating the GPCR (Vilardaga, Jean-Alphonse, and Gardella 2014).

The termination of endosomal signaling and the final step to dissociate receptor and

ligand is facilitated the recruitment of the multiprotein endosomal sorting actin-sorting

nexin 27 (SNX)-retromer tubule (ASRT) complex, which replaces β-arrestin. From

there, the receptor is either degraded or recycled in a rapid or slow fashion (Pavlos

and Friedman 2017).

1.4.4 Glucagon-like peptide 1 receptor (GLP1R)

The Glucagon-like peptide 1 receptor (GLP1R) is a class B receptor that is activated

by incretins in pancreatic β-cells (Hoare 2005). The incretins GLP1 and glucose-

dependent insulinotropic polypeptide (GIP) are produced by cells of the intestine and

released into the blood stream after meal ingestion. A third natural agonist for GLP1R

Figure 4: Graphical overview of GPCR signaling cascades and internalization mechanism. On

the left side, pathways of the four most prominent G proteins are depicted. Upon ligand binding

and receptor activation, a conformational change in the membrane protein leads to the exchange

of a GDP to a GTP in the α-subunit of a coupled G protein. This subunit hydrolyses GTP to GDP

and dissociates from its βγ-subunits. There are four prominent α-subunits: Gαs increases

intracellular cAMP concentration by activation of AC, Gαi decreases cAMP by inhibition of AC,

Gαq/11 activates PLC, which leads to an increase in intracellular Ca2+ and Gα12/13 activating the

RhoA pathway. All of these signaling cascades result in different gene regulations. On the right

side, GPCR internalization is depicted. Upon activation, receptor signaling needs to be ceased

to prevent overstimulation of the cell. GRKs phosphorylate the C-terminal part of the GPCR,

which results in the recruitment of β-arrestins. The receptor-ligand complex gets internalized

into clathrin-coated pits. After this endocytosis, receptors are either degraded or recycled back

to the plasma membrane. AC: adenylate cyclase; PKA: protein kinase A; ERK: extracellular-

signal regulated kinase; PLC: phospholipase C; GRK: GPCR kinase; Rho homology gene family,

member A; RhoGEF: Rho Guanine exchange factor

is glucagon, which is released by the pancreatic α-cells when insulin concentration in

the bloodstream is low to control glycolysis (Voet and Voet 2011). Agonist-dependent

activation of the GLP1R leads to an increase in intracellular cAMP, which results in an

accumulation of intracellular calcium and thus mediates insulin secretion, proliferation

and inhibition of apoptosis. Additionally, GLP1R is expressed in neurons of the

hypothalamus, where agonists promote reduced food intake and satiety. In the

pancreas, GLP1R signaling controls insulin secretion and biosynthesis as well as β-

cell proliferation and neogenesis (Baggio and Drucker 2007).

As mentioned above, the main signaling cascades activated by GLP1R is the GS-

pathway, accumulating intracellular cAMP (Skoglund, Hussain, and Holz 2000).

Additionally, GLP1R in rattus norvegicus is reported to induce calcium response as

well as ERK1/2 phosphorylation, suggesting coupling to the Gq/11 protein and Gi1,2

(Wheeler et al. 1993; Montrose-Rafizadeh et al. 1999). Co-expression and

consequently interaction with GIP receptor (GIPR) was shown to significantly decrease

cAMP and calcium production as well as ERK1/2 phosphorylation, whereas the

interaction with glucagon receptor (GCGR) impaired calcium response and to a lesser

extend ERK accumulation, but not cAMP production (Roed et al. 2015). As mentioned

in the last chapter, GLP1R was reported to exhibit endosomal cAMP signaling in a

pancreatic β-cell line suggesting implications for receptor-mediated regulation of

insulin secretion (Kuna et al. 2013).

The peptide GLP1 is the result of a posttranscriptional cleavage of proglucagon, coded

by the glucagon gene and is only processed in L cells, an endocrine cell type located

in the ileum and the small intestine (Doyle and Egan 2007).

1.4.5 Oxytocin receptor and vasopressin arginine 1 A receptor

1.4.5.1 Oxytocin receptor (OTR)

The oxytocin receptor (OTR) belongs to the class A GPCRs and is known to couple to

Gq/11 as well as Gi proteins resulting in increased Calcium mobilization and ERK1/2

phosphorylation, and decreased intracellular cAMP concentrations (Strakova et al.

1998). Its endogenous ligands oxytocin (OT) and the structural closely related arginine

vasopressin (AVP), as well as synthetic drugs can activate the receptor. The two

closely-related natural agonists are small neuropeptides, consisting of nine amino

acids and differ only in two amino acids: the third position, an isoleucine for OT and a

phenylalanine for AVP, and the eighth position, a leucine for OT and an arginine for

AVP (Ganten et al. 1986). They originate from the same chromosomal locus and arose

from an early gene duplication event in evolution (Gwee et al. 2009). Their structural

similarity leads to the overlap in receptor targets, including the OTR and the three

vasopressin receptors (V1AR, V1BR V2R). Both peptides are synthesized in

magnocellular neurons in the paraventricular nucleus (PVN) and the supraoptic

nucleus (SON) of the hypothalamus. From there, their producing neurons project into

the posterior lobe of the pituitary, where the neuropeptides are stored in vesicles and

can be released into the peripheral circulation to reach OTR expressing tissues

(Meyer-Lindenberg et al. 2011).

The expression of OTR is most prominent in uterus, placenta, kidney, heart and the

CNS (Kimura et al. 1992). Its biological functions include mediation of uterine

contractions, lactation, cardiovascular and kidney functions, bone and muscle

formation as well as sexual and social behaviors, such as maternal behavior and

interpersonal bonding (Gimpl and Fahrenholz 2001; Zingg, Bourque, and Bichet 2012).

1.4.5.2 Vasopressin arginine 1 A receptor (V1AR)

The vasopressin arginine 1 A receptor (V1AR) also belongs to the Class A GPCRs and

shares 45% sequence with the OTR. The receptor has a higher affinity to AVP in

comparison to OT (Åkerlund et al. 1999; Tahara et al. 1997; Tahara et al. 1998). The

main pathway is the activation of Gq/11/PLC- pathway and consequently ERK

phosphorylation (Tahara et al. 1999; Schöneberg et al. 1998; Thibonnier et al. 1994)

In the periphery, V1AR is expressed in liver, kidney and vasculature, whereas the

behavioral functions are mediated by the expression throughout the CNS, in particular

hippocampus, hypothalamus, olfactory bulb, ventral tegmental area, substantia nigra,

superior colliculus, dorsal raphe, nucleus of the solitary tract and inferior olive (Johnson

et al. 1993; Caldwell et al. 2008). It has been under intense investigation concerning

pair bonding, anxiety, circadian rhythm and renal function in rodents (Bielsky et al.

2003; Bales et al. 2007; Li et al. 2009). In humans, compared to rodents, V1AR is not

expressed in the cortex and is investigated in the context of personality traits and

autism (Loup et al. 1991; Meyer-Lindenberg et al. 2009).

1.5 The “Trojan Horse”-like mechanism

Although the efficacy of peptide-hormone conjugates has been shown in in vivo

models, the underlying mechanism behind it is not completely elucidated. It is proposed

that the GPCR-conjugate-complex is internalized upon receptor activation. Inside the

cell, the pH-sensitive linker between peptide and hormone releases the hormone,

which can then activate its respective signaling responses. Since this approach uses

an unconventional way to deliver hormones into target cell by linking it to another

compound, it has been dubbed the “Trojan Horse” approach or the “Trojan Horse”-like

mechanism. Here the hormone conjugated to a peptide is only taken up into cells

where the peptide receptor is expressed and thereby directed the hormone to specific

target cells. In case of MCT8 deficiency, it could be a therapeutic option by choosing

one or more target receptors that are expressed in hypothyroid areas of the brain

(Figure 5).

Here, the choice of guidance peptide and target receptor needs careful consideration.

The peptide should be easy to administer and have no or few adverse effects.

Optimally, it should add to the overall objective of the treatment. The expression of its

target receptors should align with the affected areas of the transporter deficiency.

Expression in additional areas and the effects on these tissues need to be kept in mind

for adverse side effects. Binding of the conjugate to the GPCR should start a signaling

cascade, which could be G protein- or β-arrestin–dependent. Most importantly, the

internalization process needs to be initiated by the conjugate. Negative effects of the

linker position on the binding capacity of the peptide to the receptor need to be avoided

or at least biased signaling should be noted as it could have diverse effect from the

administration of the peptide alone. Lastly, upon internalization the hormone should

start its respective signaling cascades similar to a hormone transported via a

transmembrane transporter. To find the perfect peptide-receptor pair for the targeted

application, in vitro studies are necessary, especially to minimize animal testing.

The first created conjugate that would target the central nervous system was GLP1-

T3. It has been developed for metabolic interventions and should primarily target

pancreatic β-cells, but was the perfect conjugate for proof-of-principle studies. The

second conjugate was designed derived from the symptoms of MCT8-deficient

patients and their probably most affected brain areas: Oxytocin-T3 (OT-T3). T3 was

attached to oxytocin at the position 8 (Leu) with a pH-sensitive linker. Here, the two

important target receptors in mind were V1AR and OTR, both highly expressed in

several brain areas.

Figure 5: MCT8-deficiency and the “Trojan Horse”-like mechanism (A) The monocarboxylate

transporter 8 (MCT8) is able to control efflux and influx of thyroid hormones (TH) T3 and T4.

The biological active T3 binds to nuclear TH receptors (TR). The complex translocates into the

nucleus, where a cell response is started. An inactivation mutation of MCT8 leads to a

decreased uptake in the central nervous system and therefore to a severe psychomoto

retardation.

(B) A therapeutic option for MCT8-deficiency could be the “Trojan Horse”-approach with the

administration of peptide-T3 conjugates. The proposed mechanism is based on the

internalization of G protein-coupled receptor (GPCR. The conjugate activates the receptor, the

GPCR-ligand complex is internalized and T3 is released into the cytoplasm. Here it can bind

to the TRs and start a cell response.

Picture credit: Heiko Krude

2 Aim of the study

From in vivo studies of already designed conjugates, it is known that the “Trojan

Horse”- mechanism is generally working. So far, this mechanism was used to increase

energy expenditure in order to treat obesity and related complications such as diabetes

(Finan et al. 2012; Finan et al. 2016). Yet, this approach was never used to treat a rare

disorder such as MCT8 deficiency. Therefore, the intention of this study was the

elucidation of in vitro aspects that are prerequisites to use the “Trojan Horse”-like

mechanisms in MCT8 deficiency. Ideally, a perfect ligand-receptor pair would be found

that can be used as treatment for this devastating disease.

New peptide-T3 conjugates have been synthesized by our cooperation partners at the

chemistry department of the Indiana University (Bloomington, USA), tested in vivo in

Munich (Helmholtz Zentrum, Institute for Diabetes and Obesity), and need to be

characterized in addition to in vivo studies. The investigation will be covering the

following questions for every tested conjugate:

a) Does the conjugate activate the receptor and its signaling cascades? Here, all

known signaling pathways for the respective receptor will be analyzed in

comparison to the unconjugated peptide.

b) Does the conjugate initiate the receptor internalization? This is the crucial step

for the “Trojan Horse”-like mechanism and will be investigated using

internalization assays as well as fluorescence-based imaging methods.

c) Does the conjugate activate both canonical and non-canonical TH-dependent

signaling pathways? Cells over-expressing the specific receptor are compared

with cells that do not express the receptor. Only if T3-dependent signaling

increases in the presence of the peptide receptor, it shows that the conjugate

enter the cell through the “Trojan Horse”-like mechanism. If possible, the

endogenously expressed TH transporter should be inhibited using known TH

transporter blocker. If T3-dependent signaling is still present after inhibition, the

T3 entered the cell through the “Trojan Horse”-like mechanism.

The first available conjugate used to establish the assays was GLP1-T3. It was

developed to treat metabolic diseases, but was the ideal compound for proof-of-

principle experiments. The conjugate that was developed with MCT8-deficiency in

mind was oxytocin-T3. It has two target receptors that are expressed in the brain: The

oxytocin receptor (OTR) and the Vasopressin arginine receptor 1 A (V1AR). It will be

investigated using the established methods.

3 Material and methods

3.1 Materials

3.1.1 Technical equipment

Table 4: Machines used and their corresponding supplier company

Machine Company

Anthos Plate reader 2001 Anthos Mikrosysteme GmbH, Krefeld,

Cell culture hood Lamin Air HBB 2448 Heraeus Instruments, Hanau, DE

Centrifuge Sorvall RC 6 Plus Thermo Fisher, Waltham, MA, USA

Deep freezer Forma -80C ULT-Freezer Thermo Fisher, Waltham, MA, USA

DNA sequencer 3130xL Genetic

Analyzer Applied Biosystem, Thermo Fisher

Gel chambers Whatman compact XS/S Biometra, Jena, DE

Gel documentation system GeneFlash Syngene, Cambridge, UK

Heating block Thermomixer Compact Eppendorf, Hamburg, DE

Incubator certomat® BS-1 B. Braun, Melsungen, DE

Incubator kelvitron® t Heraeus Instruments

Lab balance CPA 223S-OCE Satorius, Göttingen, DE

LUNA™ Automated Cell Counter Logos biosystems, Villeneuve d’Ascq,

Microscope Axiovert10 Zeiss, Oberkochen, DE

MilliQ-system Millipore Water Purification

Systems Merck Millipore, Darmstadt, DE

Mini centrifuge Galaxy Mini VWR, Darmstadt, DE

pH-meter Seven easy Mettler Toledo, Gießen, DE

Photometer BioPhotometer Eppendorf

Plate reader Mithras LB 940 Berthold Technologies, Bad Wildbad, DE

Shaker IKA-vibrax-VXR Janke&Kunkel, Staufen, DE

Shaker Vari-Shaker Dynatech Laboratories Ltd., Sussex, UK

Table centrifuge Centrifuge 5417R/C Eppendorf

Thermocycler Mastercycler® Gradient Eppendorf

3.1.2 Chemicals and consumables

The companies that supplied consumables (Table 5A) and chemicals (Table 5B) are

listed below.

Table 5: Supplier companies for consumables (A) and chemicals (B)

A B

Becton Dickinson Biosciences Invitrogen, Darmstadt, DE

Berthold Technologies Merck, Darmstadt, DE

Biozym Scientific GmbH, Hessisch

Oldendorf, DE

Promega, Mannheim, DE

BRAND GmbH + CO KG, Wertheim, DE Roth, Karlsruhe, DE

Eppendorf Sigma-Aldrich, Taufkirchen, DE

Greiner bio-one, Frickenhausen, DE

PerkinElmer, Rodgau, DE

Sarstedt, Nürnbrecht, DE

Thermo Scientific

TPP Techno Plastic Products,

Trasadingen, CH

3.1.3 Kits

Table 6: Overview of the commercially available kits that were used in the course of this project

Kit Supplier

ABI Prism® BigDye® Terminator v3.1 Cycle

Sequencing Kit

Applied Biosystems

AlphaScreen™ cAMP Assay Kit Perkin Elmer

CellTiter 96® AQueous One Solution Cell

Proliferation Assay (MTS)

Promega

Luciferase Assay System, Reporter Lysis 5x Buffer Promega

NanoBRET™ Nano-Glo® Detection System Promega

Nano-Glo® HiBiT Extracellular Detetction System Promega

Pure Yield™ Plasmid Midiprep System Promega

Pure Yield™ Plasmid Miniprep System Promega

SsoFast™ Evagreen Supermix® Bio-Rad, München, DE

Wizard® SV Gel and PCR Clean Up System Promega

3.1.4 Buffers, reagents

All buffers have been prepared using HPLC or Milli-Q water.

Table 7: Buffers and reagents commercially purchased or prepared in the lab

Buffer/solution Components/Supplier/Comments

Ampicillin stock solution 50 mg/ml Ampicillin

cAMP assay medium 138 mM NaCl, 6 mM KCl, 1mM MgCl2 * 6

H2O, 5.5 mM Glucose, 20 mM Hepes,

1mM CaCl2 * 2 H2O, 0.1% BSA, adjusted

to pH 7.4

4′,6-diamidino-2-phenylindole (DAPI)

stock solution

5 mg/ml dissolved in HPLC H2O, Roche

Applied Science, Mannheim, DE)

dNTP mix stock 50 mM, Invitrogen

dNTP mix 10 mM dNTP-Mix diluted from dNTP mix

stock

70 % Ethanol 70 % (v/v) (96 % Ethanol diluted in HPLC

water)

HPLC water HPLC Gradient Grade, purchased from

Fisher Chemicals, Schwerte, DE

3-Isobutyl-1-methylaxanthine (IBMX)

stock solution

500 mM IBMX (Sigma-Aldrich), solved in

DMSO (Dimethylsulfoxide)

Kanamycin stock solution 50 mg/ml Kanamycin

Loading dye 0.05 % (w/v) Bromphenolblue sodium salt,

0.05 % (w/v) Xylencyanol, 50 % (v/v) 1 x

TBE, 50 % (v/v) Glycerol

LO buffer

(basic solution for LI buffer)

0.1 % BSA (Bovine Serum Albumin), 0.3

% Tween® 20, 5 nM Hepes, adjusted to

pH 7.4

LI buffer (lysis buffer for AlphaScreen™) 1 mM IBMX to LO buffer

Paraformaldehyde (PFA) for cell fixation 4% PFA in PBS

PBS Dulbecco w/o Ca2+, w/o Mg2+, instamed 9.55 g/l, 1x:

diluted in 5 l Milli-Q water, purchased from

Biochrom AG, Berlin, DE

PBS-T 0.05 % (w/v) Tween® 20, in 1 x PBS

5 x TBE stock solution

(Running buffer for gel electrophoresis)

0.5 M Boracic acid, 10 mM EDTA, 0.5 M

Tris, adjusted to pH 8.0

3.1.5 Stimulation agents/Blockers

Table 8: Stimulation agents used in this project

Stimulant/Blocker Stock solution Supplier

2-aminobicyclo-(2,2,1)-

heptane-2-carboxylic acid

(BCH)

10 mM in H2O Sigma-Aldrich

3,3′,5-Triiodo-L-thyronine, T3 10 mM in DMSO Sigma-Aldrich

3-iodothyronamin (3-T1AM) 1 mM in DMSO Santa Cruz

Bromosulphtalein (BSP) 4 mM in H2O Sigma-Aldrich

Desipramine hydrochloride

(DMI)

10 mM in H2O Sigma-Aldrich

Forskolin 10 mM in DMSO BioChemica -

AppliChem, Darmstadt,

Glucagon-like-peptide 1 (GLP1) 1 mM in PBS + 0.1%

BSA

Indiana University,

Bloomington, USA

GLP1-T3 conjugate 1 mM in PBS + 0,1%

BSA

Indiana University,

Bloomington, USA

Oxytocin 1 mM in DMSO Indiana University

Oxytocin-T3 1 mM in DMSO Indiana University

Pertussis toxin (PTX) 50 µg/ml Sigma-Aldrich

Probenecid 10.3 mM in NaOH,

adjusted to pH 7.4

Sigma-Aldrich

Silychristin 20 mM in EtOH Sigma-Aldrich

Thyrotropin-stimulating

hormone from bovine pituitary

(bTSH)

10 IU/ml in PBS +

0,1% BSA

Sigma-Aldrich

BSA: Bovine Serum Albumin; PBS: Phosphate buffered saline; DMSO:

Dimethylsulfoxide; NaOH: Sodium hydroxide; EtOH: Ethanol

3.1.6 Enzymes

All reactions using the following commercially available enzymes were prepared with

the supplied appropriate buffers.

Table 9: Commercially available enzymes used for this project.

Enzyme Supplier

Carboxy-Flexi® Enzyme Blend (SgfI, EcoICRI) Promega

Flexi® Enzyme Blend (SgfI, PmeI) Promega

Mango-Taq-DNA Polymerase Bioline

Pfu-Turbo DNA Polymerase Agilent Technologies

Restriction endonucleases (DpnI, EcoRI, HindIII,

NheI, PacI, XbaI, XhoI)

New England Biolabs

T4 DNA Ligase New England Biolabs

Trypsin/EDTA-solution (0,5/0,2% in PBS) Biochrom

3.1.7 Expression vectors

Table 10: Expression vectors utilized in the course of this project; * marks

Name Supplier

pBiT3.1-secN Promega

pcDNA3 IEPE (Charité)

pcDps IEPE (Charité)

p(DR4)2-SV40-luc+ (pGL3) IEE (Charité)

pFC14A Promega

pFC32K Promega

pGL4.3-[luc2P/NFAT-RE/Hygro] Promega

pGL4.3-[luc2P/SRE/Hygro] Promega

pRS-rTRα1 IEE (Charité)

pSNAPf New England Biolabs

3.1.8 cDNA of proteins of interest

Table 11: Sources for cDNA of proteins of interest for this project

Gene Protein Supplier

rARRB2 β-arrestin2 (rattus

norvegicus)

Dr. Vera Knäuper College of Biomedical and Life

Sciences, Cardiff University, UK

GLP1R Glucagon-like peptide 1

receptor

Source Bioscience, Nottingham, UK

OTR Oxytocin receptor DNASU plasmid repository

V1AR Vasopressin Arginine

receptor 1 A

DNASU plasmid repository

3.1.9 Fluorescent dyes, gel electrophoresis ladder

Table 12: Fluorescent dyes and DNA markers implemented for this project

Agent Supplier

SNAP Surface® A488 New England Biolabs

1 kb DNA Ladder (0.5 μg/lane) Invitrogen, BIOLINE

Easy Ladder I Invitrogen, BIOLINE

3.1.10 Cell culture reagents and media

Table 13: Overview of cell cultured related reagents and media for this project

Agent Supplier

Advanced MEM (w Non-essential amino

acids, w 110 mg/l Na- pyruvate, w/o L-

glutamine)

Gibco

Charcoal-treated Fetal Bovine Serum

(FBS)

Biochrom

Dulbecco´s MEM (w 3.7 g/l NaHCO3, w

4.5 g/l D-Glucose, w/o L-Glutamine, w/o

Na- Pyruvate, low endotoxin)

Biochrom

Fetal Bovine Serum (FBS) Biochrom

FuGene HD® Promega

FBS (charcoal-stripped, sterile-filtered) Sigma-Aldrich

MEM Earle´s (w 2.2 g/l NaHCO3, w stable

glutamine, low endotoxin)

Biochrom

Metafectene® Biontex Laboratories GmbH,

Martinsried, DE

Mounting medium with DAPI Vectastain

Non-essential amino acids (NEA) Biochrom

Opti-MEM I Reduced Serum Medium (1x) Gibco

Opti-MEM I Reduced Serum Medium

(1x), w/o phenol red

Gibco

Penicillin/streptomycin Biochrom

Phenol/water/chloroform Applied Bioscience, Beverly Hills, CA,

USA

Phosphate buffered saline (PBS), 1x,

Dulbecco (w/o Ca2+, w/o Mg2+, low

endotoxin)

Gibco

Poly-L-lysine (0.1 mg/ml) Biochrom

Cells were cultivated in complete media with additive specified as:

Table 14: Composition for complete media for the different cell lines used in the course of this

project

Complete medium

for

Basis medium additives

HEK 293 MEM 5% (v/v) FBS, 0.5% (v/v) NEA

COS7 DMEM 10% (v/v) FBS, 0.5% (v/v) Pen/Strep

Freezing medium FBS 10% (v/v) DMSO

3.1.11 Bacteria culture reagents and media

All media are prepared using MilliQ water.

Table 15: Composition for bacterial culture reagents and media implemented for this project

Medium Composition Antibiotics

LB medium 0.5% (w/v) yeast extract, 1% (w/v)

tryptone/peptone from casein, 1% (w/v)

sodium chloride, adjusted to pH 7.4

100 µg/ml

ampicillin or

kanamycin

LB agar plates 1,5% (w/v) agar-agar in LB medium 100 µg/ml

ampicillin or

kanamycin

SOB medium 0.5% (w/v) yeast extract, 2% (w/v)

tryptone/peptone from casein, 0.05% (w/v)

sodium chloride, 25 mM potassium

chloride, adjusted to pH 7.4

SOC medium 20 mM magnesium chloride, 20 mM

glucose in SOB medium

3.1.12 Bacterial strains and eukaryotic cell lines

Table 16: Bacterial strains and eukaryotic cell lines used in the course of this project

Strain/cell line Properties

E. coli Max

Efficiency® DH5α™

Chemically competent bacterial strain, 109 transformants/µg

plasmid DNA, purchased from Invitrogen

HEK 293 Human embryonic kidney cell line, immortalized by

transformation with adenoviral fragments, adherent

COS7 Cercopithecus aethiops, African green monkey kidney

fibroblast-like cell line, immortalized by transformation with

SV40 T antigen, adherent

3.1.13 Computer software

Table 17: Computer software used for this project

Program Supplier

GraphPad Prism 6 GraphPad Software, La Jolla, CA, USA

Microsoft Office Microsoft, Unterschließheim, DE

MikroWin 2000 Mikrotek Laborsysteme, Overath, DE

ImageJ National institute of health, MD, USA

3.2 Methods

3.2.1 Molecular biology methods

In this project, peptide-T3 conjugates were validated using numerous in vitro assays.

Therefore, the target receptors were cloned into the expression vector specific for the

various in vitro assays. Afterwards, characterization of the conjugates was performed

focusing on two main topics: 1) GPCR activation and internalization by the compound

and 2) the TH signaling as proof of the “Trojan Horse”-like mechanism. Figure 6

depicts the workflow for the project:

3.2.1.1 Cloning strategy

For the validation of the peptide-hormone conjugates, several in vitro assays were

employed. For these, cDNA of the target receptors was cloned into different expression

vectors following the scheme in Figure 7:

Figure 6: Schematical depiction of the workflow for this project. The cDNA of the target receptors

was cloned into various expression vectors that are specific for the in vitro assays. Afterwards,

GPCR signaling and internalization as well as TH signaling was investigated using various in

vitro assa

The untagged versions of the GPCRs have been cloned into the expression vector

pcDps, which utilizes a SV40 promotor for expression. The optical visualization to

investigate the behavior of GPCRs towards the peptide-hormone conjugates used

several fusion proteins generated by expression vectors. These needed to be cloned

into different expression vectors in the course of this project. The first generated N-

terminally SNAP-tagged constructs of GPCRs with the pSNAP

expression vector.

Additionally, fusion proteins for a new proximity assay (NanoBRET™) were generated,

adding either the luciferase NanoLuc or the protein tag HaloTag®, both C-terminally,

using the expression vector pFC32K or pFC14A respectively. Lastly, a small protein

tag called HiBiT has been added to the N-terminus of the GPCRs by employing the

expression vectors pBiT3.1-N or pBiT3.1-secN. The GPCRs used for tagging were

GLP1R, V1AR, and OTR. All of these plasmids express the insert under the control of

the CMV promotor. Furthermore, all cDNA sequences were checked for cutting sites

beforehand, to make sure that the enzymes do not cut inside of the amplicon.

3.2.1.1.1 Cloning of SNAP-tagged GPCRs

To investigate the internalization of GPCRs, N-terminally SNAP-tagged receptors have

been generated. The advantage of the pSNAP

expression vector was the fact that it

has two multiple cloning sites (MCS) to be able to generate both, N- and C-terminally

tagged fusion proteins. Since the GLP1R includes a signal peptide, a two-step cloning

strategy was used to position the signal peptide N-terminally of the SNAP-tag, whereas

the rest of the GPCR is fused to the C-terminus of the tag. The receptors V1AR and

OTR do not need signal peptides to be delivered to the plasma membrane, so no

additional cloning step was needed.

Figure 7: Overview of in vitro assay and the expression vectors used for cloning for the specific

assays. The in vitro assays were used to monitor signaling and GPCR internalization. Signaling

assays were either luciferase-based reporter gene assays or a competitive-based AlphaScreen™

assay. For all of them, GPCRs were cloned into the expression vector pcDps. For the

internalization assays, cell surface expression was measured using either the HiBiT system,

which uses the expression vectors pBiT3.1-N or pBiT-3.1secN, or the SNAP-tag system that

requires cloning into the expression vector pSNAPf. Part of the internalization mechanism is the

β-arrestin2 recruitment of GPCRs, which was measured using a protein interaction assay, the

NanoBRET™, a system that uses pFC14A as the acceptor fusion vector and pFC32K as the

donor fusion vector.

For DNA amplification of all GPCRs, primers were designed to include the restriction

enzymes XhoI for the forward primer and PacI for the reverse primer. The GLP1R

signal peptide cDNA was amplified using primers containing NheI for the forward

primer and EcoRI for the reverse primer. All restriction enzymes are contained either

in the first MCS of the expression vector, located N-terminally from the SNAP-tag

(NheI, EcoRI) or in the second MCS on the C-terminal site of the tag (XhoI, PacI).

3.2.1.1.2 Cloning fusion proteins for NanoBRET™

To generate the GPCRs fused with the BRET partners, the expression vectors from

the Flexi cloning system (Promega) were utilized. These vectors enclose a protein-

encoding region flanked by the restriction enzymes SgfI and EcoICRI as cloning sites,

endonucleases that do not affect most human open reading frames. The PCR

amplicons were generated using primers that add a SgfI restriction site to the 5’ end

and a PmeI restriction site on the 3’ end. After digestion, these proceed into a sticky

and a blunt end site respectively. Cloning primers were designed omitting the stop

codon and append a GTT codon (encoding a valine). Ligation to the blunt end

produced by EcoICRI in the plasmid results in the elimination of the stop codon in the

PmeI restriction site, allowing the fusion with one of the two BRET partners. All GPCRs

were cloned into both expression vectors for NanoBRET™, resulting in fusion proteins

with NanoLuc and HaloTag® on the C-termini of the receptors.

3.2.1.1.3 Cloning HiBiT-tagged GPCRs

A new surface expression assay employs a small protein tag (eleven amino acids)

called HiBiT. To generate HiBiT-tagged GPCRs, the HiBiT MCS-cloning system

(Promega) was used. Here, choosing different restriction sites in the vector can employ

a variable Gly/Ser linker. For the GLP1R a very long spacer of 24 amino acids was

chosen and the cDNA was cloned into the pBiT3.1-secN expression vector, containing

an N-terminal IL6 signal peptide to replace the signal peptide of the GLP1R. The used

restriction enzymes were XbaI and HindIII and were added via PCR to the amplified

sequence.

3.2.1.2 Polymerase chain reaction (PCR)

Polymerase chain reaction is an enzyme-dependent method to amplify specific double-

stranded DNA sequences exponentially (Schorderet 1994). This method was used to

amplify target sequences from expression vectors with the polymerase PfuTurbo. The

reaction was prepared according to Table 18:

Table 18: PCR preparation and cycler program

PCR preparation Cycler program

ingredient amount time Temperature cycles

template DNA 50 – 300 ng 2 min 95°C 1

PfuTurbo-Buffer 1x 30 s 95°C

30 dNTP mix 0.4 mM 30 s 55°C

Forward primer 125 ng 5 min 72°C

Reverse primer 125 ng 10 min 72°C 1

PfuTurbo 2.5 U 30 min 4°C 1

HPLC water Add to 25 µl ∞ 20°C 1

After the reaction, the amplicon was purified using the column-based “wizard SV gel

and PCR clean-up system” (Promega) according to manufacturer’s protocol to remove

proteins, salts and unused nucleotides. For further cloning, the DNA underwent

digestion with the appropriate restriction enzymes, another purification and ligation into

the respective expression vector. Afterwards the ligated vectors were propagated in

chemically competent E.coli (DH5α) by transformation.

3.2.1.2.1 Site-directed mutagenesis

In order to introduce insertion, deletion or substitutions within the target DNA, site-

directed mutagenesis can be used. This method utilizes a customized PCR with

primers, complementary to the template, containing the desired modification,

preferably in the middle of the sequence. After the reaction, the amplicon is not

methylated in contrast to the template sequence, which does not contain the desired

modifications. To remove the template, a digestion with DpnI is used. This restriction

enzyme recognizes and cleaves methylated DNA, which evolve by DNA-methylases

in E.coli, and can be used to remove the methylated template DNA, leaving the

amplicon intact (Kunkel 1985). In the course of this project, the PCR was performed

with PfuTurbo. The reaction was prepared according to Table 19:

Table 19: Site-directed mutagenesis PCR preparation and cycler program

PCR preparation Cycler program

ingredient amount time Temperature cycles

template DNA 10 – 200 ng 30 s 95°C 1

PfuTurbo-Buffer 1x 1 min 95°C

dNTP mix 0.4 mM 1 min 55°C

Forward primer 125 ng 5 min 68°C

Reverse primer 125 ng 10 min 68°C 1

PfuTurbo 2.5 U

HPLC water Add to 50 µl ∞ 4°C 1

After the reaction, digestion with DpnI (20 units) and transformation into chemically

competent E.coli (DH5α) was performed.

3.2.1.2.2 DNA sequencing

DNA sequencing is a standard procedure to verify the correctness of distinct

sequences. In the course of this project, sanger-sequencing was performed (Sanger,

Nicklen, and Coulson 1977). This technique relies on the principle of PCR with

additional differently fluorescently labeled dideoxynucleotides (ddNTPs). Here, only

one primer is used to amplify single-stranded DNA with randomly introduced ddNTPs,

which results in the termination of the elongation due to the lack of the 3’-hydroxy

group. This produces a mix of DNA fragments of various lengths. To analyze the

sequence, these fragments are then sorted by length in automated sequencing

machines that read out the fluorescent labels on the 3’-end (Shendure et al. 2011). In

the course of this project, DNA sequencing was conducted using the Big Dye

polymerase. The reaction was prepared according to Table 20:

Table 20: Sequencing PCR preparation and PCR program

PCR preparation Cycler program

ingredient amount time Temperature cycles

template DNA 100 – 300 ng 2 min 95°C 1

Big Dye

sequencing

buffer

0.75x 30 s 95°C

30 s 55°C

primer 0.5 µM 5 min 60°C

Big Dye Mix V3.1 0.75 µl 7 min 60°C 1

HPLC water Add to 10 µl ∞ 4°C 1

After the reaction, the DNA was precipitated using 0.1 M sodium acetate and 20 µl 96

% ethanol. After a centrifugation step at 15°C for at least 30 min, the supernatant was

removed by suction and the pellet was washed with 70% ethanol. After a second

centrifugation step for 15 min the supernatant was removed again and pellets were

dried for 5 min at 37°C. The dry pellet was stored in -20°C until it was suspended in

sequencing buffer and analyzed using the sequencing machine (ABI PRISM 3130xl)

3.2.1.3 Restriction digestion

For the insertion of double-stranded DNA into an expression vector, digestion with

restriction enzymes is used to create compatible ends to join. Depending on the

enzyme used for the digestion, 3’- or 5’-sticky ends or blunt ends can be created by a

break of the phosphodiester link of the DNAs backbone. The specific cutting sites are

palindromic, so that sticky ends can only be joined with the complementary end

respectively, whereas blunt end can be joined with any other blunt end. To inhibit

spontaneous re-ligation of the expression vector after restriction, the 5’-phosphate

group needs to be removed by additional incubation with calf intestinal alkaline

phosphatase (CIP) (Arber and Linn 1969). The reaction was prepared according to

Table 21:

Table 21: Restriction digestion preparation for expression vector and insert

Digestion preparation for expression

vector

Digestion preparation for DNA insert

ingredient amount ingredient amount

Digestion buffer 1x Digestion buffer 1x

Expression vector 4 µg Expression vector 1 µg

Restriction enzyme I 1,5 µl Restriction enzyme I 2,5 µl

Restriction enzyme II 1,5 µl Restriction enzyme II 2,5 µl

HPLC water Add to 30 µl HPLC water Add to 50 µl

Incubation at 37°C for 45 min

Incubation at 37°C for 1h

CIP 1,5 µl

Incubation at 37°C for 15 min

To identify and purify the desired fragments, agarose gel electrophoresis was

implemented. Respective bands were cut out, identified by their length (in bp) and

purified using the column-based “wizard SV gel and PCR clean-up system” (Promega)

according to manufacturer’s protocol. The next step is the joining of expression vector

and insert using ligation.

3.2.1.4 Ligation

The actual joining reaction of two double-stranded DNA molecules is executed by the

T4 ligase enzyme and is called ligation. The ligase joins a 5’-phosphate with a 3’-

hydroxy group to create a phosphodiester bond between two nucleotides and uses

ATP (Lehnman 1974). The reaction was prepared with a volume ratio between vector

and insert ranging from 1:1 to 1:5, depending on the length of insert. The reaction was

prepared and modified according to Table 22:

Table 22: Ligation preparation

ingredient amount

Ligation buffer 1x

Digested expression vector 1 µl

Digested insert DNA 1 µl /3 µl /5 µl

T4 ligase 200 units

HPLC water Add to 20 µl

Incubation at room temperature for 1-2h or 16°C overnight

For propagation of correctly joined expression vector and insert, the ligation reaction

was transformed into chemically competent E.coli (DH5α).

3.2.1.5 Transformation

Bacterial transformation of plasmid DNA is a technique to amplify generated

expression vectors that include the desired insert sequence. An E.coli strain DH5α,

made chemically competent by the rubidium chloride method (Hanahan, Jessee, and

Bloom 1991) is forced to take up and incorporate exogenous DNA by the heat shock

method. The DNA is then amplified through bacterial replication and selected using

resistance encoded by the expression vector. The desired plasmid DNA can be stored

by freezing glycerol stocks of transformed bacteria stored at -80°C.

For this project, 50 µl of competent DH5α cells were placed on ice for 10 to 15 min,

then incubated with either 20 µl of ligation preparation or 50 µl of DpnI digested

mutagenesis preparation for 20 to 30 min on ice and afterwards the heat shock was

applied for 90 sec at 42°C. In the following, the cells chilled for 2 min on ice and 250 µl

pre-warmed SOC-medium was added. The mixture was incubated while shaking for

30 to 60 min at 37°C. Afterwards cells were plated on agar plates supplemented

selective antibiotics (ampicillin or kanamycin) and incubated at 37°C overnight. A

successful transformation yielded colonies of clones, which were picked to inoculate

mini preparations of 5 ml LB-medium supplemented with selective antibiotics

overnight. The plasmid DNA was isolated using “Pure Yield™ Plasmid Miniprep

System” (Promega) according to manufacturer’s protocol. The correctness of inserted

DNA sequences was checked using sequencing PCR. Upon conformation, midi

preparations (200 ml) were inoculated overnight and the 500 µl culture was used to

generate glycerol stocks. DNA isolation was performed using “Nucleobond Xtra Midi

Prep” (Macherey & Nagel).

3.2.2 Cell culture methods

3.2.2.1 Cultivation of cell lines

The cultivation of mammalian cell lines took place in tissue culture treated polystyrene

flask (75 cm2) at standard cell culture conditions of 37°C and 5% CO2 in appropriate

cell culture media.

Passaging was accomplished by washing the cells with 10 ml Dulbecco’s PBS and

trypsinizing with 1x Trypsin/EDTA for 3 to 6 min at 37°C until cells are detached. Light

tapping was performed to support detachment and cells were then harvested by re-

suspended with complete cell culture media added to 10 ml volume. As an ingredient

of the media, serum oversaturates trypsin and stops the enzymatic digestion of cell

matrix. The re-suspended cells were then split and seeded.

For performance of a functional assay, cells were counted using a cell counting device.

The appropriate cell concentration was set and the cells were seeded in the respective

dish format. For TH-signaling assays, cells were seeded with MEM, containing

charcoal-treated and therefore TH-free FBS (5%).

For storage of cell aliquots, the cell suspension was centrifuged at 200 g for 10 min,

the pellet was re-suspended using freeze media containing FBS with 10% DMSO and

placed into a cryo tube. The freezing process was conducted using a “Mr. Frosty”

cooling device for gradually freezing and placed at -80°C. For re-cultivation, cells were

thawed in a water bath at 37°C shortly and seeded in pre-warmed complete cell culture

media in a 75 cm2 culture flask. The next day, the media was exchanged to remove

DMSO and cell debris.

3.2.2.2 Transfection

Transfection is a method to incorporate expression vectors into mammalian cells in

order to create cells overexpressing a desired protein. Depending upon whether a

selective pressure is applied (e.g. using a selective antibiotic) or not, the transfection

will be permanent (stable transfection) or only for a few days (transient transfection).

For cell lines that are easy to transfect, transfection reagents can be used that form a

complex with the DNA to mask its negative charge. This results in the DNA entering

the cell and an expression in a high rate.

For all studies conducted in this project, only transient transfections were performed.

3.2.2.2.1 Metafectene™

Functional studies were performed in HEK293 or COS7 cells cultured in MEM (Earl’s

minimal essential medium) supplemented with 5% FBS (fetal bovine serum) and 1%

non-essential amino acids or DMEM (Dulbecco’s MEM) containing 10% FBS and 100

U/ml penicillin, 100 µg/ml streptomycin respectively. For that, cells were seeded into

96 well plates at a cell density of 1.5x104 cells/well and 5% CO2. For HEK 293 cells,

plates were pretreated with poly-L-lysine (1:2 in PBS) for 10 min and washed twice

with PBS. Transfection was performed with 45 ng plasmid DNA/well and 0.45 µl

Metafectene™/well 24h after seeding. The transfection reagent was premixed in half

of the needed volume of serum-free medium and incubated for 5 min at room

temperature. The DNA was added to the other half of the volume and the prepared

Metafectene™ mix was added to the DNA preparation. After incubation for 15-20 min

at room temperature, the complete medium was sucked off the cells and replaced with

the transfection mix. Cultivation was continued at 37°C and 5% CO2. For cAMP assays,

cells were transfected with the appropriate GPCR expression vector or empty vector.

For reporter gene assays, equal amounts of the appropriate reporter construct

containing the firefly luciferase gene and the respective GPCR were co-transfected.

For T3 response reporter gene assays, equal amounts of constructs containing rTRα1

have also been co-transfected to have an abundance of TR in the cell.

3.2.2.2.2 FuGene® HD

In comparison to Metafectene™, FuGene® HD has a lower autofluorescence and can

therefore be used for imaging studies that involve fluorescent dyes. This transfection

reagent was used to transfect HEK293 cells on cover slips in a 6 well plate (1.7x105

cells/well) 24h after seeding. Transfection of SNAP-tagged GPCRs was performed

using FuGene® HD in a reagent-DNA-ratio of 2.5:1 and according to manufacturer’s

protocol. In details, 3.3 µg/well DNA was added to Opti-MEM, mixed with 8.3 µl

FuGene® HD and incubated for 5-10 min at room temperature. For transfection, 150

µl of the transfection preparation was added dropwise to each well and cultivation was

continued at 37°C and 5% CO2.

Additionally, FuGene® HD was used for the NanoBRET™ assay. Here, HEK293 cells

in 6 well plates (8.0x105 cells/well) were transfected 4-6 h after seeding in a reagent-

DNA-ratio of 3:1 and according to manufacturer’s protocol. The BRET partners were

kept in a donor-acceptor-ratio of 1:10, diluted in carrier DNA (pGEM3Z, Promega). In

details, a transfection mixture was prepared with 2 µg of HaloTag® plasmid and 0.2

µg of NanoLuc plasmid and 0.2 µg carrier DNA in 100 µl Opt-MEM for each well. To

each mixture, 8 µl FuGene® HD was added and incubated for 10 min. The mixture

was then added dropwise to each well with attached cells and incubated for

approximately 20h at 37°C, 5% CO2 to allow protein expression.

3.2.3 Biochemical methods

3.2.3.1 Characterization of intracellular signaling transduction

GPCR signaling can be influenced by many factors, including fusion with other proteins

or modified ligands. In order to make sure that T3 conjugated peptide ligands are able

to activate a signaling cascade in the same manner as the unmodified ligand, a

functional characterization of the conjugates took place. Additionally, modifications of

any kind can also lead to deviating behavior of GPCRs, known as biased signaling

(Shukla, Singh, and Ghosh 2014). Depending on the type of signaling pathway the

respective receptor was involved, different approaches were used to measure the

accumulation of second messengers.

Next to the pathway activations of GPCRs, intracellular signaling can be activated by

TH. The T3 conjugated ligands are supposed to release the T3 in the intracellular

space, where several signaling cascades can be activated. Therefore, reporter gene

assays were used to analyze different signaling pathways.

3.2.3.1.1 Intracellular cAMP accumulation assay for measurement of GS- and Gi-

activation

Cyclic AMP is a downstream product of the interaction between the GPCR and the GS

protein. It is enzymatically produced by the adenylate cyclase, which is activated by

the α-subunit of the GS protein. For a direct conclusion of receptor activation to

intracellular cAMP concentration, the degrading enzyme phosphodiesterase needs to

be inhibited by adding 3-Isobutyl-1-methylxanthine (IBMX) while receptor stimulation

is in progress. Intracellular cAMP concentration was determined using the

AlphaScreen™ (Amplified Luminescent Proximity Homogeneous Assay, PerkinElmer)

method. This competitive assay is based on the reaction of two different kinds of beads.

The acceptor beads are conjugated to a cAMP antibody and bind to accumulated

cAMP in the sample. The streptavidin-coated donor beads are added together with

biotinylated cAMP, which is then bound to both donor and acceptor beads. Excitation

of the donor beads at 680 nm leads to a conversion of ambient oxygen to a more

excited singlet state. In close proximity, this molecule diffuses across to react with the

acceptor bead, which generates a chemiluminescence emitting light in the range of

520 to 620 nm. Since this reaction only occurs, when donor and acceptor beads are

bound to the same molecule, the biotinylated cAMP is creating signals, being in

competition to the intracellular cAMP. Therefore, this method is an indirect

determination of cAMP concentration (Eglen et al. 2008). The principle of the

AlphaScreen™ is depicted in Figure 8:

After seeding (1.5x104 cell/well) and transfection with vectors containing the respective

GPCR or empty vector using Metafectene™, HEK293 or COS7 cells were stimulated

with various ligands (50 µl/well) in stimulation buffer containing IBMX (500 µM) for 40

min at 37°C. For Gi –activity measurements, cells were stimulated with forskolin in

parallel, which penetrates the cell membrane to activate the adenylate cyclase

unspecifically. Afterwards, cells were lysed using lysis buffer containing IBMX at 4°C

on a horizontal shaker. Cell lysates were stored at -20°C. In order to conclude from

raw emission data to concentration, a cAMP standard is prepared according to

manufacturer’s protocol using lysis buffer. 5 µl of standard and samples are transferred

to a 384 well plate and 10 µl of diluted acceptor beads (1:100 in lysis buffer) are added

in minimal lighting. After incubation for 30 min at room temperature in the dark, 10 µl

of diluted donor beads containing biotinylated cAMP (1:100 for beads, 1:16 for cAMP

in lysis buffer) is added in minimal lighting. After incubation for an additional hour in the

dark, the emission of the acceptor beads are measured using a plate reader (Berthold

Mithras LB 940).

Figure 8: Schematical depiction of the AlphaScreen™ assay used for determination o

intracellular cAMP concentration. Upon stimulation of cells expressing Gs-coupled GPCRs with

their specific ligands, intracellular cAMP increases. IBMX, a PDE inhibitor blocks the degradation

of cAMP in the cell. The determination is then performed with cell lysate after 45 min o

stimulation. The competitive-based AlphaScreen™ assay utilizes a biotinylated cAMP tracer that

competes with intracellular cAMP for the anti-cAMP-conjugated acceptor beads. Only the trace

can be bound by the streptavidin-coated donor beads, which brings donor and acceptor in close

proximity. Excitation of the donor bead leads to the conversion of ambient oxygen to a more

excited singlet state, which reacts with the acceptor beads nearby. This leads to the generation

of chemiluminescent light. Binding of endogenous cAMP does not lead to light emission.

Therefore, the cAMP assay is a competitive measurement. GPCR: G protein-coupled receptor;

AC: adenylate cyclase; cAMP: cyclic AMP; PDE: phosphodiesterase; IBMX: 3-Isobutyl-1-

methylxanthin

3.2.3.1.2 Reporter gene assays to determine PLC activation (read out used for

Gq/11), MAPK-signaling and canonical T3-dependent signaling

Luciferase-based reporter gene assays are used to identify the activation of different

signaling pathways such as Gq/11 and MAPK-signaling. The promotor region of the

luciferase gene contains a response element, which starts the expression of luciferase

when bound by the respective second messenger. For Gq/11/PLC activation, the

response element NFAT RE (nuclear factor of activated T-cell response element) was

used, whereas the ERK1/2 phosphorylation (MAPK pathway), which can be activated

by both GPCR activity and intracellular T3, was determined using the SRE (serum

response element) (Singleton 2010). To analyze the canonical nuclear T3-pathway, a

DR4 (Thyroid hormone receptor response element) was used (Hofmann, Schomburg,

and Kohrle 2009). An overview of response elements used in luciferase-based reporter

gene assays in the course of this project give Table 23:

Table 23: Overview of the pathways investigated with luciferase-based reporter gene assays

Pathway Response

element

Vector References

Gq/11/PLC activation NFAT pGL4.3-luc2 (Boss, Talpade, and

Murphy 1996)

ERK 1/2 phosphorylation SRE pGL4.3-luc2 (Hawes et al. 1995)

Canonical TH-dependent

signaling

DR4 pGL3-luc2 (Hofmann, Schomburg,

and Kohrle 2009)

After seeding (1.5x104 cells/well) and co-transfection with the reporter gene, the

appropriate GPCR or empty vector and thyroid hormone receptor α (only for

investigation of T3-dependent signaling) using Metafectene™, HEK293 cells were

stimulated with the respective ligands (50 µl/well) in MEM without supplements at

37°C. Depending on the assay, stimulation time varied. For GPCR functional

characterization studies, cells were stimulated for 6h. If PTX pretreatment was

necessary, cells medium was changed 16-18h before stimulation to complete media

containing 50 ng/ml PTX. Non-pretreated cells were also changed to complete media

without PTX. For T3-dependent signaling studies, stimulation was either performed in

a time-dependent or concentration-dependent manner. For time course experiments,

cells were incubated with 1 µM T3 for 10, 30, 60, 180 and 360 min. Afterwards, the

cells were washed with 50 µl PBS and incubation continued with 50 µl with T3-depleted

media until 24h were completed to allow sufficient luciferase expression. For

concentration-response experiments, cells were incubated with different

concentrations of T3 for 1h and again washed and incubated as stated above. For

experiments using TH transport blockers, the cells were incubated with the compounds

with or without 1 µM T3 for 1h, followed again by a washing step and incubation for

another 23h in T3-depleted medium. Subsequently, cells were lysed using passive

lysis buffer (Promega) at room temperature according to manufacturer’s protocol. Cell

lysates were stored at -20°C. The measurement of luciferase activity was performed

by transfer of 10 µl cell lysate to black 96 well plates, injection of luciferase substrate

(40 µl/well) and measurement of emission at 560 nm using a plate reader (Berthold

Mithras LB 940).

3.2.3.1.3 NanoBRET™ assay to determine the interaction between GPCR and β-

arrestin2

The interaction with β-arrestins is the initiation of internalization and subsequently

desensitization of GPCRs and their signal transduction. Arrestins can also act as

scaffolding proteins to activate several signaling pathways, which they share with G

protein-dependent signaling, such as MAPK pathways. Therefore, it is difficult to

characterize the activation of β-arrestins by determining downstream messengers. To

study this important step for GPCR internalization, a Bioluminescence resonance

energy transfer (BRET) assay was employed. It is a method to detect protein-protein

interaction in live cells. Figure 9 depicts the concept of a BRET assay:

NanoBRET™, a very recent version of this assay, uses a smaller and brighter bio-

engineered version of the Renilla luciferase, called the NanoLuciferase (NanoLuc) as

an energy donor. The second interaction partner is tagged with a protein tag called

HaloTag®, which can bind covalently to an energy acceptor, the NanoBRET™ ligand

618. In comparison to the usual used BRET methods, this version allows a convenient

determination of background by omitting the addition of the energy acceptor. The

combination of these BRET partners achieved a good spectral separation to avoid

Figure 9: Concept of a bioluminescence resonance energy transfer (BRET) assay. This method

is used to detect protein-protein interactions in lives cells. Putative interaction partners are

fused with either an energy donor, a luciferase or an energy acceptor, a fluorophore. The

luciferase can oxidize its substrate and emits light. In case of an interaction between the two

BRET partners, the energy acceptor is in close proximity to the donor and radiation-free energy

transfer takes place. This leads to the emission of fluorescence from the acceptor fluorophore.

Both the luminescent signal of the donor and the fluorescent light of the acceptor are measured

and the ratio between these signals.

bleed-through of the donor emission into the acceptor channel, increasing the

detection sensitivity of the assay (Machleidt et al. 2015).

After seeding the cells in 6 well plates (8.0x105 cells/well) and transfection using

FuGene® HD, cells were detached approximately 20h after transfection. After a

centrifugation step (130 g, 5 min, RT), the medium containing phenol red was removed,

and cells were re-suspended in Opt-MEM without phenol red containing 4% FBS. The

cell suspension was diluted to 2.2x105/ml and separated into two pools adding either

NanoBRET™ Ligand 618 (1 µl/ml cells of 0.1 mM solution) or the same amount of

DMSO to the cells. Afterwards, cells were reseeded into opaque white 96 well plates

(90 µl/well) in triplicates for each condition tested and let cells attach about 4-6h at

37°C and 5% CO2.

Due to spatial configurations of the putative interaction partners, not every combination

of tagged donor/acceptor pair is optimal for BRET measurements. Therefore, the ideal

combination needed to be established for every investigated GPCR, before

concentration-response assay could be performed. Here, all available tag variants of

receptor and arrestin was combined and stimulated with 1 µM of agonist for 5 min.

Since the interaction between receptor and β-arrestin2 is located in the intracellular

part, the receptors were tagged at the C-terminus only, limiting the combination of

donor/acceptor pairs. The BRET pair with the highest ligand-promoted NET BRET was

chosen for further assays. For concentration-dependent experiments, cells were

incubated with a serial dilution of compound at 37°C for 5 min.

For measurement, 25 µl/well of Nano-Glo substrate (50 µM) were injected using a

plate reader (Berthold Mithras LB 940). The donor and acceptor emission were

measured at 460 nm and 610 nm respectively. The BRET ratio was calculated using

the formula:

BRETratio emission

emission

Afterwards, the ratio was corrected by subtraction of the background (no acceptor

control) and conversion to milliBRET Units (mBU):

correctedBRETratio󰇟mBU󰇠  󰇛BRETratio BRETratio

󰇜  1000

To calculate the difference between stimulated and basal interaction, the NET BRET

was acquired using the formula:

NETBRET  correctedBRETratio correctedBRETratio

3.2.3.2 HiBiT Internalization assay

The HiBiT Extracellular Detection Assay can be utilized to monitor internalization of

GPCRs. The assay is based on a split luciferase, the NanoLuc, divided into a very

small (eleven amino acids, SmBiT) and a very large part of the protein (LgBiT). The

SmBiT is used as an N-terminal tag at the GPCR and should be expressed very low in

the cell. The transfection was performed with 0.45 ng DNA/well of HiBiT and Carrier-

DNA (pGEM-3Zf(+)) was used to add to 45 ng/Well. Measurement took place after 30

minutes of stimulation with varying concentration of stimulant by mixing LgBiT protein

(1:100) and luciferase substrate (1:50) in substrate buffer. This mixture (50 µl/well) was

injected using a plate reader (Berthold Mithras LB 940) with a low speed at room

temperature and the plate was not shaken to mix. The reagent is slightly denser than

culture media for rapid equilibration of HiBiT-tagged proteins with LgBiT Protein and

the substrate. After four minutes of incubation, luminescence was measured at 460

nm. For a background control, data from cells transfected with pcDNA3 were recorded

and subtracted from the sample data. Figure 10 depicts the principle of this assay.

3.2.4 Imaging methods

3.2.4.1 Imaging of SNAP-tagged GPCRs

Fusion proteins between a GPCR and the SNAP-tag can be imaged using a

fluorescent dye, which covalently binds to the tag. It is a highly engineered version of

AGT (alkylguanine DNA alkyltransferase) and forms a covalent thioester bond with the

dye (Gautier et al. 2008). In the course of this project, GPCRs were tagged N-terminally

with the SNAP-tag. Thus, tagged GPCRs in the plasma membrane had the tag

localized in the extracellular space. Addition of a lipophobic SNAP-dye, that is not able

to overcome the plasma membrane, could only bind to the extracellular tags and

Figure 10: Graphical depiction of the HiBiT Cell surface expression assay. A: When the receptor is

located on the cell surface, the N-terminal HiBiT-tag is situated in the extracellular space. It is then

available for the LgBiT-protein that cannot overcome the cell membrane. Once HiBiT and LgBiT are

complemented, they resemble a functional luciferase that is able to oxidize its substrate and produces

luminescence. B: The target receptor was stimulated with a serial dilution of compounds for 30 min.

Afterwards, the LgBiT and substrate were injected into the wells and luminescence was measured using

a plate reader after 4 min of incubation.

therefore only receptors localized on the cell membrane were labeled. After the

addition of an agonist, internalization of the GPCR is triggered. This process could be

detected by microscopy in a time-dependent manner.

48 h after transfection, cells were incubated with SNAP surface A488 (1:500 in HEK

complete media) for 30 min at 37°C. Afterwards, cells were washed briefly with DMEM

three times and stimulated with appropriate ligand or conjugate (1 µM) for various time

periods at 37°C. To stop the stimulation, cells were washed with PBS three times for 5

min and fixed with 4% paraformaldehyde (PFA) for 20 min at room temperature. After

an additional two washing steps with PBS for 5 min, cells were mounted with mounting

media including DAPI (Vectastain) and dried at 4°C overnight. In order to display the

internalization process, microscopy was applied (Zeiss Axiovert). Images were edited

using ImageJ software.

3.2.5 Cell viability assay

Treatment with various compounds can lead to poor cell viability due to cytotoxicity. To

measure survival of cells during an assay, cell viability assays can be utilized. Here,

the CellTiter 96® AQueous One Solution Cell Proliferation Assay (MTS) from Promega

was used. Cells were treated as described for luciferase-based reporter gene assays

with additional non-transfected cells as well as just medium for background controls.

After stimulation, 10 µl solution per well was added and incubated at 37°C for an hour.

Calorimetric measurements were carried out with a plate reader (Anthos reader 2000)

at 492 nm with a reference wavelength of 620 nm. Background values were subtracted

from the sample values and data was normalized to mean of background.

3.2.6 Statistical analysis

The statistical analysis of data collected from experiments is necessary to evaluate the

results and be able to discuss and conclude a possible explanation. In the course of

this project, statistical analysis was performed using the GraphPad Prism 6 software.

For Concentration-response experiments, X-values were transformed into log10 (x) and

non-linear regression was applied using the equation “log(agonist) vs. response (three

parameters)” for stimulants to calculate EC50 or IC50 and span values or “log(inhibitor)

vs. response (three parameters)” for blockers.

To evaluate the significant difference between transfections, treatments or time points,

a two-way ANOVA (analysis of variance) with multiple comparison was implemented

with an α = 0.05. Statistical significance assessed by the p-value was indicated with *

for p≤0.05, ** for p≤0.01 and *** or ### for p≤0.001.

4 Results

The result part of this project is organized in two main subchapters. First, the findings

of the functional characterization are depicted for both investigated conjugates, GLP1-

T1 and OT-T3. Here, the focus will be on activation of signaling cascades as well as

internalization processes. These actions are initiated by the peptide part of the

conjugates. The second part will then present the results obtained from TH signaling

assays, initiated by the T3 part of the conjugate. The general structure of the result

part is depicted in Figure 11:

4.1 Functional characterization of the conjugates at their respective GPCR

4.1.1 Activation of GLP1R signaling

The prerequisite of the “Trojan horse”-paradigm is internalization of a GPCR for which

the ligand was used to couple to T3. The process of GPCR internalization and

subsequently desensitization for extracellular stimuli is preceded by receptor

activation. Therefore, the activation of the GLP1R by GLP1-T3 is crucial for the

channeling of T3 into the cell. Here, we performed functional characterization studies

in order to determine whether GLP1-T3 activates the signaling cascades of GLP1R,

and subsequently internalization, in the same manner as the endogenous ligand GLP1.

GLP1R is part of the GPCR family B. Its main pathway is reported to be G

coupling,

leading to an increase in intracellular cAMP concentration (Skoglund, Hussain, and

Holz 2000). HEK293 cells were transfected with either a mock control or GLP1R,

stimulated with T3, GLP1 or GLP1-T3 (each 1 µM) and intracellular accumulation of

cAMP was determined using the AlphaScreen™ assay. GLP1-T3 was able to activate

the GLP1R in the same manner as GLP1, leading to an accumulation for both of

around 12 fold over the basal control (GLP1: 12.24 ± 2.87, GLP1-T3: 12.79 ± 2.20)

(Figure 12A). Additionally, GLP1R of rattus norvegicus is reported to induce calcium

Figure 11: The general structure of the results part. First, findings of the functional

characterization will be shown, afterwards results for TH signaling assays are presented.

response as well as ERK1/2 phosphorylation, suggesting coupling to Gq/11 (Wheeler et

al. 1993; Montrose-Rafizadeh et al. 1999). Therefore, we measured NFAT as read-out

for Gq/11 phospholipase C (PLC) activation, and SRE for activation of the MAPK

pathway. In these experiments, GLP1R exhibited a mean basal activity of 14.36 ± 0.84

fold over basal control in PLC activation that could not be increased further with

stimulation with either T3, GLP1 nor GLP1-T3 (Figure 12B). In HEK293 cells, we were

not able to determine significant ERK phosphorylation at the human GLP1R (Figure

12C).

Figure 12: GLP1-T3 is able to activate GLP1R in the same manner as GLP1. Functional studies

of the GLP1R (in green) to induce cAMP accumulation (A), Calcium mobilization (B) and ERK

phosphorylation (C) were performed in comparison to mock transfected cells (in blue). Cells

were stimulated with T3, GLP1 or GLP1-T3 (each 1 µM). Data represent four independent

experiments, each performed in triplicates. For statistical analysis, a two-way ANOVA with

multiple comparison was performed comparing different transfections. Values represent mean

± SEM

4.1.2 Initiation of GLP1R internalization

In order to monitor receptor internalization, GLP1R tagged with a SNAP-tag was

transfected was into HEK293 cells. The SNAP-tag is a recombinant derivate of the O6-

guanine nucleotide alkyl transferase, which covalently binds to benzyl guanine

substrates conjugated to fluorescent dyes.

Living cells were stained with SNAP Surface®-A488, a dye that cannot overcome the

plasma membrane, before stimulation with GLP1 or GLP1-T3 for 30 min. In the basal

state, receptors are distributed evenly throughout the cell. Pre-stained receptors on the

cell surface accumulated within the cell plasma upon activation (Figure 13A).

Additionally, we investigated β-arrestin2 recruitment using BRET. The interaction with

arrestins is known to mediate receptor internalization by clathrin-coated pits and also

scaffolding G protein independent signaling, such as ERK-phosphorylation (Moore,

Milano, and Benovic 2007). Class B receptors, such as GLP1R, are reported to bind

to both β-arrestin1 and β-arrestin2 with high affinity (Tobin, Butcher, and Kong 2008).

Conjugate-dependent recruitment of β-arrestin2 would therefore indicate receptor

internalization. Before a concentration-dependent interaction was investigated, the

optimal BRET partner combination needed to be established. Therefore, all available

combination of BRET pairings were tested to find the combination with the highest

ligand-promoted NET BRET. The results are depicted in the supplements (Figure S

2A and B). Afterwards, a concentration-response assay was performed with the best

BRET pair, which was ARRB2-NL and GLP1R-HT with a NET BRET of 14.18 ± 0.39

mBU. Both unconjugated GLP1 and GLP1-T3 were able to induce β-arrestin2

interaction in a similar manner with a higher EC

for GLP1-T3 (188.0 ± 0.15 nM)

compared to the EC

of GLP1 (50.06 ± 0.18 nM) (Figure 13B).

Furthermore, internalization would reduce the amount of receptors present on the

plasma membrane that can be monitored using a cell surface expression assay. GLP1,

as well as the conjugate were able to reduce the amount of GLP1R expressed on the

cell surface in a concentration-dependent manner with a surface expression decrease

of 83.93 ± 5.44% for GLP1 and 96.63 ± 5.23% for GLP1-T3. The compounds had a

very similar IC

with 1.78 ± 0.15 nM for GLP1 and 2.31 ± 0.13 nM for GLP1-T3 (Figure

13C).

Figure 13: GLP1-T3 is able to initiate GLP1R internalization in a similar fashion to GLP1.

Imaging studies in (A) to monitor receptor internalization in HEK293 cells pre-stained with SNAP

Surface-A488 (green) stimulated with GLP1 (1 µM) or GLP1-T3 (1 µM), and subsequently fixated.

Nuclei were stained with DAPI (blue) and cell outline was added digitally (white dashed line). In

the basal state, SNAP-tagged GLP1R are evenly distributed over the cell surface immediatel

after live staining (t = 0 min) and after an hour of no stimulation (t = 60 min). Following

stimulation with either GLP1 or GLP1-T3 for 30 min, pre-stained receptors accumulated in the

cytosol due to internalization processes.

β-arrestin2 recruitment in (B) was measured using NanoBRET™ in HEK293 cells afte

stimulation with either GLP1 (1 µM) or GLP1-T3 (1 µM) for 5 min and determined as the difference

between basal and stimulated BRET ratios. Studies reveal a similar interaction profile after the

stimulation of GLP1 (blue) and GLP1-T3 (yellow). The recruitment of β-arrestin2 can hint to the

initiation of internalization processes after stimulation with specific agonists. Data represent

three independent experiments, each performed in triplicates. Values represent mean ± SEM.

GLP1R shows a concentration-dependent decrease of cell surface expression after challenge

with either GLP1 (blue) or GLP-T3 (yellow) (each 1 µM) in (C) Cells surface expression as

measured using a small N-terminal protein tag that is part of a split luciferase. N-terminally

tagged receptors were stimulated with GLP1 (1 µM) or GLP1-T3 (1 µM) for 30 min. Data represent

three independent experiments, each performed in triplicates. Values represent mean ± SEM.

Basal GLP1 GLP1-T3

t = 30 min t = 30 min

Basal

t = 60 min

4.1.3 Activation of OTR and V1AR

Functional characterization of OT-T3 was performed on two target receptors, the OTR

and the V1AR. These receptors have different affinities towards OT, with Ki(OTR) lower

than Ki(V1AR) (Åkerlund et al. 1999). Both GPCRs are reported to activate the Gq/11

protein as their main signaling pathway. Additionally, OTR initiates Gi protein

activation, lowering cAMP production in the cell as well as increasing ERK

phosphorylation. In fact, the latter is reported to be partly PTX-sensitive and partly PLC

inhibitor-sensitive, meaning both activated G proteins lead to ERK phosphorylation

with different biological functions depending on the cell type (Ohmichi et al. 1995;

Strakova et al. 1998). Similarly, Calcium mobilization/PLC activation following OT

stimulation has been shown to also be a mixed result of the βγ-subunits of Gi and Gq/11

due to decreased, but not fully abolished signal after PTX-treatment (Molnár and

Hertelendy 1990; Phaneuf et al. 1993). For V1AR, ERK phosphorylation following an

OT challenge was reported in CHO cells (Tahara et al. 1999).

For the OTR, cAMP accumulation assay (AlphaScreen™) was performed with cells

pretreated with forskolin (50 µM) to activate the adenylate cyclase, so a decrease in

intracellular cAMP concentrations would imply Gi protein activation. The stimulation

with OT lead to significant decrease of 66.55 ± 11.12% in OTR-overexpressing cells in

comparison with mock transfected cells. OT-T3 was not able to decrease cAMP

content in OTR expressing cells (Figure 14A) indicating that OT-T3 is unable to

activate Gi mediated inhibition of AC.

Calcium mobilization/PLC-activation was measured using the NFAT-luciferase

reporter assay, whereas ERK phosphorylation was assessed by a SRE-luciferase

based reporter. To gain more information about the G protein involvement in the PLC

activation and ERK phosphorylation signals we obtained, cells expressing OTR were

additionally pretreated with PTX, a Gi inhibitor. For OTR, calcium mobilization was

identical for both agonists with an EC50 of 3.77 ± 0.18 nM for OT and 2.87 ± 0.21 nM

for OT-T3. Treatment with PTX significantly reduced signal to around 60% of non-

treated cells for both agonists in a similar fashion. However, there was no shift in EC50

compared to non-treated cells with 4.18 ± 0.2 nM for OT + PTX and 2.76 ± 0.23 nM

for OT-T3 + PTX (Figure 14B).

ERK phosphorylation by activated OTR was induced by stimulation with OT (EC50 2.11

± 0.32 nM). Pretreatment with PTX resulted in the abolishment of the Gi effect with an

EC50 of 0.8 ± 0.43 nM and a significantly reduced signal to around 47% of OT

stimulation. Challenge with OT-T3 mimics this attenuated effect of PTX treatment with

an EC50 of 1.23 ± 0.30 nM. No additional decrease of signal was obtained for OT-T3

stimulation when pretreated with PTX (EC50 of 1.19 ± 0.37 nM) (Figure 14C), indicating

that ERK phosphorylation resulting from OT-T3 challenge is solely due to Gq/11

activation, but not Gi.

Figure 14: Functional studies on the OTR showed biased agonism of OT-T3.

For Gi activation in (A), COS7 cells were transfected with OTR (red) or empty vector (mock, blue)

and stimulated with either T3, OT, OT-T3 (each 1 µM) or not (basal) while treated with forskolin

(50 µM) to activate adenylate cyclase. Data represent four independent experiments, each

performed in triplicates. For statistical analysis, a two-way ANOVA with multiple comparison was

performed, comparing different transfections. Values represent mean ± SEM. Statistical

significance is indicated by *p≤0.05.

Calcium mobilization was assessed by a NFAT-reporter in (B) and ERK phosphorylation by a

SRE-reporter in (C). HEK293 cells were stimulated with OT (yellow) or OT-T3 (blue) (each 1 µM)

and pretreated with PTX (red and violet) or not. Data represent five to nine independent

experiments, each performed in triplicates. For statistical analysis, a two-way ANOVA with

multiple comparison was performed, comparing different treatments. Values represent mean ±

SEM. Statistical significance is indicated by *** p≤0.001

Determination of calcium mobilization and ERK phosphorylation at the V1AR was

executed in the identical fashion to OTR. The NFAT-reporter showed identical

signaling properties for both agonists on the receptor with an EC50 of 6.83 ± 0.31 nM

for OT and 6.64 ± 0.30 nM for OT-T3 (Figure 15A). ERK phosphorylation on the other

hand was significantly increased for the highest OT-T3 concentration of 10-6 M and a

shifted EC50 from 12.69 ± 0.48 nM for OT to 66.4 ± 0.31 nM for OT-T3 (Figure 15B).

Figure 15: Functional characterization on the V1AR showed biased agonism of OT-T3.

HEK293 cells overexpressing V1AR were stimulated with either oxytocin or OT-T3 (each 1 µM)

and calcium mobilization (A) and ERK phosphorylation (B) were measured using a reporter gene

assay. Data represent three independent experiments, each performed in triplicates. Values

represent mean ± SEM. For statistical analysis, a two-way ANOVA with multiple comparison was

performed, comparing different treatments, Statistical significance is indicated by *p≤0.05

4.1.4 Internalization of OTR and V1AR

Both target receptors were investigated for β-arrestin2 recruitment after OT-T3

stimulation using BRET again as an indication for receptor internalization. For

establishment of the optimal BRET pairing, preliminary experiments were performed

and the combination with the highest ligand-promoted NET BRET was chosen for

concentration-dependent experiments. The BRET pairs chosen was ARRB2-NL and

OTR-/V1AR-HT with a NET BRET of 5.13 ± 0.63 mBU for OTR and 5.18 ± 0.10 mBU

(Figure S 2C-F). In these assays, OTR showed identical agonist-promoted BRET

signals for both OT and OT-T3 (EC50 6.71 ± 0.15 nM for OT and 5.47 ± 0.11 nM for

OT-T3) (Figure 16A). Surprisingly, at the V1AR the signal from OT-T3 was reduced to

around 9% of maximal OT recruitment of β-arrestin2 (EC50 5.27 ± 0.14 nM for OT). For

OT-T3 EC50 calculation with appropriate accuracy is not possible. (Figure 16B).

Figure 16: Stimulation of OTR and V1AR with OT and OT-T3 resulted in different β-arrestin2

recruitment pattern. HEK293 cells overexpressing β-arrestin-NanoLuc fusion protein and OTR

(A) or V1AR (B) C-terminally tagged with the protein tag HaloTag, binding to the energy acceptor,

were stimulated with either OT (yellow) or OT-T3 (blue) in different concentrations and BRET

ratios were measured after 5 min of stimulation. Data represent three independent experiments,

each performed in triplicates. Values represent mean ± SEM.

4.1.5 Summary of the functional characterization

In summary, two different T3-conjugates have been investigated. They were

characterized by their ability to initiate receptor activation and internalization, as well

as TH-dependent signaling in presence of the receptor as proof for the “Trojan Horse”-

like mechanism. The first investigated conjugate GLP1-T3 was used to establish a

toolbox for the investigation of future conjugates. It was able to activate receptor

signaling, especially the main Gs-pathway, and internalization, by the means of β-

arrestin2 recruitment and decrease of receptor on the cell surface after stimulation in

the identical manner to the unconjugated GLP1.

The second conjugate OT-T3, which was designed for correction of symptoms of

MCT8-deficiency, has been tested using the toolbox established with GLP1-T3. In

comparison to the unconjugated peptide, at the OTR OT-T3 was able to activate the

main pathway Gq/11, but not the Gi-pathway. This led to a reduced phosphorylated ERK

signal comparable to the PTX-treated receptor, confirming the missing Gi coupling. The

main signaling pathway, Gq/11 at the V1AR was not attenuated and ERK

phosphorylation was even significantly increased compared to unconjugated peptide

in the highest concentration. Nevertheless, this receptor did not show β-arrestin2

recruitment when stimulated with OT-T3 in comparison to a specific interaction after

stimulation with unconjugated OT.

The functional characterization of the conjugates resulted in the findings summarized

in Table 24 for GLP1-T3 and Table 25 for OT-T3. A check mark indicates the activation

of the respective signaling pathway or the internalization process, whereas a cross

illustrate a missing signal for the respective assay. Arrows indicate the decrease

(downwards) or increase (upwards) of signal in comparison to the unconjugated ligand.

Table 24: Summary of the results for the functional characterization of GLP1-T3 at the GLP1R

Ligand Signaling cascades Internalization

Gs PLC

activation

ERK phos-

phorylation

Visually

(SNAP-

tag)

β-arrestin2

recruitment

Decrease of

cell surface

expression

GLP1 ✔ ❌ ❌ ✔ ✔ ✔

GLP1-T3 ✔ ❌ ❌ ✔ ✔ ✔

Table 25: Summary of the results for the functional characterization of OT-T3 at the OTR and

V1AR

4.2 Determination of TH-dependent signaling of the conjugates

In addition to activation of signaling and internalization of the target receptors, it was

determined whether the conjugates were able to activate canonical and non-canonical

signaling of thyroid hormones. For the canonical signaling, it is known that T3 binds to

the TRα1 and relocate into the nucleus. Here, the complex binds to thyroid hormone

receptor response elements (TRE), which starts the expression of target genes. To

investigate whether T3 is released by the conjugate after internalization and trigger

TH-dependent signaling – in other words, whether the “Trojan Horse”-like mechanism

works – we used a reporter gene that expressed firefly luciferase under the control of

a TRE of four direct repeat (DR4) (Hofmann, Schomburg, and Kohrle 2009). In the

following part, results from these reporter gene assays are presented.

Ligand Target

receptor

Signaling cascades Internalization

Gi PLC

activation

ERK phos-

phorylation

β-arrestin2

recruitment

OT OTR

✔ ✔ ✔ ✔

V1AR - ✔ ✔ ✔

OT-T3 OTR ❌ ✔ ↓ ✔

V1AR - ✔ ↑ ❌

4.2.1 Finding a suitable inhibitor for TH transport in HEK293 cells

Membrane transporters are essential for cell regulations and therefore ubiquitous in all

cell types (Giacomini et al. 2010). Especially since THs are important for cell function,

various sets of transporters, able to transport THs, are endogenously expressed in

different cell types (Friesema et al. 2005). In a functional assay for TH-dependent

signaling, cell lines like HEK293 cells always show high responses to T3 stimulation.

For most of the membrane transporter, that are able to transport THs, one or more

blocker has been found to inhibit the influx and efflux of extracellular T3 or T4. In order

to find a suitable TH transporter blocker, we used published inhibitors for all known TH

transporters. The drug probenecid is used to increase uric acid excretion in the kidney

by inhibition of OATPs (Ho et al. 2000). It is also known to block OATPs in the rat

choroid plexus (Pritchard et al. 1999) and therefore is known to block OATP1C1. The

tricyclic anti-depressant desipramine (DMI) is known for the inhibition of LATs and

MCT8 and MCT10 (Taylor and Ritchie 2007; Roth, Kinne, and Schweizer 2010). LAT1

and 2 are reported to be blocked by 2-aminobicyclo-(2,2,1)-heptane-2-carboxylic acid

(BCH) (Christensen 1990). The TH derivate 3-iodothyronamin (3-T1AM) has been

shown to block T4 and/or T3 transport of OATP1A2, OATP1C1 and MCT8 (Ianculescu

et al. 2010), whereas Bromosulphtalein (BSP) is a known blocker for MCT8 (Friesema

et al. 2003; Jayarama-Naidu et al. 2015). Silychristin (SY), a flavonolignan derived from

the milk thistle, is a very recent addition to the specific MCT8 blockers (Johannes,

Jayarama-Naidu, et al. 2016).

As a preliminary experiment, these blockers were tested in for canonical TH signaling

in a reporter gene assay (Figure 17A). Cells were co-stimulated with T3 and the

different blockers with ten to fifty-fold lower concentrations than previously published,

since incubation times were increased from a couple of minutes to an hour (Roth,

Kinne, and Schweizer 2010; Zhang et al. 2010; Taylor and Ritchie 2007).

Only one blocker was able to inhibit TH signaling in HEK293 cells, the most

promiscuous DMI. It could significantly decrease the cells response to T3 compared to

the untreated cells from 10.92 ± 1.01 to 2.62 ± 0.55 fold. Visual control of the cells

through the microscope during the assay presented the question of whether the

decrease in signaling is due to cytotoxicity of the compound rather than inhibition of

transport. Therefore, a cell viability test was performed (Figure 17B). In concentrations

higher than 50 µM, cell viability was diminished. To find a concentration that is not yet

toxic, but sufficiently inhibit TH-dependent signaling, both datasets are depicted

together in Figure 17C. For ascending concentrations of DMI both cell response and

viability decreased in a similar fashion.

Figure 17: Preliminary assays to determine a suitable blocker for endogenous expressed TH

transporter. A: Luciferase based reporter gene studies to compare the different inhibitors that

were published as TH transporter blocker. B: Viability test to ensure cell viability at high

concentrations of DMI. C: Combined depiction of canonical TH-dependent signaling and cell

viability after incubation with varying concentration of DMI. Data represent three to four

independent experiments, each performed in triplicates. For statistical analysis, a two-way

ANOVA was performed, comparing different treatments with DMI. Values represent mean ± SEM.

Statistical significance is indicated by *p≤0.05, ***p≤0.001

DMI: desipramine; BCH: 2-aminobicyclo-(2,2,1)-heptane-2-carboxylic acid; 3-T1AM: 3-

iodothyronamine; BSP: bromosulphtalein; SY: silychristin

4.2.2 Determination of TH-dependent signaling after challenge with GLP1-T3

In order to ascertain whether the conjugate can activate canonical and non-canonical

pathways of TH action, both signaling cascades were investigated by the

aforementioned reporter gene assays. The “Trojan Horse”-like mechanism should only

work in the presence of the GLP1R and cells that do not express the receptor should

not experience T3-dependent signaling after stimulation with GLP1-T3.

Cells were stimulated with GLP1-T3 or T3 in presence or absence of GLP1R for

various time periods and concentrations. A significant increase of T3-dependent

signaling in the presence of GLP1R was determined after GLP1-T3 challenge for more

than 60 min, in comparison to cells that do not express GLP1R. More specifically,

signaling increased from 1.527 ± 0.233 x106 RLU in the basal state to 11.251 ± 1.360

x106 RLU after 6h of incubation with GLP1-T3 (Figure 18A). This effect has been

shown to be concentration-dependent, though significance between transfections was

only present for the highest concentration of 1 µM GLP1-T3 raising from 0.817 ± 0.118

x106 RLU to 3.344 ± 0.348 x106 RLU (Figure 18B, significance indicated with *).

Additionally, the mock control showed a significant increase of signaling for this

concentration as well (1.922 ± 0.255 x106 RLU, significance indicated with #). The

stimulation with T3 resulted in an immediate high response irrespective of the

expression of GLP1R for different stimulation periods (Figure 18C). Furthermore, it

rose concentration-dependently for both transfections (Figure 18E). Stimulation with

GLP1 did not have any effect on T3-dependent signaling, neither time- (Figure 18D)

nor concentration-dependent (Figure 18F).



Figure 18: Luciferase

based reporter gene

studies of nuclear

response to TH

(canonical pathway) in a

time- and concentration-

dependent manner. Cells

have been co-transfected

with the luciferase

reporter gene DR4, the

thyroid hormone receptor

α (TRa) and either GLP1R

(red) or empty vector

(blue). Cells were

stimulated with GLP1-T3

(1 µM) in (A), T3 (B) or

GLP1 (C) for various

periods (10 min, 30 min,

1h, 3h, 6h) or in different

concentrations of GLP1-

T3 (D), T3 (E) or GLP1 (F)

for one hour. Data

represent three

independent

experiments, each

performed in triplicates.

For statistical analysis, a

two-way ANOVA with

multiple comparison was

performed, comparing

different transfections (*)

or different

concentrations (#).

Values represent mean ±

SEM. Statistical

significance is indicated

by *p≤0.05, **p≤0.01,

***/###p≤0.001



In addition, we measured non-canonical signaling of rTRα1 that results in a rapid

phosphoinositide 3-kinase activation and mitogen-activated ERK1/2 phosphorylation

and used MAPK activation as read-out (Makino et al. 2008; Kalyanaraman et al. 2014).

Stimulation with GLP1-T3 resulted a significant increase in T3-dependent signaling in

the presence of GLP1R in comparison to the absence of GLP1R (mock transfection

with empty vector). More specifically, it increased from 0.3548 ± 0.4468 x106 RLU in

the basal state to 4.947 ± 0.641 x106 RLU after 6h of stimulation(Figure 19A). As a

control, cells were also treated with T3, which lead to an immediate response with no

difference in T3-dependent signaling for cells expressing either GLP1R or empty vector

(Figure 19B). Additionally, cells treated with GLP1 showed no response in T3-

dependent signaling (Figure 19C).

Since we were not able to find a transporter blocker that would inhibit endogenous TH

transporter in our cell system without also having inhibitory effects on cell viability, no

blocking experiments were performed, which would confirm the entrance of T3 through

the receptor.

Figure 19: Luciferase based reporter gene studies of non-canonical response to TH in a time -

dependent manner. Cells have been co-transfected with the luciferase reporter gene DR4, the

thyroid hormone receptor α (TRa) and either GLP1R (red) or empty vector (blue). Cells were

stimulated with GLP1-T3 (1 µM) in (A), T3 (B) or GLP1 (C) for various periods (10 min, 30 min, 1h,

3h, 6h). Data represent three independent experiments, each performed in triplicates. For

statistical analysis, a two-way ANOVA with multiple comparison was performed, comparing

different transfections. Values represent mean ± SEM. Statistical significance is indicated by

*p≤0.05, **p≤0.01

4.2.3 Determination of TH-dependent signaling after challenge with OT-T3

Similar to the investigation of GLP1-T3, T3-dependent signaling after stimulation with

OT-T3 was investigated in the absence and presence of both target receptors, OTR

and V1AR. Again, HEK293 cells were used and a time-dependent assay was

employed. The treatment with OT-T3 of OTR expressing cells did not result in a

significant increase in canonical signaling compared to mock expressing cells (Figure

20A). Similarly, no significant change was measured for V1AR expressing cells,

although it showed a tendency to higher values, which were not significant in the

statistical analysis (two-way ANOVA, comparing the transfections for every time point)

(Figure 20B). As a positive control for this assay, cells were also treated with T3 alone.

For both receptor transfections, stimulation with T3 resulted in similar values than

treatment with OT-T3 with no significant difference for OTR expressing cells in

comparison with mock cells (Figure 20C) and also again not for V1AR expressing

cells, but with a higher tendency (Figure 20D). As a negative control, stimulation with

OT alone showed overall low values for both receptors (Figure 20E and

F).Conspicuously, especially when compared to the results from GLP1-T3, values of

cells treated with OT-T3 were in the same range with cells treated with T3 for all

transfections (around 4 x 106 to 1 x107 RLU). Since these results raised the question

of whether or not the conjugate is stable enough in cell culture conditions,

concentration-dependency and non-canonical signaling of OT-T3 was not

investigated.

Figure 20: Luciferase

based reporter gene

studies of canonical

response to TH in a time -

dependent manner. Cells

have been co-transfected

with the luciferase

reporter gene DR4, the

thyroid hormone receptor

α (TRa) and either OTR

(green), V1AR (red) or

empty vector (blue). Cells

expressing OTR were

stimulated with OT-T3 (1

µM) in (A) or T3 (C).

Stimulation of cells

expressing V1AR were

treated the same with OT-

T3 in (B) and T3 in (D). All

stimulation took place for

various periods (10 min,

20 min, 30 min, 1h, 3h, 6h).

Data represent three

independent experiments,

each performed in

triplicates. For statistical

analysis, a two-way

ANOVA with multiple

comparison was

performed, comparing

different transfections.

Values represent mean ±

SEM.

4.2.4 Summary of the TH signaling assay results

Additionally, I was able to show increased canonical and non-canonical TH signaling

in the presence of GLP1R in comparison to mock transfected cells. These results are

depicted in Figure 21.

When investigating canonical TH-dependent signaling, it was revealed that OT-T3

activated the pathway in the same manner as a stimulation with T3 alone and no

difference between cells that express either one of the target receptors and mock

transfected cells was present. A possible reason is discussed in chapter 5. The results

are summarized in Figure 22.

Figure 21: Graphical depiction of the

results for canonical and non-

canonical TH signaling assays of GLP1-

T3. Cells were transfected with a

luciferase-based reporter, the TRα and

either an empty vector as a mock

control or the GLP1R. Cells were then

challenged with either T3 as a positive

control, GLP1 as a negative control o

GLP1-T3, the conjugate between the

two compounds. Significant (though

not as high as the positive control)

increase of both canonical and non-

canonical signaling for the conjugate

was measured only in the presence o

the receptor GLP1R, whereas the mock

control did not show any significantly

increased values

Figure 22: Graphical depiction

of the results for canonical TH

signaling of OT-T3. Cells were

transfected with a luciferase-

based reporter, the TRα and

either an empty vector as a

mock control or the target

vectors OTR or V1AR. Cells

were then challenged with

either T3 as a positive control,

OT as a negative control or OT-

T3, the conjugate between the

two compounds. For all cell

populations, OT-T3 increased

the TH-dependent signaling in

the same range with no

significant difference between

mock expressing and target

receptor expressing cells.

5 Discussion

In the course of this project two T3-conjugated peptides were characterized and

validated in terms of receptor activation and internalization, as well as TH-dependent

signaling activation. These investigations highlight whether or not the conjugates are

working in the intended way, via the “Trojan-Horse”-like approach, as well as identifying

possible adverse effects even before in vivo investigations. The main aim is to find a

perfect receptor-ligand pair that could be used for MCT8-deficiency treatment with

minimal side effects. Figure 23 depicts an overview of general considerations that

were made about the perfect receptor-ligand pair as well as the aspects that were

investigated in the course of this project.

In the following part, the results from this investigation will be discussed. First, the

results of the functional characterization at their respective target GPCRs will be

examined. Additionally, it focusses on the internalization of the receptor-ligand

complex that follows activation. The second part discusses the activation of TH-

dependent signaling which hints to the liberation of T3 after internalization and can be

regarded as a proof that the “Trojan Horse”-like mechanism is at play.

5.1 Functional characterization of the conjugates at their respective target

GPCR

The “Trojan Horse”-like mechanism utilizes the internalization mechanism of GPCRs.

This process is part of the desensitization mechanism of receptors that prevents

overstimulation. Hence, a receptor stimulation precedes the endocytosis of the whole

receptor-ligand complex. Afterwards T3 should be released within the cytoplasm and

Figure 23: Graphical overview of the considerations regarding the perfect receptor-

ligand pair as well as an outline of the aspects investigated in the course of this

project.

would be able to initiate TH-dependent signaling within the cell. A characterization of

the conjugates must therefore happen with a view to functionality on their respective

receptors by their peptide part, and to the activation of TH signaling by their T3 part.

First, the functional studies were performed including activation of signaling cascades

and internalization.

5.1.1 GLP1-T3 activates GLP1R

For the functional characterization of GLP1-T3, the unconjugated GLP1 was used as

a positive control. The activation by GLP1-T3 should mimic the activation of the

endogenous ligand GLP1. Three signaling cascades have been reported for GLP1R:

cAMP as the main pathway, IP3/PLC-activation and MAPK pathway ERK1/2, all of

which have been investigated to identify possible adverse effects of the conjugate.

5.1.1.1 Gs pathway at the GLP1R

In the main pathway, the Gs coupled cascade (Baggio and Drucker 2007), GLP1-T3

and unconjugated GLP1 show a similar cAMP concentration that was not significantly

different from one another. Therefore, the conjugation of T3 to GLP1 does not influence

the pathway that is most important for the receptor activation and a major adverse

effect on the receptor can be excluded.

5.1.1.2 Gq/11/PLC-activation pathway at the GLP1R

There are hints from the literature that GLP1R is also able to activate Gq/11/PLC

activation pathway (Montrose-Rafizadeh et al. 1999; Thompson and Kanamarlapudi

2015), although it is unclear, whether this effect is due to overexpression in in vitro

assays and/or activation of PLC and plasma membrane calcium channels by the cAMP

pathway (Holz, Leech, and Habener 1995). Nevertheless, a basal activity was not

reported in any study and therefore, might be a pleiotropic effect of the HEK cells used

for this project. Neither GLP1 nor GLP1-T3 were able to increase this basal activity

upon stimulation. This leads to the conclusion that the addition of T3 to GLP1 does not

lead to diverse activation of the Gq-associated PLC activation pathway and adverse

effects can also be excluded for this signaling cascade.

5.1.1.3 MAPK pathway activation at the GLP1R

Additionally, ERK1/2 phosphorylation was tested, since it was also reported for the rat

GLP1R in CHO cells and suspected to be caused by the βγ-subunits of Gs (Montrose-

Rafizadeh et al. 1999). In the course of this project and with the human GLP1R, we

were not able to recreate these findings, since the reporter gene assay for ERK1/2

phosphorylation did not report any activation. One explanation could be the

endogenous expression of GIPR (Atwood et al. 2011), another incretin receptor that is

reported to impair ERK phosphorylation and calcium release of GLP1R and decrease

to 25% of the maximal response when challenged with GLP1 (Roed et al. 2015).

In conclusion, all three assays showed no significant difference between GLP1-T3 and

the uncoupled GLP1. From this, it can be concluded, that the hybridization with T3 did

not affect the functional properties of GLP1 at its target receptor.

5.1.2 GLP1-T3 initiates GLP1R internalization

Internalization of the receptor after the stimulation with the conjugate is the crucial step

in the “Trojan Horse”-like mechanism. Therefore, investigations were performed using

several methods.

1. The SNAP-tag staining was used to visualize the internalization. When stained

with a non-permeable dye (SNAP Surface 488) and challenged with GLP1 or

GLP1-T3, the receptor was clearly internalized for both ligands after 30 min of

stimulation. This seems to be a reasonable time frame and similar results have

been published already where the SNAP-tag technology was also used in a

similar fashion (Roed et al. 2014).

2. This was also confirmed by an internalization assay that utilized a split luciferase

(HiBiT), where the surface lost more than 80% of receptor expression in a

concentration-dependent manner for both GLP1 and GLP1-T3 with very similar

IC50 for the two ligands. The cell surface expression of GLP1R was measured

after 30 min, verifying the results from the SNAP-tag staining.

3. Recruitment of β-arrestin precedes the internalization process and

NanoBRET™ assay (in this case β-arrestin2) was used for these investigations.

Similar results for GLP1 have been published with an EC50 of around 10-8 M

(Hager et al. 2016). GLP1-T3 showed a slightly lower potency to recruit β-

arrestin2, but was still able to interact with the protein in a concentration-

dependent manner.

These results allow the conclusion that the conjugate is able to activate both signaling

cascades and the internalization process in a similar fashion to the unconjugated

peptide. The conjugate does not exhibit biased signaling on the receptor, a fact that

should exclude unexpected adverse effects in vivo on the cellular level.

5.1.3 Oxytocin-T3 exhibits biased signaling at both target receptors

Similar functional characterization was performed with the two target receptors for OT-

T3, OTR and V1AR. The results show a differential picture of activation by OT-T3. The

pathways tested for OTR are Gi, Gq/11/PLC-activation and ERK1/2 phosphorylation, all

pathways have been published for the receptor.

5.1.3.1 OTR signaling

5.1.3.1.1.1 Gi signaling at the OTR

Gi coupling has been known for a long time and confirmed in many publications, and

seems to be most important for uterine contractions (Busnelli et al. 2012; Strakova and

Soloff 1997). The conjugate was not able to activate Gi mediated AC inhibition upon

stimulation. This showed an attenuation of OT function at the OTR when hybridized

with T3.

5.1.3.1.1.2 Gq/11/PLC-activation at the OTR

PLC- activation, mostly described as the main pathway for OTR, has been reported to

be partly activated by Gq/11 and partly by the βγ-subunits of Gi (Phaneuf et al. 1993;

Molnár and Hertelendy 1990). Pertussis toxin (PTX), a Gi inhibitor was used to confirm

the results from the cAMP assay. PTX is an ADP-ribosylating toxin that adds a ribosyl

group to a cysteine residue at the α-subunit of Gi. This results in an uncoupling of the

G protein with its receptor and therefore the α-subunit does not dissociate from its βγ-

subunit (Locht, Coutte, and Mielcarek 2011). The main pathway of OTR, PLC-

activation was not attenuated by the hybridization with T3 in comparison to OT alone.

This was surprising, since signaling was PTX-sensitive for both peptides, which clearly

suggests that part of the activation is still due to activation of the βγ-subunits of Gi even

for OT-T3. It is published that the subunits of Gi do not necessarily dissociate upon

activation (Frank et al. 2005; Bünemann, Frank, and Lohse 2003). It could be

hypothesized that OT-T3 binding to OTR rearranged the receptor in a way such that it

still couples to Gi without activation of the α-subunit but βγ-subunits, which would be

fully inhibited by PTX-pretreatment.

5.1.3.1.1.3 MAPK pathway at the OTR

To confirm this, MAPK pathway was investigated. For the OTR, it has been published

to be partly activated by PLC activation and partly due to activation of Gi. However, it

is not entirely clear whether the Gi part of ERK phosphorylation is due to the βγ-

subunits or the activity of the α-subunit and could be a pleiotropic effect, since different

sensitivities of MAPK pathway for OTR have been published in different cell lines

(Zhong, Yang, and Sanborn 2003; Strakova et al. 1998). PTX-pretreatment lead to a

significant decrease of signals, indicating that Gi protein activation is partly responsible

for ERK phosphorylation in the used cell system. The signals were significantly

dampened when cells were challenged with OT-T3 in comparison to unconjugated OT.

The curve resulting from OT-T3 stimulation mimics the curve of PTX-treated cells,

leading to the conclusion that the activation of ERK phosphorylation by OT-T3 is

produced by the activation of Gq/11 alone, but not Gi.

Together with the results from the AlphaScreen™ and NFAT assay, these findings

suggest that OT-T3 is not able to activate the G

protein fully, but only its βγ-subunits.

In contrast to PTX, the interaction between receptor and G protein still takes place,

when stimulated with the conjugate, but the conformational change of the receptor

does not activate the α-subunit of G

and therefore, no inhibition of AC and a decrease

in ERK phosphorylation was measured. Figure 24 depicts an overview of the biased

signaling at the OTR for a better understanding.

In brief, OT-T3 exhibits biased signaling on the OTR with attenuated activation of the

pathway

with abolished AC inhibition and reduced MAPK activation, but persistent

PLC activation. These results hint to possible adverse effects on a cellular level.

5.1.3.2 V1AR signaling

For the second receptor, V1AR two pathways have been investigated: G

q/11

/PLC-

activation and ERK phosphorylation.

5.1.3.2.1.1 G

11/PLC-activation at the V1AR

The main pathway, the PLC-activation, has been published for V1AR for both

endogenous ligands, AVP and OT (Thibonnier et al. 1994). In comparison to OTR, the

maximal response of V1AR was lower, verifying the lower affinity of V1AR to OT

(Åkerlund et al. 1999). The activation of this pathway was identical for both peptides,

Figure 24: Depiction of the signaling pathways at the OTR and the biased signaling caused by

PTX and OT-T3.

A: OTR is known to couple to two different G proteins, Gi and Gq/11. Upon activation, the trimeric

G proteins dissociate in their α-subunit, which has a GTPase function, and their βγ-subunits.

The Gαi inhibits AC, whereas Gαq/11 activates PLC. Additionally, it is known to start MAPK

pathway, here depicted as ERK. For this particular GPCR, both, PLC activation and ERK

phosphorylation have been published to be caused by activation of both G proteins.

B: Pertussis toxin (PTX) inhibits the coupling of Gi to the receptor and therefore blocks the

dissociation of this G protein entirely. This leads to a decrease in PLC activity and ER

phosphorylation.

C: OT-T3 was able to activate the Gq/11 protein to a full extend, but lacked a certain part of the Gi

protein pathway, such as AC inhibition and ERK phosphorylation. Since PLC activity was

identical to OT stimulation alone and could be decreased by PTX treatment, the βγ-subunits are

assumed to be active, whereas Gαi activity was missing.

which leads to the conclusion that the conjugate activates the receptor in the same

manner as the unconjugated OT.

5.1.3.2.1.2 MAPK pathway at the V1AR

Activation of MAPK pathway at the V1AR has been published for both natural agonists,

AVP and, with a slightly lower affinity, OT (Tahara et al. 1999). For the performed assay

in the course of this project, signals were significantly increased with high

concentrations of OT-T3. This could result from the instability of the conjugate and the

resulting activation of non-canonical TH signaling. This has been explained in the

introduction (1.1.6.4) and demonstrated to be activated (results part, 0). The reporter

used for these assays is the same as for non-canonical TH signaling, but it misses the

co-expression of TRα. Still, high concentrations of free T3 might be enough to activate

the reporter by endogenous TRs and could be the reason for the significant increase

in ERK phosphorylation. This would explain that the effect is most prominent for the

highest concentration of conjugate.

In conclusion, both target receptors exhibit biased signaling for OT-T3. Although these

effects have to be kept in mind for potential adverse effects of the conjugate, the

internalization process is most important for the “Trojan Horse”-like approach.

Depending on the signaling effects of the receptors in the target cells, it could even be

beneficial to decrease signaling, but still activate the endocytosis. Therefore, the

internalization process of the two target receptors was investigated next.

5.1.4 Oxytocin-T3 initiates internalization of OTR, but not V1AR

Afterwards, the recruitment of β-arrestin2 was tested for both target receptors as an

indicator for internalization. β-arrestins have been shown to be important mediators of

internalization independently of G protein activation (Grundmann et al. 2018), and

therefore the recruitment needed to be tested separately.

5.1.4.1 β-arrestin2 recruitment at the OTR

The recruitment of β-arrestins by OTR have been shown to be important for

endocytosis of the receptor (Oakley et al. 2001). Our assays showed the identical

response to both conjugated and unconjugated peptide in a concentration-dependent

manner. The receptor seems to be internalized upon the stimulation with the conjugate,

which is an important step in the “Trojan Horse”-like mechanism.

5.1.4.2 β-arrestin2 recruitment at the V1AR

As with most Class A GPCRs, V1AR is also known to be internalized by β-arrestins

(Terrillon, Barberis, and Bouvier 2004). Surprisingly, the receptor was not able to

recruit β-arrestin2 when challenged with OT-T3, but did so after OT stimulation. This

seems to be a specific effect for OT-T3 at the V1AR, since the assay for OTR executed

in the same manner showed an identical recruitment for both tested agonists.

Therefore, it is unlikely that the instability of the conjugate and free T3 had any

influence on the assay, although the effect of stimulation with T3 alone has not been

controlled for. This means that in contrast to OTR, V1AR is not internalized upon

stimulation with OT-T3.

From these results, it can be concluded that OTR internalizes when stimulated with

OT-T3, whereas V1AR does not. Unfortunately, V1AR was the target receptor that was

most attractive for the treatment of MCT8-deficient patients due to the high expression

in motoneurons (Caldwell et al. 2008), a potential target for a treatment.

Combining these results with the signaling assays, it is clear that the OT-T3 conjugate

does not meet optimal requirements for the “Trojan Horse”-like approach as it was

determined for GLP1-T3. Final validation however was made by investigating the

canonical TH-dependent signaling in the absence and presence of both target

receptors.

5.2 TH-dependent signaling of the conjugates

Stimulation of cells with T3 leads to the regulation of gene expression within every

cell. This is mediated by transmembrane transporters that are able to transport TH

over the membrane. Within the cells, T3 binds to TRs that function as transcription

factors and regulate the expression of target genes by binding to TRE in the promotor

region (Lazar, Berrodin, and Harding 1991). This has been defined as the canonical

pathway of TH and can be used by a reporter gene assay that utilizes a TRE in

control of a luciferase gene. The expression of luciferase is proportional to the TH-

dependent signaling that is activated by free T3 within the cell (Hofmann,

Schomburg, and Kohrle 2009).

After the functional characterization of the peptide part of the conjugate has been

performed, TH-dependent signaling was investigated as a proof that after

internalization of the receptor-conjugate complex, T3 is liberated and can act in a

similar manner to T3 that would be transported via the transmembrane transporter.

First, preliminary experiments were performed to find a suitable TH transporter

blocker for the used cell system. This could then be used to exclude that free T3 is

being transported by TH transporters and is proof of a functioning “Trojan Horse”-like

mechanism. Every cells type is expressing its unique set of TH transporters. To find a

suitable blocker, all published TH transporter blockers were tested.

Afterwards, activation of TH-dependent signaling of both conjugates were tested in

absence and presence of their target receptors. The “Trojan Horse”-like mechanism

would be proven by an increase of TH signaling only in the presence, but not in the

absence of their respective receptors.

5.2.1 No tested blocker could inhibit endogenous TH transporters in HEK293

cells

We tested published TH transporter blockers in previously reported concentrations or

lower, but for longer incubation periods. Only DMI was able to decrease T3 signaling

by a significant amount, but when cell viability was tested, it was discovered that DMI

might not block the transport of T3 into the cell, but is in fact a toxic compound. Hence,

it probably decreased the TH signaling by lowering the amount of live cells when

incubated with high concentrations or for long periods. To find a concentration that

might be able to block TH transport in non-toxic concentrations, cells were incubated

with varying concentrations of DMI and both TH-dependent signaling and cell viability

were determined. Unfortunately, no concentration was found that would still be able to

block TH transport, but would not influence cell viability. It has to be concluded, that

the inhibition of TH transport is due to toxicity of the compound rather than a true

blockage of the endogenous transporter. Later, a study was published that confirmed

toxicity of DMI for concentrations higher than 500 µM, five times the highest amount

that was used for the experiments in this project (Dong and Wade 2017). It has to be

noted that the incubation time in the published study was 10 min, whereas the

incubation time for the experiments in this project was 1 h. For time-dependent

experiments, incubation periods up to 6 h hours were planned.

From those preliminary experiments it can only be concluded that DMI is potentially

toxic for living cells even in low concentrations when incubated for long periods.

5.2.2 GLP1-T3 activates TH-dependent signaling in the presence of GLP1R

After it was confirmed that GLP1-T3 activates GLP1R and its internalization, TH-

dependent signaling was investigated. It was hypothesized that signaling should only

occur in the presence of the respective receptor through the “Trojan Horse”-like

mechanism when cells are challenged with the conjugate. Two different pathways have

been investigated, the canonical pathway utilizing a TRE-controlled luciferase reporter

and the non-canonical pathway which uses a SRE-controlled luciferase reporter. Both

are co-expressed with the rTRα, a TH receptor that is necessary to activate the

reporter.

5.2.2.1 GLP1-T3 activates the canonical pathway

Indeed, GLP1-T3 was able to activate canonical TH-dependent signaling pathway in

the presence of the receptor. The signals were significantly increased in comparison

to the cells where the receptor was absent. This seems to be time- and concentration-

dependent. In comparison to the positive control, the stimulation with T3 alone, where

we could see an immediate increase to maximal signals, the values for GLP1-T3 were

more than 50% lower and the effect was slower with significant results after 60 min of

stimulation. This is a strong indicator that the “Trojan Horse”-like mechanism is at work

with GLP1R being internalized and recycled back to the membrane or degraded rather

than a transporter that permanently transports TH over the plasma membrane.

Regarding the concentration-ascending experiments, it has to be noted that signaling

also increased in mock controlled cells for the highest concentration compared to the

basal state. This is most likely due to endogenous expressed GPCRs of the used cell

system. It has been published that HEK293 cells express GIPR (Atwood et al. 2011),

another incretin receptor that is able to bind to GLP1 with very low affinity (Roed et al.

2015). The interactions with possible endogenously expressed GLP1R or other

unknown receptors cannot be excluded.

5.2.2.2 GLP1-T3 activates the non-canonical pathway

A similar experimental set-up was used to measure the non-canonical pathway. Here,

the reporter gene was controlled by a SRE. This response element is identical to the

one that was used for ERK phosphorylation assays, since the non-canonical pathway

is known to activate PI3K as well as MAPK pathways (Makino et al. 2008;

Kalyanaraman et al. 2014). The difference to the experiments performed for the

activation of MAPK at the GLP1R is the co-expression of transfected rTRα1, which

mediates the phosphorylation of ERK. Results from the functional characterization at

the receptor show that the abundance of endogenously expressed TRs is not enough

to activate the reporter (chapter 4.1.2). GLP1-T3 was able to activate this pathway in

a time-dependent manner in the presence of GLP1R that was significantly higher than

in mock transfected cells. The curve progression is very similar to the canonical

pathway with a significant increase after 60 min that was slowly increasing, but did not

reach signals similar to T3 challenge. Again, this indicates that the increase is due to

the “Trojan Horse”-like mechanism rather than instability of the conjugate.

Unfortunately, since we were not be able to find a suitable TH transporter blocker, we

could not confirm our findings with the inhibition of endogenously expressed

transporters. Nevertheless, the significant increase of signaling in comparison to mock

controlled cells is evidence that GLP1-T3 is entering the cells via the “Trojan Horse”-

like mechanism. Since we have not been able to find a suitable blocker for these

experiments that could block endogenous TH transporters in the used cell system,

further experiments have not been performed.

In conclusion, GLP1-T3 was able to activate both canonical and non-canonical TH-

dependent signaling, probably by internalization of the receptor-conjugate complex

and subsequent liberation of T3 within the cell. The fact that challenge with GLP1-T3

only increased signals in cells that were expressing GLP1R is a strong indicator that

this conjugate is working in the intended way and would be suitable for in vivo studies.

The performed assays for this conjugate helped building a toolbox that can be used for

future peptide candidates.

5.2.3 Oxytocin-T3 does not display increased TH-dependent signaling in the

presence of both target receptors

The second conjugate investigated, OT-T3 already displayed biased signaling on both

target receptors with one of them potentially failing internalization after stimulation.

Both of the receptors were tested in a time-dependent experiment for TH-dependent

canonical signaling. There was no significant difference between GPCR expressing

and mock expressing cells. Additionally, the values were in the same range as the T3-

stimulated positive controls indicating that the conjugate is not stable in cell culture

conditions and T3 was released prematurely outside of the cells before OT-T3 even

binds to the receptor. Later, our cooperation partners, who synthesized the peptide-T3

hybrid, confirmed the instability of the compound when incubated with serum. The

biased signaling that the conjugate still exhibits is most likely due to the residual linker

on the peptide. This confirms that even slight modifications on small peptides can

cause a biased response on the receptor.

These findings lead to the conclusion that OT-T3 is not stable enough to be used as a

“Trojan Horse”. Our cooperation partners are currently working on designing a new

conjugate that is more stable.

5.3 Summary of all findings and conclusions

5.3.1 Summary for GLP1-T3

For the first conjugate, GLP1-T3, that was used to establish a toolbox for investigations

of peptide-hormone conjugates in general and functioned as a proof-of-principle

compound, we found that it was able to activate its target receptor GLP1R in the

identical manner as the unconjugated GLP1. In addition, it can initiate the

internalization process of GLP1R, which is crucial for the “Trojan Horse”-like

mechanism. The hybridization with T3 seems to have no attenuating effect on receptor

binding and the desensitization mechanism. GLP1R is a class B GPCR that binds

ligands with its long N-terminal domain rather than a binding pocket in the

transmembrane domain (Hoare 2005). This might be one reason the receptor

activation is less sensitive to ligand modifications. Additionally, its size of around 30

amino acids (Baggio and Drucker 2007) makes it a good candidate for hybridization

with several target side chains that are probably not involved in receptor binding. In

this project, it was used to establish the toolbox for validation of future conjugates and

originally was intended as a treatment of metabolic syndromes. Therefore, a similar

activation and internalization of GLP1R by GLP1-T3 is beneficial for the treatment,

since both GLP1 and T3 have metabolic effects in their target tissues. In the case of a

conjugate that is intended as a treatment for AHDS, possible side effects by the

activation of receptor pathways have to be considered, though the internalization of

the receptor is a necessity for the “Trojan Horse”-like mechanism to work.

This mechanism was proven by canonical and non-canonical TH-signaling that was

significantly increased in the presence, but not in the absence of GLP1R. This confirms

that liberation of T3 after internalization is working rather than a premature release

outside of the cell. Since we could not find a suitable TH transport blocker, the results

could not be confirmed by inhibition of endogenous TH transporter. All results for

GLP1-T3 are depicted in a graphical abstract (Figure 25).

5.3.2 Summary for OT-T3

The second conjugate, OT-T3 that was thought to be a candidate to treat MCT8-

deficieny was investigated with the established toolbox. In contrast to GLP1, oxytocin

is a small nonapeptide that binds to class A GPCRs that have small binding pockets

(Barberis, Mouillac, and Durroux 1998).

Any kind of addition to or modification of the peptide can lead to attenuated binding at

the receptor and therefore altered signaling. This has been already shown for

carbetocin (de-amino-1-monocarba-(2-O-methyltyrosine)-oxytocin), an oxytocin

analog that has been deaminated at the N-terminus and the disulphide bridge between

cysteine 1 and 6 has been replaced with CH

-S. This peptide is no longer able to

activate G

coupling and β-arrestin1/2 recruitment (Passoni et al. 2016).

This illustrates that, especially for small peptides and Class A GPCRs, nearly every

residue seems to be crucial for ligand binding and receptor conformational changes

that result in altered signaling. As stated above, depending on the cell response,

altered receptor signaling could be beneficial to minimize adverse effects. On the other

hand, receptor desensitization could also have adverse effects on the cell, but it is

necessary for the intake of T3 over the “Trojan Horse”-like mechanism. Therefore, a

receptor should be chosen that is quickly recycled to the cell surface.

TH-dependent signaling assays revealed instability of the compound which lead to the

increase of signal independently of receptor expression, concluding that no “Trojan-

Horse”-like mechanism is at work. Therefore, no further experiments were conducted.

Figure 25: Graphical abstract for the first investigated conjugate GLP1-T3. The ligand was able

to activate the main pathway for the respective GPCR, the GLP1R in the identical manner to the

unconjugated GLP1. It has also been shown that the conjugate initiated the recruitment of β-

arrstin2 to the receptor and decreased the amount of surface-localized GLP1R, suggesting

receptor internalization.

In presence of the GLP1R, the conjugate was able to activate both canonical and non-canonical

signaling pathways of TH, suggesting T3 entered the cell through the “Trojan Horse”-like

mechanism.

All results for OT-T3 are depicted in a graphical abstract (Figure 26). Interesting to

note however, is the fact that even though the conjugate is most likely unstable and T3

is released prematurely, the signaling properties on both tested target receptors are

different from unconjugated OT. It might be concluded that the linker is still attached to

OT and seems to impede interaction with the receptor binding pocket.

5.4 Suitability of small peptides as backbone for “Trojan-Horse”-conjugates

In order to discuss the suitability of the two different peptides used in the course of this

project, the different classes of their respective GPCRs and their binding pockets have

to be reviewed shortly. The main difference between class A and class B is the long

extracellular N-terminal domain of class B GPCRs. For most receptors, the peptide

agonists bind to this long extracellular part and the N-terminal region then activates the

receptor by interaction with the helical transmembrane domains (Lee, Booe, and

Pioszak 2015). Most of the known class A GPCR on the other hand have shorter

extracellular domains and agonists bind directly into their binding pocket. Therefore,

ligands that activate the receptor at their orthosteric binding site are relatively small.

As the results from this project show, the choice of peptide as the backbone of “Trojan

Horse”-conjugates can have a profound effect on signaling behavior exhibited by their

Figure 26: Graphical abstract for the second investigated conjugate OT-T3. The ligand was able to

activate the main pathway for the two target GPCRs, OTR and V1AR. For other pathways, biased

signaling was identified. OT-T3 was not able to activate Gi-mediated AC inhibition and showed a

significantly increased ERK phosphorylation in comparison to the unconjugated OT. It was also

able to initiate recruitment of β-arrstin2 to OTR, but not to V1AR, suggesting no internalization at

the V1AR.

The conjugate showed stability issues, suggesting that high TH-dependent signaling measured in

this project were due to free T3 that has already been cleaved outside of the cell.

respective receptor. The main difference between GLP1 and OT is their size (GLP1:

30 amino acids, ~ 3500 Da; OT: 8 amino acids, ~1000 Da) and the GPCR class of their

target receptors. As a class B receptor, GLP1R binds its ligands in two steps. First, the

N-terminal long extracellular part of GLP1R interacts with the C-terminal part of GLP1,

then the N-terminus of the ligands is associated with the core of the receptor in the

transmembrane domain (Underwood et al. 2010; Runge et al. 2008). Combining this

knowledge with the overall size of the peptide, GLP1 displays plenty of options to

attach a small hormone, such as estrogen (as was done in (Finan et al. 2012)) or T3

without hindrance of receptor binding.

OTR and V1AR, on the other hand, have small binding pockets within their

transmembrane domain. Due to their overall homology to the many existing class A

receptors, the residues that are important for specific interaction with the ligands are

oriented towards the binding pocket. These have been shown to be conserved over

many species in TMH6, whereas residues in TMH7 are more variable to determine the

selectivity of the receptor towards its ligands (Koehbach et al. 2013). For OT it has

been found that residues in the extracellular N-terminal part of OTR close to the TMH1,

as well as ECL1 and 2, are essential for high-affinity binding of agonists, but not for

antagonists (Gimpl et al. 2008). Intense research on these receptors show that plenty

of contact points between peptide and receptor are necessary for binding specificity

and signaling activation. Associated with the small size of the ligand, it is easy to

imagine that almost every residue of this peptide has a role in receptor binding.

Furthermore, the fact that the closely related peptide AVP, which differs in only two

amino acid positions (Ganten et al. 1986), has a vastly different affinity to V1AR than

OT (Åkerlund et al. 1999), makes clear that the family of oxytocin and vasopressin

receptors and peptides are a fine-tuned network. Therefore, the attachment or

modification on these peptides poses a difficult task.

Before synthesis, a structural homology model was used to identify possible

modification sites of OT (Figure 27). In this model, the C-terminal part pointed to the

extracellular site and was therefore chosen as the hybridization site for T3. In detail, it

was position 8 (Leu) that was linked to the TH. Since the synthesis of the conjugate

was performed by our cooperation partners (Indiana University, Bloomington), the

actual linker chemistry is not known.

Recently, Busnelli et. al. published a

study with bivalent OTR agonists that are

targeted towards OTR dimers. Here, OT

was deaminated at position 1 (Cys) and

alkane linkers of various length were

added at the leucine residue of position

8. They then characterized these

agonists and found that a linker containing more than eight alkane groups was able to

maintain receptor function (Busnelli et al. 2016). This confirms that the hybridization

site for OT-T3 was a good choice, but the chemical composition needs to be improved.

This is especially complicated, since the linker should also be able to release T3 within

the cell, but not before entering it.

In conclusion, hybridization of larger peptides such as glucagon or GLP1 with small

hormones is possible without losing receptor functions. With smaller peptides,

modification should be made with caution. The here investigated OT-T3 is not advised

to be tested in vivo.

5.5 Suitability of OT-T3 for treatment of MCT8-deficiency

The aim of the study was to find a perfect ligand-receptor pair that would be suitable

for treatment of MCT8-defciciency. For that, an already available peptide-T3 conjugate,

GLP1-T3, was used to establish a toolbox of in vitro assays. This conjugate was not

intended for MCT8-deficiency treatment, but as a proof-of-principle compound. The

candidate that was thought to be suitable for the intended purpose was OT-T3.

Irrespective of the fact that the investigated OT-T3 conjugate is not stable enough in

cultured conditions and the size of OT makes the hybridization of T3 difficult without

losing receptor functions, its suitability for the intended purpose as a treatment for

AHDS has to be discussed. Initially, it was thought to be a perfect candidate due to its

known positively associated functions in social behavior and bonding. Additionally, its

target receptors are highly expressed in the brain overall, especially in the areas that

would benefit from increased T3 levels in MCT8-deficient patients, such as areas that

are associated with locomotor defects.

Figure 27: Structural homology model of OTR

(backbone ribbon presentation) based on a

recently described protocol for modeling of the

oxytocin/vasopressin receptor group (Busnelli et

al. 2013). OT (blue sticks) is docked at the

endogenous orthosteric binding site in the active

conformation. The inset shows the surface

representation receptor/ligand complex and

visualizes that specifically the C-terminal ligand

part should be used to modify the ligand,

because it points towards the extracellular site.

T3 (magenta) is added to the C-terminal part o

OT as the suggested hybridization site.

This image by Gunnar Kleinau was used as an

indicator as to where T3 should be conjugated to

OT.

However, whether or not OT even crosses the BBB is still up for debate (Leng and

Ludwig 2016). The BBB is comprised of epithelia cells, glia cells, and extracellular

matrix that separates the peripheral circulation from the CNS. Its permeability is

regulated by tight junctions between the epithelium (Lawther, Kumar, and Krovvidi

2011). Several studies claim an intranasal (IN) administration of OT as a way to target

the brain directly and bypass the BBB. This is supposed to be facilitated by two routes:

the olfactory pathway over olfactory nerve in the nasal mucosa, which directly extend

to the cranial cavity (Bahadur and Pathak 2012) and the trigeminal nerve, a cranial

nerve which projects from the nasal passages to the brain stem. OT is thought to be

internalized into these neurons, transported along the axon followed by exocytosis.

Although this method of administration supposedly reaches the brain, most of the IN

administered OT will leak into the circulation through the nasal mucosa (Lochhead and

Thorne 2012). Another, more reasonable, but debated explanation is the uptake of IN

administered drugs into the perivascular space, a lymphatic system of the brain,

dubbed the glymphatic system. Whether this system actually exists or is just an artifact

in MRI scans is highly controversial (Engelhardt, Vajkoczy, and Weller 2017;

Naganawa et al. 2017). From here it can be absorbed by the CSF and would be

distributed throughout the whole brain through the perivascular space to the neurons

(Lochhead et al. 2015). Since most studies are performed in rodents, they should be

regarded with caution. Diffusion and distribution of drugs throughout the brain regions

might be faster than in humans due to the considerably smaller size of rodent brains.

After IN administration, OT concentrations are measured in the cerebrospinal-fluid

(CSF), a fluid that surrounds the brain and the spinal cord (Davson 1967), in saliva

and/or blood. Since a lot of the studies are performed in rodents and in most of them,

overall OT concentrations have been determined, several of these investigations are

subject to strong criticism. Many suggest that administration of OT to the periphery (for

example intravenous=IV) would lead to a release of central OT (Ermisch et al. 1985)

and critics argue that this would be also the case for IN administered OT that distributes

in the periphery in high concentrations (Leng and Ludwig 2016). A very recent study

suggested that this is not the case. Lee and colleagues administered d5-deuterated

OT IV and IN to non-human primates (rhesus macaques). Afterwards they measured

this labeled peptide in peripheral blood and CSF samples with a novel highly sensitive

and specific quantitative mass spectrometric assay, but could not find increased

endogenous, non-labeled OT concentrations. Nevertheless, they were also not able to

find an advantage of IN versus IV administrations, since there were no increased

labeled OT concentrations in the CSF compared to IV administration (Lee et al. 2017).

Critics also point out that in most studies measured OT concentrations in the CSF are

a small fraction of the originally administered OT and therefore are due to spillage of

high concentrations in the circulation and accumulation in the subarachnoid space.

This could have adverse effect on organs in the periphery, such as the gastro-intestinal

tract and reproductive system (Leng and Ludwig 2016).

Another noticeable fact is the inconsistent methodology used in the published articles.

Numerous studies use vastly different methods of measurement of plasma OT. Some

studies used ELISA (enzyme-linked immunosorbent assays) (Feldman, Gordon, and

Zagoory-Sharon 2011), radioimmunoassays (Szeto et al. 2011; Feldman, Gordon, and

Zagoory-Sharon 2011) or LC/MS (Liquid chromatography/mass spectrometry) (Zhang

et al. 2011) of unextracted or extracted plasma (Christensen et al. 2014), making it

difficult to compare values from different reports.

In clinical psychology OT is studied for its behavior effects. Patients with schizophrenia

or on the autism spectrum disorders are mostly subjected to IN administration of high

doses of OT and many studies report reduction of symptoms. However, many of these

studies have only a small number of test subjects, short treatment periods and

investigate social cognition using tests that are standardized, but can be hard to

interpret (Feifel et al. 2010; Pedersen et al. 2011).

In summary, many questions surrounding OT administration are still open. In case

there will be a new and stable OT-T3 conjugate, in vivo studies in rodents would be

necessary in addition to the here presented in vitro studies. Presuming the conjugate

was able to surpass the BBB after IN administration, T3 would be released in the

olfactory bulb, the trigeminal nerve, and/or the glymphatic system and could be

transported into other brain regions.

The expression of MCT8 within the human brain is still mostly unknown, since the

majority of studies have been performed in mice. As mentioned in the introduction,

mct8-KO does not lead to the severe phenotype in mice. Another TH transporter, most

likely oatp1c1, is expressed on the murine BBB which can compensate for the

decreased influx of T3 into the brain (Mayerl et al. 2014). Fluorescence-based

immunostaining of the human brain performed in our institute suggest that MCT8 is

mainly expressed in both apical and basolateral membrane of epithelia cells of the

human BBB, but not in neurons or the supporting glia cells (unpublished data, Nina-

Maria Wilpert). This contrasts with previously published findings in human neurons of

several brain areas using chromogenic immunohistochemistry staining (Chan et al.

2014; Alkemade et al. 2011; Wirth, Roth, Blechschmidt, Hölter, Becker, Racz, Zimmer,

Klopstock, Gailus-Durner, Fuchs, et al. 2009). Other studies support our findings

(Vatine et al. 2017; Roberts et al. 2008), leading to believe that once the BBB was

circumvented by IN administration, TH transport into neurons with OT-T3 would be at

least possible. This is supported by the fact that OTR is found in an early olfactory

neuronal cell line (Gravati et al. 2010) and in the olfactory bulb (Stoop 2012) as well as

trigeminal neurons (Tzabazis et al. 2015) of rat brains.

5.6 Prospects of AHDS treatment

Due to the controversy around OT administration, however, other ways for the

treatment of MCT8-deficiency to bypass the BBB should be considered. IN

administration seems to be very dependent on the properties of the compound used,

including but not limited to lipophilicity, acid dissociation constant, and charge density

(Chou and Donovan 1998; Kao et al. 2000). A recent publication shows the IN

administration of labeled full-length immunoglobulin G, a rather large peptide, into the

nervous system of rats, highlighting the brain areas that were reached by the

compound, including olfactory bulb, brainstem and part of the frontal cortex (Kumar et

al. 2018). Other substances that have been postulated to be suitable for IN

administration are dopamine (Dahlin et al. 2000), insulin (Kern et al. 1999) and insulin-

like growth factor (Thorne et al. 2004). IN administration of TH has not been tested yet

and could be of interest, but needs intensive investigations. Many compounds have

been capsulated with nanoparticles such as methoxy poly(ethylene glycol)-poly(lactic

acid) (MPEG-PLA) that improves the passage of the nasal mucosae due to its

hydrophobic core, the PLA-part, and its hydrophilic shell, the PEG-part (Zhang et al.

2006). This could be a promising alternative way to deliver TH into the brain.

The last question remaining is whether major improvement of AHDS by any treatment

is even possible, since it is a genetic disorder with developmental impairments. Here,

the only similar disease that comes to mind is CH, which can be caused genetically,

but involves global hypothyroidism. Major improvements in growth and development

of CH patients have been observed by treatment with T4 from early childhood (Krude,

Kühnen, and Biebermann 2015). As described in the introduction, one of the main

differences of AHDS in comparison to CH is the missing supply of maternal THs even

during the first trimester of gestation, a crucial phase of neuronal development (de

Escobar et al. 2008), due to the inactive transporter on the BBB. Therefore, treatment

within pregnancy would be most desirable, although it is rarely tested for an uncommon

genetic disorder during pregnancy and most cases are diagnosed postnatally. This is

only the case in families with known cases of AHDS. Nevertheless, treatment with OT-

T3 during pregnancy would not be advised anyway, due to the labor-inducing

properties of OT (Dawood et al. 1978) and the high concentrations of oxytocinase, the

enzyme that degrades oxytocin during pregnancy to prevent premature labor (Fekete

1932).

Treatment would most likely start after birth and diagnosis, but how promising the

therapy would be is unknown. TH derivates have been used to treat AHDS patients,

but were not able to improve the neurological phenotype (Verge et al. 2012). In a

mouse study, pregnant mice carrying mct8-deficient and wild type embryos were

treated with L-T4 or DITPA. After birth, TH-regulated gene expression was assessed

by qPCR to show that the derivate was able to cross the placenta in a similar fashion

to L-T4 (Ferrara et al. 2014). This study has to be interpreted with caution, since the

mct8-KO mouse model has its limitations as stated above (introduction, chapter 1.2.3).

Especially the missing neurological problems due to an additional TH transporter on

the BBB of rodents, make this model not ideal for treatment testing. It has also been

shown that DITPA is not able to cross a human BBB-model derived from iPSCs (Vatine

et al. 2017) and most likely does not so in vivo, which explains the results of the human

trails.

The main histological characteristics resulting from this are structural immaturity and

deficient myelination of CNS-neurons (López-Espíndola et al. 2014), a shared

phenotype with untreated CH (Berbel et al. 1994; Koromilas et al. 2010). Myelination

has been shown to continue postnatally to the age of four (Bernal 2007). Moreover, TH

treatment of an adult mouse model for multiple sclerosis, a demyelinating disease,

resulted in regeneration of myelin and differentiation of oligodendrocyte progenitor

cells (Harsan et al. 2008). Additionally, case reports of CH show improvements of

myelination through magnetic resonance spectroscopy (Jagannathan et al. 1998),

leading to the belief that postnatal treatment in the first years of life might also improve

the conditions of MCT8-deficient patients. Nevertheless, due to the aforementioned

differences between these conditions, it is still unclear at what point in time and to what

extent some manifestations are reversible, if at all. Therefore, more research in this

field is clearly necessary.

6 List of references

Abe, Satoko, Toyomasa Katagiri, Akihiko Saito-Hisaminato, Shin-ichi Usami, Yasuhiro

Inoue, Tatsuhiko Tsunoda, and Yusuke Nakamura. 2003. 'Identification of

CRYM as a candidate responsible for nonsyndromic deafness, through cDNA

microarray analysis of human cochlear and vestibular tissues', The American

Journal of Human Genetics, 72: 73-82.

Abramowicz, M. J., H. M. Targovnik, V. Varela, P. Cochaux, L. Krawiec, M. A. Pisarev,

F. V. Propato, G. Juvenal, H. A. Chester, and G. Vassart. 1992. 'Identification

of a mutation in the coding sequence of the human thyroid peroxidase gene

causing congenital goiter', The Journal of clinical investigation, 90: 1200-04.

Åkerlund, Mats, Thomas Bossmar, Rémi Brouard, Anna Kostrzewska, Tadeusz

Laudanski, Adam Lemancewicz, Claudine Secradeil‐Le Gal, and Margareta

Steinwall. 1999. 'Receptor binding of oxytocin and vasopressin antagonists and

inhibitory effects on isolated myometrium from preterm and term pregnant

women', BJOG: An International Journal of Obstetrics & Gynaecology, 106:

1047-53.

Alkemade, Anneke, Edith CH Friesema, Andries Kalsbeek, Dick F Swaab, Theo J

Visser, and Eric Fliers. 2011. 'Expression of thyroid hormone transporters in the

human hypothalamus', The Journal of Clinical Endocrinology & Metabolism, 96:

E967-E71.

Allan, William, CN Herndon, and Florence C Dudley. 1944. 'Some examples of the

inheritance of mental deficiency: apparently sex-linked idiocy and

microcephaly', Am J Ment Defic, 48: 325-34.

Araki, O, H Ying, XG Zhu, MC Willingham, and SY Cheng. 2009. 'Distinct dysregulation

of lipid metabolism by unliganded thyroid hormone receptor isoforms', Molecular

endocrinology, 23: 308-15.

Arber, W., and S. Linn. 1969. 'DNA modification and restriction', Annu Rev Biochem,

38: 467-500.

Arjona, Francisco J, Erik de Vrieze, Theo J Visser, Gert Flik, and Peter HM Klaren.

2011. 'Identification and functional characterization of zebrafish solute carrier

Slc16a2 (Mct8) as a thyroid hormone membrane transporter', Endocrinology,

152: 5065-73.

Atwood, Brady K, Jacqueline Lopez, James Wager-Miller, Ken Mackie, and Alex

Straiker. 2011. 'Expression of G protein-coupled receptors and related proteins

in HEK293, AtT20, BV2, and N18 cell lines as revealed by microarray analysis',

BMC genomics, 12: 14.

Baggio, Laurie L, and Daniel J Drucker. 2007. 'Biology of incretins: GLP-1 and GIP',

Gastroenterology, 132: 2131-57.

Bahadur, Shiv, and Kamla Pathak. 2012. 'Physicochemical and physiological

considerations for efficient nose-to-brain targeting', Expert opinion on drug

delivery, 9: 19-31.

Bales, K. L., P. M. Plotsky, L. J. Young, M. M. Lim, N. Grotte, E. Ferrer, and C. S.

Carter. 2007. 'Neonatal oxytocin manipulations have long-lasting, sexually

dimorphic effects on vasopressin receptors', Neuroscience, 144: 38-45.

Barberis, C, B Mouillac, and T Durroux. 1998. 'Structural bases of vasopressin/oxytocin

receptor function', Journal of Endocrinology, 156: 223-29.

Bárez-López, Soledad, Maria Jesus Obregon, Raquel Martínez-de-Mena, Juan Bernal,

Ana Guadaño-Ferraz, and Beatriz Morte. 2016. 'Effect of triiodothyroacetic acid

treatment in Mct8 deficiency: a word of caution', Thyroid, 26: 618-26.

Bassett, JH Duncan, and Graham R Williams. 2003. 'The molecular actions of thyroid

hormone in bone', Trends in Endocrinology & Metabolism, 14: 356-64.

Berbel, Pere, Ana Guadan˜o-Ferraz, Antonia Angulo, and Jose´Ramo´n Cerezo. 1994.

'Role of thyroid hormones in the maturation of interhemispheric connections in

rats', Behavioural Brain Research, 64: 9-14.

Bergh, Joel J, Hung-Yun Lin, Lawrence Lansing, Seema N Mohamed, Faith B Davis,

Shaker Mousa, and Paul J Davis. 2005. 'Integrin αVβ3 contains a cell surface

receptor site for thyroid hormone that is linked to activation of mitogen-activated

protein kinase and induction of angiogenesis', Endocrinology, 146: 2864-71.

Bernal, Juan. 2007. 'Thyroid hormone receptors in brain development and function',

Nature Clinical Practice Endocrinology &Amp; Metabolism, 3: 249.

Bernal, Juan, Ana Guadaño-Ferraz, and Beatriz Morte. 2015. 'Thyroid hormone

transporters [mdash] functions and clinical implications', Nature Reviews

Endocrinology, 11: 406-17.

Bernier-Valentin, F, Z Kostrouch, R Rabilloud, Y Munari-Silem, and B Rousset. 1990.

'Coated vesicles from thyroid cells carry iodinated thyroglobulin molecules. First

indication for an internalization of the thyroid prohormone via a mechanism of

receptor-mediated endocytosis', Journal of Biological Chemistry, 265: 17373-

80.

Bialer, Martin G, Laura Lawrence, Roger E Stevenson, Gittel Silverberg, Marjorie K

Williams, J Fernando Arena, Herbert A Lubs, and Charles E Schwartz. 1992.

'Allan‐Herndon‐Dudley syndrome: clinical and linkage studies on a second

family', American journal of medical genetics, 43: 491-97.

Biebermann, Heike, Petra Ambrugger, Patrick Tarnow, Arpad von Moers, Ulrich

Schweizer, and Annette Grueters. 2005. 'Extended clinical phenotype,

endocrine investigations and functional studies of a loss-of-function mutation

A150V in the thyroid hormone specific transporter MCT8', European Journal of

Endocrinology, 153: 359-66.

Bielsky, Isadora F., Shuang-Bao Hu, Kathleen L. Szegda, Heiner Westphal, and Larry

J. Young. 2003. 'Profound Impairment in Social Recognition and Reduction in

Anxiety-Like Behavior in Vasopressin V1a Receptor Knockout Mice',

Neuropsychopharmacology, 29: 483.

Bochukova , Elena, Nadia Schoenmakers , Maura Agostini , Erik Schoenmakers ,

Odelia Rajanayagam , Julia M. Keogh , Elana Henning , Jana Reinemund ,

Evelien Gevers , Margarita Sarri , Kate Downes , Amaka Offiah , Assunta

Albanese , David Halsall , John W.R. Schwabe , Murray Bain , Keith Lindley ,

Francesco Muntoni , Faraneh Vargha-Khadem , Mehul Dattani , I. Sadaf

Farooqi , Mark Gurnell , and Krishna Chatterjee 2012. 'A Mutation in the Thyroid

Hormone Receptor Alpha Gene', New England Journal of Medicine, 366: 243-

49.

Boelaert, K, and JA Franklyn. 2005. 'Thyroid hormone in health and disease', Journal

of Endocrinology, 187: 1-15.

Boss, Valerie, Deepa J Talpade, and Thomas J Murphy. 1996. 'Induction of NFAT-

mediated Transcription by G-coupled Receptors in Lymphoid and Non-lymphoid

Cells', Journal of Biological Chemistry, 271: 10429-32.

Braun, Doreen, Iva Lelios, Gerd Krause, and Ulrich Schweizer. 2013. 'Histidines in

potential substrate recognition sites affect thyroid hormone transport by

monocarboxylate transporter 8 (MCT8)', Endocrinology, 154: 2553-61.

Brucker-Davis, Francoise, Monica C Skarulis, Marcy B Grace, Jacques Benichou,

Peter Hauser, Edythe Wiggs, and Bruce D Weintraub. 1995. 'Genetic and

clinical features of 42 kindreds with resistance to thyroid hormone: the National

Institutes of Health Prospective Study', Annals of Internal Medicine, 123: 572-

83.

Bünemann, Moritz, Monika Frank, and Martin J Lohse. 2003. 'Gi protein activation in

intact cells involves subunit rearrangement rather than dissociation',

Proceedings of the National Academy of Sciences, 100: 16077-82.

Busnelli, Marta, Elisabetta Bulgheroni, Maurice Manning, Gunnar Kleinau, and Bice

Chini. 2013. 'Selective and potent agonists and antagonists for investigating the

role of mouse oxytocin receptors', Journal of Pharmacology and Experimental

Therapeutics, 346: 318-27.

Busnelli, Marta, Gunnar Kleinau, Markus Muttenthaler, Stoytcho Stoev, Maurice

Manning, Lucka Bibic, Lesley A Howell, Peter J McCormick, Simona Di Lascio,

and Daniela Braida. 2016. 'Design and characterization of superpotent bivalent

ligands targeting oxytocin receptor dimers via a channel-like structure', Journal

of medicinal chemistry, 59: 7152-66.

Busnelli, Marta, Aude Saulière, Maurice Manning, Michel Bouvier, Celine Galés, and

Bice Chini. 2012. 'Functional selective oxytocin-derived agonists discriminate

between individual G protein family subtypes', Journal of Biological Chemistry,

287: 3617-29.

Caldwell, Heather K, Heon-Jin Lee, Abbe H Macbeth, and W Scott Young III. 2008.

'Vasopressin: behavioral roles of an “original” neuropeptide', Progress in

neurobiology, 84: 1-24.

Calebiro, Davide, Viacheslav O Nikolaev, Maria Cristina Gagliani, Tiziana de Filippis,

Christian Dees, Carlo Tacchetti, Luca Persani, and Martin J Lohse. 2009.

'Persistent cAMP-signals triggered by internalized G-protein–coupled

receptors', PLoS biology, 7: e1000172.

Chan, Shiao Y, Laura A Hancox, Azucena Martín-Santos, Laurence S Loubière, Merlin

NM Walter, Ana-Maria González, Phillip M Cox, Ann Logan, Christopher J

McCabe, and Jayne A Franklyn. 2014. 'MCT8 expression in human fetal

cerebral cortex is reduced in severe intrauterine growth restriction', Journal of

Endocrinology, 220: 85-95.

Chassande, O. 2003. 'Do unliganded thyroid hormone receptors have physiological

functions?', Journal of molecular endocrinology, 31: 9-20.

Chelikani, Prashen, Viktor Hornak, Markus Eilers, Phillip J Reeves, Steven O Smith,

Uttam L RajBhandary, and H Gobind Khorana. 2007. 'Role of group-conserved

residues in the helical core of β2-adrenergic receptor', Proceedings of the

National Academy of Sciences, 104: 7027-32.

Cheng, Sheue-Yann, Jack L Leonard, and Paul J Davis. 2010. 'Molecular aspects of

thyroid hormone actions', Endocrine reviews, 31: 139-70.

Chou, Kang-Jye, and Maureen D. Donovan. 1998. 'Distribution of antihistamines into

the CSF following intranasal deliviery', Biopharmaceutics & Drug Disposition,

18: 335-46.

Christensen, Halvor N. 1990. 'Role of amino acid transport and countertransport in

nutrition and metabolism', Physiological reviews, 70: 43-77.

Christensen, James C, Pavel A Shiyanov, Justin R Estepp, and John J Schlager. 2014.

'Lack of association between human plasma oxytocin and interpersonal trust in

a prisoner’s dilemma paradigm', PLoS one, 9: e116172.

Clifton-Bligh, Roderick J., John M. Wentworth, Peter Heinz, Michele S. Crisp, Rhys

John, John H. Lazarus, Marian Ludgate, and V. Krishna Chatterjee. 1998.

'Mutation of the gene encoding human TTF-2 associated with thyroid agenesis,

cleft palate and choanal atresia', Nature genetics, 19: 399.

Dahlin, Maria, Ulrika Bergman, Björn Jansson, Erik Björk, and Eva Brittebo. 2000.

'Transfer of dopamine in the olfactory pathway following nasal administration in

mice', Pharmaceutical research, 17: 737-42.

Davis, Paul J, Fernando Goglia, and Jack L Leonard. 2016. 'Nongenomic actions of

thyroid hormone', Nature Reviews Endocrinology, 12: 111.

Davis, Paul J, Ai Shih, Hung-Yun Lin, Leon J Martino, and Faith B Davis. 2000.

'Thyroxine promotes association of mitogen-activated protein kinase and

nuclear thyroid hormone receptor (TR) and causes serine phosphorylation of

TR', Journal of Biological Chemistry, 275: 38032-39.

Davson, Hugh. 1967. Physiology of the cerebrospinal fluid (Churchill: London).

Dawood, M Yusoff, KS Raghavan, C Pociask, and Fritz Fuchs. 1978. 'Oxytocin in

human pregnancy and parturition', Obstetrics & Gynecology, 51: 138-43.

De Deken, Xavier, Dantong Wang, Marie-Christine Many, Sabine Costagliola,

Frédérick Libert, Gilbert Vassart, Jacques E Dumont, and Françoise Miot. 2000.

'Cloning of two human thyroid cDNAs encoding new members of the NADPH

oxidase family', Journal of Biological Chemistry, 275: 23227-33.

de Escobar, Gabriella Morreale, Susana Ares, Pere Berbel, María Jesus Obregón, and

Francisco Escobar del Rey. 2008. "The changing role of maternal thyroid

hormone in fetal brain development." In Seminars in perinatology, 380-86.

Elsevier.

de Escobar, Gabriella Morreale, Marıa Jesús Obregón, and Francisco Escobar del

Rey. 2004. 'Maternal thyroid hormones early in pregnancy and fetal brain

development', Best practice & research Clinical endocrinology & metabolism,

18: 225-48.

Devriendt, Koenraad, Christine Vanhole, Gert Matthijs, and Francis de Zegher. 1998.

'Deletion of thyroid transcription factor-1 gene in an infant with neonatal thyroid

dysfunction and respiratory failure', New England Journal of Medicine, 338:

1317-18.

DeWire, Scott M, Seungkirl Ahn, Robert J Lefkowitz, and Sudha K Shenoy. 2007. 'β-

arrestins and cell signaling', Annu. Rev. Physiol., 69: 483-510.

Di Cosmo, Caterina, Xiao-Hui Liao, Alexandra M Dumitrescu, Nancy J Philp, Roy E

Weiss, and Samuel Refetoff. 2010. 'Mice deficient in MCT8 reveal a mechanism

regulating thyroid hormone secretion', The Journal of clinical investigation, 120:

3377-88.

Dohan, Orsolya, Antonio De la Vieja, Viktoriya Paroder, Claudia Riedel, Mona Artani,

Mia Reed, Christopher S Ginter, and Nancy Carrasco. 2003. 'The sodium/iodide

symporter (NIS): characterization, regulation, and medical significance',

Endocrine reviews, 24: 48-77.

Dohlman, Henrik G, Jeremy Thorner, Marc G Caron, and Robert J Lefkowitz. 1991.

'Model systems for the study of seven-transmembrane-segment receptors',

Annu Rev Biochem, 60: 653-88.

Dong, Hongyan, and Michael G Wade. 2017. 'Application of a nonradioactive assay

for high throughput screening for inhibition of thyroid hormone uptake via the

transmembrane transporter MCT8', Toxicology in Vitro, 40: 234-42.

Doyle, Máire E, and Josephine M Egan. 2007. 'Mechanisms of action of glucagon-like

peptide 1 in the pancreas', Pharmacology & therapeutics, 113: 546-93.

Dumitrescu, Alexandra M, Xiao-Hui Liao, Thomas B Best, Knut Brockmann, and

Samuel Refetoff. 2004. 'A novel syndrome combining thyroid and neurological

abnormalities is associated with mutations in a monocarboxylate transporter

gene', The American Journal of Human Genetics, 74: 168-75.

Dumitrescu, Alexandra M, Xiao-Hui Liao, Roy E Weiss, Kathleen Millen, and Samuel

Refetoff. 2006. 'Tissue-specific thyroid hormone deprivation and excess in

monocarboxylate transporter (mct) 8-deficient mice', Endocrinology, 147: 4036-

43.

Duprez, Laurence, Jasmine Parma, Jacqueline Van Sande, Anouk Allgeier, Jacques

Leclère, Claire Schvartz, Marie-Joëlle Delisle, Marc Decoulx, Jacques Orgiazzi,

Jacques Dumont, and Gilbert Vassart. 1994. 'Germline mutations in the

thyrotropin receptor gene cause non–autoimmune autosomal dominant

hyperthyroidism', Nature genetics, 7: 396.

Eglen, Richard M, Terry Reisine, Philippe Roby, Nathalie Rouleau, Chantal Illy, Roger

Bossé, and Martina Bielefeld. 2008. 'The use of AlphaScreen technology in

HTS: current status', Current chemical genomics, 1: 2.

Eichel, K., D. Jullié, and M. von Zastrow. 2016. 'β-Arrestin drives MAP kinase signalling

from clathrin-coated structures after GPCR dissociation', Nature Cell Biology,

18: 303.

Engelhardt, Britta, Peter Vajkoczy, and Roy O. Weller. 2017. 'The movers and shapers

in immune privilege of the CNS', Nature Immunology, 18: 123.

Ermisch, A, H-J Rühle, R Landgraf, and J Hess. 1985. 'Blood—brain barrier and

peptides', Journal of Cerebral Blood Flow & Metabolism, 5: 350-57.

Faure, Michel, Tatyana A Voyno-Yasenetskaya, and Henry R Bourne. 1994. 'cAMP

and beta gamma subunits of heterotrimeric G proteins stimulate the mitogen-

activated protein kinase pathway in COS-7 cells', Journal of Biological

Chemistry, 269: 7851-54.

Feifel, David, Kai Macdonald, Angel Nguyen, Patrice Cobb, Heather Warlan, Barbara

Galangue, Arpi Minassian, Olga Becker, Jason Cooper, William Perry,

Mischelle Lefebvre, James Gonzales, and Allison Hadley. 2010. 'Adjunctive

Intranasal Oxytocin Reduces Symptoms in Schizophrenia Patients', Biological

Psychiatry, 68: 678-80.

Feinstein, Timothy N, Naofumi Yui, Matthew J Webber, Vanessa L Wehbi, Hilary P

Stevenson, J Darwin King, Kenneth R Hallows, Dennis Brown, Richard Bouley,

and Jean-Pierre Vilardaga. 2013. 'Noncanonical control of vasopressin receptor

type 2 signaling by retromer and arrestin', Journal of Biological Chemistry, 288:

27849-60.

Fekete, K von. 1932. 'Gibt es während der Schwangerschaft ein aktives Hypophysen-

Hinterlappenhormon im Blute?', Endokrinologie, 10: 16.

Feldman, Ruth, Ilanit Gordon, and Orna Zagoory-Sharon. 2011. 'Maternal and paternal

plasma, salivary, and urinary oxytocin and parent–infant synchrony: considering

stress and affiliation components of human bonding', Developmental Science,

14: 752-61.

Feng, Weijun, Ralff CJ Ribeiro, Richard L Wagner, Hoa Nguyen, James W Apriletti,

Robert J Fletterick, John D Baxter, Peter J Kushner, and Brian L West. 1998.

'Hormone-dependent coactivator binding to a hydrophobic cleft on nuclear

receptors', Science, 280: 1747-49.

Ferrara, Alfonso Massimiliano, Xiao-Hui Liao, Pilar Gil-Ibáñez, Juan Bernal, Roy E.

Weiss, Alexandra M. Dumitrescu, and Samuel Refetoff. 2014. 'Placenta

Passage of the Thyroid Hormone Analog DITPA to Male Wild-Type and Mct8-

Deficient Mice', Endocrinology, 155: 4088-93.

Ferrara, Alfonso Massimiliano, Xiao-Hui Liao, Honggang Ye, Roy E Weiss, Alexandra

M Dumitrescu, and Samuel Refetoff. 2015. 'The thyroid hormone analog DITPA

Ameliorates metabolic parameters of male mice with Mct8 deficiency',

Endocrinology, 156: 3889-94.

Finan, Brian, Christoffer Clemmensen, Zhimeng Zhu, Kerstin Stemmer, Karine

Gauthier, Luisa Müller, Meri De Angelis, Kristin Moreth, Frauke Neff, and Diego

Perez-Tilve. 2016. 'Chemical hybridization of glucagon and thyroid hormone

optimizes therapeutic impact for metabolic disease', Cell, 167: 843-57. e14.

Finan, Brian, Bin Yang, Nickki Ottaway, Kerstin Stemmer, Timo D Müller, Chun-Xia Yi,

Kirk Habegger, Sonja C Schriever, Cristina García-Cáceres, and Dhiraj G

Kabra. 2012. 'Targeted estrogen delivery reverses the metabolic syndrome',

Nature medicine, 18: 1847-56.

Fischer, Jana, Gunnar Kleinau, Annette Grüters-Kieslich, Heiko Krude, and Heike

Biebermann. 2013. 'Specific pathogenic mutations modulate monocarboxylate

transporter 8 oligomerization.' in, Molecular Mechanisms in Thyroid Hormone

Action & Cancer (Endocrine Society).

Flamant, Frédéric, John D Baxter, Douglas Forrest, Samuel Refetoff, Herbert Samuels,

Tom S Scanlan, Bjorn Vennström, and Jacques Samarut. 2006. 'International

Union of Pharmacology. LIX. The pharmacology and classification of the

nuclear receptor superfamily: thyroid hormone receptors', Pharmacological

reviews, 58: 705-11.

Flamant, Frédéric, Sheue-Yann Cheng, Anthony N. Hollenberg, Lars C. Moeller,

Jacques Samarut, Fredric E. Wondisford, Paul M. Yen, and Samuel Refetoff.

2017. 'Thyroid Hormone Signaling Pathways: Time for a More Precise

Nomenclature', Endocrinology, 158: 2052-57.

Frank, Monika, Leonore Thümer, Martin J Lohse, and Moritz Bünemann. 2005. 'G

protein activation without subunit dissociation depends on a Gαi-specific region',

Journal of Biological Chemistry, 280: 24584-90.

Fredriksson, Robert, Malin C Lagerström, Lars-Gustav Lundin, and Helgi B Schiöth.

2003. 'The G-protein-coupled receptors in the human genome form five main

families. Phylogenetic analysis, paralogon groups, and fingerprints', Molecular

pharmacology, 63: 1256-72.

Friesema, Edith CH, Roel Docter, Ellis PCM Moerings, François Verrey, Eric P

Krenning, Georg Hennemann, and Theo J Visser. 2001. 'Thyroid hormone

transport by the heterodimeric human system L amino acid transporter',

Endocrinology, 142: 4339-48.

Friesema, Edith CH, Sumita Ganguly, Amal Abdalla, Jocelyn E Manning Fox, Andrew

P Halestrap, and Theo J Visser. 2003. 'Identification of monocarboxylate

transporter 8 as a specific thyroid hormone transporter', Journal of Biological

Chemistry, 278: 40128-35.

Friesema, Edith CH, Annette Grueters, Heike Biebermann, Heiko Krude, Arpad von

Moers, Maarten Reeser, Timothy G Barrett, Edna E Mancilla, Johan Svensson,

and Monique HA Kester. 2004. 'Association between mutations in a thyroid

hormone transporter and severe X-linked psychomotor retardation', The Lancet,

364: 1435-37.

Friesema, Edith CH, Jurgen Jansen, Jan-willem Jachtenberg, W Edward Visser,

Monique HA Kester, and Theo J Visser. 2008. 'Effective cellular uptake and

efflux of thyroid hormone by human monocarboxylate transporter 10', Molecular

endocrinology, 22: 1357-69.

Friesema, Edith CH, Jurgen Jansen, Carmelina Milici, and Theo J Visser. 2005.

'Thyroid hormone transporters', Vitamins & Hormones, 70: 137-67.

Friesema, Edith CH, George GJM Kuiper, Jurgen Jansen, Theo J Visser, and Monique

HA Kester. 2006. 'Thyroid hormone transport by the human monocarboxylate

transporter 8 and its rate-limiting role in intracellular metabolism', Molecular

endocrinology, 20: 2761-72.

Friesema, Edith CH, W Edward Visser, and Theo J Visser. 2010. 'Genetics and

phenomics of thyroid hormone transport by MCT8', Molecular and Cellular

Endocrinology, 322: 107-13.

Fujiwara, Hirokazu, Ke-ita Tatsumi, Kazunori Miki, Tokuzo Harada, Kiyoshi Miyai,

Shin-ichiro Takai, and Nobuyuki Amino. 1997. 'Congenital hypothyroidism

caused by a mutation in the Na+/l− symporter', Nature genetics, 16: 124-25.

Ganten, Detlev, Donald Pfaff, JPH Burbach, ML Forsling, and R Ivell. 1986.

Neurobiology of oxytocin (Springer Berlin et al.).

Gautier, Arnaud, Alexandre Juillerat, Christian Heinis, Ivan Reis Corrêa, Maik

Kindermann, Florent Beaufils, and Kai Johnsson. 2008. 'An engineered protein

tag for multiprotein labeling in living cells', Chemistry & biology, 15: 128-36.

Giacomini, Kathleen M, Shiew-Mei Huang, Donald J Tweedie, Leslie Z Benet, Kim LR

Brouwer, Xiaoyan Chu, Amber Dahlin, Raymond Evers, Volker Fischer, and

Kathleen M Hillgren. 2010. 'Membrane transporters in drug development',

Nature Reviews Drug Discovery, 9: 215.

Gika, Artemis D, Ata Siddiqui, Anthony J Hulse, Selvakumari Edward, Penny Fallon,

Meriel E Mcentagart, Wajanat Jan, Dragana Josifova, TALLY LERMAN‐SAGIE,

and James Drummond. 2010. 'White matter abnormalities and dystonic motor

disorder associated with mutations in the SLC16A2 gene', Developmental

Medicine & Child Neurology, 52: 475-82.

Gimpl, Gerald, and Falk Fahrenholz. 2001. 'The oxytocin receptor system: structure,

function, and regulation', Physiological reviews, 81: 629-83.

Gimpl, Gerald, Julian Reitz, Sabine Brauer, and Conny Trossen. 2008. 'Oxytocin

receptors: ligand binding, signalling and cholesterol dependence.' in Inga D.

Neumann and Rainer Landgraf (eds.), Progress in Brain Research (Elsevier).

Göthe, Sten, Zhendong Wang, Lily Ng, Jenny M Kindblom, Angel Campos Barros,

Claes Ohlsson, Björn Vennström, and Douglas Forrest. 1999. 'Mice devoid of

all known thyroid hormone receptors are viable but exhibit disorders of the

pituitary–thyroid axis, growth, and bone maturation', Genes & development, 13:

1329-41.

Gravati, Marta, Marta Busnelli, Elisabetta Bulgheroni, Alessandra Reversi, Paolo

Spaiardi, Marco Parenti, Mauro Toselli, and Bice Chini. 2010. 'Dual modulation

of inward rectifier potassium currents in olfactory neuronal cells by promiscuous

G protein coupling of the oxytocin receptor', Journal of Neurochemistry, 114:

1424-35.

Groeneweg, Stefan, Edith CH Friesema, Simone Kersseboom, Wim Klootwijk, W

Edward Visser, Robin P Peeters, and Theo J Visser. 2013. 'The role of Arg445

and Asp498 in the human thyroid hormone transporter MCT8', Endocrinology,

155: 618-26.

Groeneweg, Stefan, Robin P Peeters, Theo J Visser, and W Edward Visser. 2016.

'Diagnostic and Therapeutic Challenges in the Allan–Herndon–Dudley

Syndrome', US Endocrinology, 12: 90-93.

Grøntved, Lars, Joshua J Waterfall, Dong Wook Kim, Songjoon Baek, Myong-Hee

Sung, Li Zhao, Jeong Won Park, Ronni Nielsen, Robert L Walker, and Yuelin J

Zhu. 2015. 'Transcriptional activation by the thyroid hormone receptor through

ligand-dependent receptor recruitment and chromatin remodelling', Nature

communications, 6: 7048.

Grundmann, Manuel, Nicole Merten, Davide Malfacini, Asuka Inoue, Philip Preis,

Katharina Simon, Nelly Rüttiger, Nicole Ziegler, Tobias Benkel, Nina Katharina

Schmitt, Satoru Ishida, Ines Müller, Raphael Reher, Kouki Kawakami, Ayumi

Inoue, Ulrike Rick, Toni Kühl, Diana Imhof, Junken Aoki, Gabriele M. König,

Carsten Hoffmann, Jesus Gomeza, Jürgen Wess, and Evi Kostenis. 2018. 'Lack

of beta-arrestin signaling in the absence of active G proteins', Nature

communications, 9: 341.

Grüters, Annette, and Heiko Krude. 2012. 'Detection and treatment of congenital

hypothyroidism', Nature Reviews Endocrinology, 8: 104-13.

Gruters, Annette, Heiko Krude, and Heike Biebermann. 2004. 'Molecular genetic

defects in congenital hypothyroidism', European Journal of Endocrinology, 151:

U39-U44.

Guadano-Ferraz, Ana, Ruth Benavides-Piccione, César Venero, C Lancha, Björn

Vennström, Carmen Sandi, Javier DeFelipe, and Juan Bernal. 2003. 'Lack of

thyroid hormone receptor α1 is associated with selective alterations in behavior

and hippocampal circuits', Molecular psychiatry, 8: 30-38.

Gwee, Pai-Chung, Boon-Hui Tay, Sydney Brenner, and Byrappa Venkatesh. 2009.

'Characterization of the neurohypophysial hormone gene loci in elephant shark

and the Japanese lamprey: origin of the vertebrate neurohypophysial hormone

genes', BMC evolutionary biology, 9: 47.

Hagenbuch, Bruno, and Peter J Meier. 2004. 'Organic anion transporting polypeptides

of the OATP/SLC21 family: phylogenetic classification as OATP/SLCO

superfamily, new nomenclature and molecular/functional properties', Pflügers

Archiv, 447: 653-65.

Hager, Marlies V., Lisa M. Johnson, Denise Wootten, Patrick M. Sexton, and Samuel

H. Gellman. 2016. 'β-Arrestin-Biased Agonists of the GLP-1 Receptor from β-

Amino Acid Residue Incorporation into GLP-1 Analogues', Journal of the

American Chemical Society, 138: 14970-79.

Halestrap, Andrew P. 2012. 'The monocarboxylate transporter family—structure and

functional characterization', IUBMB life, 64: 1-9.

Hammes, Stephen R, and Paul J Davis. 2015. 'Overlapping nongenomic and genomic

actions of thyroid hormone and steroids', Best practice & research Clinical

endocrinology & metabolism, 29: 581-93.

Hanahan, Douglas, Joel Jessee, and Fredric R Bloom. 1991. '[4] Plasmid

transformation of Escherichia coli and other bacteria', Methods in enzymology,

204: 63-113.

Harsan, Laura-Adela, Jérôme Steibel, Anita Zaremba, Arnaud Agin, Rémy Sapin,

Patrick Poulet, Blandine Guignard, Nathalie Parizel, Daniel Grucker, Nelly

Boehm, Robert H. Miller, and M. Said Ghandour. 2008. 'Recovery from Chronic

Demyelination by Thyroid Hormone Therapy: Myelinogenesis Induction and

Assessment by Diffusion Tensor Magnetic Resonance Imaging', The Journal of

Neuroscience, 28: 14189-201.

Harvey, Clare B, and Graham R Williams. 2002. 'Mechanism of thyroid hormone

action', Thyroid, 12: 441-46.

Hashimoto, Koshi, Flavio H Curty, Patricia P Borges, Charlotte E Lee, E Dale Abel,

Joel K Elmquist, Ronald N Cohen, and Fredric E Wondisford. 2001. 'An

unliganded thyroid hormone receptor causes severe neurological dysfunction',

Proceedings of the National Academy of Sciences, 98: 3998-4003.

Hawes, Brian E, Tim van Biesen, Walter J Koch, Louis M Luttrell, and Robert J

Lefkowitz. 1995. 'Distinct Pathways of G-and G-mediated Mitogen-activated

Protein Kinase Activation', Journal of Biological Chemistry, 270: 17148-53.

Hayashizaki, Yoshihide, Yoshiki Hiraoka, Yuichi Endo, K Miyai, and K Matsubara.

1989. 'Thyroid‐stimulating hormone (TSH) deficiency caused by a single base

substitution in the CAGYC region of the beta‐subunit', The EMBO journal, 8:

2291-96.

Hennemann, Georg, Roelof Docter, Edith CH Friesema, Marion de Jong, Eric P

Krenning, and Theo J Visser. 2001. 'Plasma membrane transport of thyroid

hormones and its role in thyroid hormone metabolism and bioavailability',

Endocrine reviews, 22: 451-76.

Hiroi, Yukio, Hyung-Hwan Kim, Hao Ying, Fumihiko Furuya, Zhihong Huang,

Tommaso Simoncini, Kensuke Noma, Kojiro Ueki, Ngoc-Ha Nguyen, and

Thomas S Scanlan. 2006. 'Rapid nongenomic actions of thyroid hormone',

Proceedings of the National Academy of Sciences, 103: 14104-09.

Ho, Edmund S, Deborah C Lin, Dirk B Mendel, and Tomas Cihlar. 2000. 'Cytotoxicity

of antiviral nucleotides adefovir and cidofovir is induced by the expression of

human renal organic anion transporter 1', Journal of the American Society of

Nephrology, 11: 383-93.

Hoare, Sam RJ. 2005. 'Mechanisms of peptide and nonpeptide ligand binding to Class

B G-protein-coupled receptors', Drug discovery today, 10: 417-27.

Hofmann, P. J., L. Schomburg, and J. Kohrle. 2009. 'Interference of endocrine

disrupters with thyroid hormone receptor-dependent transactivation', Toxicol

Sci, 110: 125-37.

Holz, G. G. th, C. A. Leech, and J. F. Habener. 1995. 'Activation of a cAMP-regulated

Ca(2+)-signaling pathway in pancreatic beta-cells by the insulinotropic hormone

glucagon-like peptide-1', The Journal of biological chemistry, 270: 17749-57.

Hurowitz, Evan H, James M Melnyk, Yu-Jiun Chen, Hosein Kouros-Mehr, Melvin I

Simon, and Hiroaki Shizuya. 2000. 'Genomic characterization of the human

heterotrimeric G protein α, β, and γ subunit genes', DNA research, 7: 111-20.

Ianculescu, Alexandra G, Edith CH Friesema, Theo J Visser, Kathleen M Giacomini,

and Thomas S Scanlan. 2010. 'Transport of thyroid hormones is selectively

inhibited by 3-iodothyronamine', Molecular BioSystems, 6: 1403-10.

Ieiri, T, Pascale Cochaux, Hector M Targovnik, M Suzuki, S Shimoda, Jason Perret,

and Gilbert Vassart. 1991. 'A 3'splice site mutation in the thyroglobulin gene

responsible for congenital goiter with hypothyroidism', Journal of Clinical

Investigation, 88: 1901.

Irannejad, Roshanak, Jin C Tomshine, Jon R Tomshine, Michael Chevalier, Jacob P

Mahoney, Jan Steyaert, Søren GF Rasmussen, Roger K Sunahara, Hana El-

Samad, and Bo Huang. 2013. 'Conformational biosensors reveal GPCR

signalling from endosomes', Nature, 495: 534-38.

Jagannathan, N. R., N. Tandon, P. Raghunathan, and N. Kochupillai. 1998. 'Reversal

of abnormalities of myelination by thyroxine therapy in congenital

hypothyroidism: localized in vivo proton magnetic resonance spectroscopy

(MRS) study1A preliminary report was presented at the 5th scientific meeting of

the International Society for Magnetic Resonance in Medicine, Vancouver,

Canada, April 1997.1', Developmental Brain Research, 109: 179-86.

Jansen, Jurgen, Edith CH Friesema, Monique HA Kester, Charles E Schwartz, and

Theo J Visser. 2008. 'Genotype-phenotype relationship in patients with

mutations in thyroid hormone transporter MCT8', Endocrinology, 149: 2184-90.

Jayarama-Naidu, Roopa, Jörg Johannes, Franziska Meyer, Eva Katrin Wirth, Lutz

Schomburg, Josef Köhrle, and Kostja Renko. 2015. 'A nonradioactive uptake

assay for rapid analysis of thyroid hormone transporter function', Endocrinology,

156: 2739-45.

Jeremy M Berg, John L Tymoczko, Lubert Stryer. 2002. Biochemistry (W.H. Freeman:

New York).

Johannes, Jörg, Doreen Braun, Anita Kinne, Daniel Rathmann, Josef Köhrle, and

Ulrich Schweizer. 2016. 'Few Amino Acid Exchanges Expand the Substrate

Spectrum of Monocarboxylate Transporter 10', Molecular endocrinology, 30:

796-808.

Johannes, Jörg, Roopa Jayarama-Naidu, Franziska Meyer, Eva Katrin Wirth, Ulrich

Schweizer, Lutz Schomburg, Josef Köhrle, and Kostja Renko. 2016.

'Silychristin, a flavonolignan derived from the milk thistle, is a potent inhibitor of

the thyroid hormone transporter MCT8', Endocrinology, 157: 1694-701.

Johnson, Allan E, Sylvie Audigier, Ferdinand Rossi, Serge Jard, Eliane Tribollet, and

Claude Barberis. 1993. 'Localization and characterization of vasopressin

binding sites in the rat brain using an iodinated linear AVP antagonist', Brain

research, 622: 9-16.

Johnson, Neil. 1955. 'The blood‐supply of the human thyroid gland under normal and

abnormal conditions', British Journal of Surgery, 42: 587-94.

Jones, Iwan, Maya Srinivas, Lily Ng, and Douglas Forrest. 2003. 'The thyroid hormone

receptor β gene: structure and functions in the brain and sensory systems',

Thyroid, 13: 1057-68.

Kalyanaraman, Hema, Raphaela Schwappacher, Jisha Joshua, Shunhui Zhuang,

Brian T Scott, Matthew Klos, Darren E Casteel, John A Frangos, Wolfgang

Dillmann, and Gerry R Boss. 2014. 'Nongenomic thyroid hormone signaling

occurs through a plasma membrane–localized receptor', Sci. Signal., 7: ra48-

ra48.

Kamegai, J, H Tamura, S Ishii, H Sugihara, and I Wakabayashi. 2001. 'Thyroid

hormones regulate pituitary growth hormone secretagogue receptor gene

expression', Journal of neuroendocrinology, 13: 275-78.

Kao, Huaihung Danny, Ashraf Traboulsi, Soichi Itoh, Lewis Dittert, and Anwar Hussain.

2000. 'Enhancement of the systemic and CNS specific delivery of L-dopa by the

nasal administration of its water soluble prodrugs', Pharmaceutical research,

17: 978-84.

Kapoor, Richa, Max van Hogerlinden, Karin Wallis, Himanish Ghosh, Kristina

Nordstrom, Bjorn Vennstrom, and Vidita A Vaidya. 2010. 'Unliganded thyroid

hormone receptor α1 impairs adult hippocampal neurogenesis', The FASEB

Journal, 24: 4793-805.

Kendall, Edward C. 1915. 'The isolation in crystalline form of the compound containing

iodin, which occurs in the thyroid: its chemical nature and physiologic activity',

Journal of the American Medical Association, 64: 2042-43.

Kern, Werner, Jan Born, Herbert Schreiber, and Horst L Fehm. 1999. 'Central nervous

system effects of intranasally administered insulin during euglycemia in men',

Diabetes, 48: 557-63.

Kersseboom, Simone, Sigrun Horn, W Edward Visser, Jiesi Chen, Edith CH Friesema,

Catherine Vaurs-Barrière, Robin P Peeters, Heike Heuer, and Theo J Visser.

2014. 'In vitro and mouse studies supporting therapeutic utility of

triiodothyroacetic acid in MCT8 deficiency', Molecular endocrinology, 28: 1961-

70.

Kim, Robert Y, Robin Gasser, and Graeme J Wistow. 1992. 'mu-crystallin is a

mammalian homologue of Agrobacterium ornithine cyclodeaminase and is

expressed in human retina', Proceedings of the National Academy of Sciences,

89: 9292-96.

Kimura, Tadashi, Osamu Tanizawa, Kensaku Mori, Michael J Brownstein, and Hiroto

Okayama. 1992. 'Structure and expression of a human oxytocin receptor',

Nature, 356: 526.

Kinne, Anita, Gunnar Kleinau, Carolin S Hoefig, Annette Grüters, Josef Köhrle, Gerd

Krause, and Ulrich Schweizer. 2010. 'Essential molecular determinants for

thyroid hormone transport and first structural implications for monocarboxylate

transporter 8', Journal of Biological Chemistry, 285: 28054-63.

Kinne, Anita, Stephan Roth, Heike Biebermann, Josef Köhrle, Annette Grüters, and

Ulrich Schweizer. 2009. 'Surface translocation and tri-iodothyronine uptake of

mutant MCT8 proteins are cell type-dependent', Journal of molecular

endocrinology, 43: 263-71.

Kinne, Anita, Ralf Schülein, and Gerd Krause. 2011. 'Primary and secondary thyroid

hormone transporters', Thyroid research, 4: S7.

Kinne, Anita, Melanie Wittner, Eva K Wirth, Katrin M Hinz, Ralf Schülein, Josef Köhrle,

and Gerd Krause. 2015. 'Involvement of the L-type amino acid transporter Lat2

in the transport of 3, 3′-diiodothyronine across the plasma membrane',

European thyroid journal, 4: 42-50.

Kleinau, Gunnar, Ulrich Schweizer, Anita Kinne, Josef Köhrle, Annette Grüters, Heiko

Krude, and Heike Biebermann. 2011. 'Insights into molecular properties of the

human monocarboxylate transporter 8 by combining functional with structural

information', Thyroid research, 4: S4.

Koehbach, Johannes, Thomas Stockner, Christian Bergmayr, Markus Muttenthaler,

and Christian W. Gruber. 2013. 'Insights into the molecular evolution of oxytocin

receptor ligand binding', Biochemical Society Transactions, 41: 197-204.

Kohrle, J. 2002. 'Iodothyronine deiodinases', Protein Sensors and Reactive Oxygen

Species, Pt a, Selenoproteins and Thioredoxin, 347: 125-67.

Kopp, P. 2001. 'Human Genome and Diseases: Review The TSH receptor and its role

in thyroid disease', Cellular and Molecular Life Sciences, 58: 1301-22.

Koromilas, Christos, Charis Liapi, Kleopatra H. Schulpis, Konstantinos Kalafatakis,

Apostolos Zarros, and Stylianos Tsakiris. 2010. 'Structural and functional

alterations in the hippocampus due to hypothyroidism', Metabolic Brain

Disease, 25: 339-54.

Kotowski, Sarah J, F Woodward Hopf, Taban Seif, Antonello Bonci, and Mark von

Zastrow. 2011. 'Endocytosis promotes rapid dopaminergic signaling', Neuron,

71: 278-90.

Krude, Heiko, Peter Kühnen, and Heike Biebermann. 2015. 'Treatment of congenital

thyroid dysfunction: Achievements and challenges', Best practice & research

Clinical endocrinology & metabolism, 29: 399-413.

Kumar, Niyanta N, Jeffrey J Lochhead, Michelle E Pizzo, Geetika Nehra, Sam

Boroumand, Gretchen Greene, and Robert G Thorne. 2018. 'Delivery of

immunoglobulin G antibodies to the rat nervous system following intranasal

administration: Distribution, dose-response, and mechanisms of delivery',

Journal of Controlled Release, 286: 467-84.

Kuna, Ramya S, Shravan Babu Girada, Suman Asalla, Joyprashant Vallentyne,

Subbareddy Maddika, James T Patterson, David L Smiley, Richard D DiMarchi,

and Prasenjit Mitra. 2013. 'Glucagon-like peptide-1 receptor-mediated

endosomal cAMP generation promotes glucose-stimulated insulin secretion in

pancreatic β-cells', American journal of physiology-endocrinology and

metabolism, 305: E161-E70.

Kunkel, T. A. 1985. 'Rapid and efficient site-specific mutagenesis without phenotypic

selection', Proc Natl Acad Sci U S A, 82: 488-92.

Lafrenière, Ronald G, Laura Carrel, and Huntington F Willard. 1994. 'A novel

transmembrane transporter encoded by the XPCT gene in Xq13. 2', Human

Molecular Genetics, 3: 1133-39.

Lawther, Bradley K., Sajith Kumar, and Hari Krovvidi. 2011. 'Blood–brain barrier',

Continuing Education in Anaesthesia Critical Care & Pain, 11: 128-32.

Lazar, MITCHELL A, THOMAS J Berrodin, and HEATHER P Harding. 1991.

'Differential DNA binding by monomeric, homodimeric, and potentially

heteromeric forms of the thyroid hormone receptor', Molecular and cellular

biology, 11: 5005-15.

Lee, M. R., K. B. Scheidweiler, X. X. Diao, F. Akhlaghi, A. Cummins, M. A. Huestis, L.

Leggio, and B. B. Averbeck. 2017. 'Oxytocin by intranasal and intravenous

routes reaches the cerebrospinal fluid in rhesus macaques: determination using

a novel oxytocin assay', Molecular psychiatry, 23: 115.

Lee, Sang-Min, Jason M. Booe, and Augen A. Pioszak. 2015. 'Structural insights into

ligand recognition and selectivity for classes A, B, and C GPCRs', European

Journal of Pharmacology, 763: 196-205.

Lehnman, IR. 1974. 'DNA ligase: structure, mechanism, and function', Science, 186:

790-97.

Leng, Gareth, and Mike Ludwig. 2016. 'Intranasal Oxytocin: Myths and Delusions',

Biological Psychiatry, 79: 243-50.

Li, Jia-Da, Katherine J. Burton, Chengkang Zhang, Shuang-Bao Hu, and Qun-Yong

Zhou. 2009. 'Vasopressin receptor V1a regulates circadian rhythms of

locomotor activity and expression of clock-controlled genes in the

suprachiasmatic nuclei', American Journal of Physiology - Regulatory,

Integrative and Comparative Physiology, 296: R824-R30.

Lin, Hung-Yun, ShenLi Zhang, Brian L West, Heng-Yuan Tang, Teresa Passaretti,

Faith B Davis, and Paul J Davis. 2003. 'Identification of the putative MAP kinase

docking site in the thyroid hormone receptor-β1 DNA-binding domain: functional

consequences of mutations at the docking site', Biochemistry, 42: 7571-79.

Lochhead, Jeffrey J, Daniel J Wolak, Michelle E Pizzo, and Robert G Thorne. 2015.

'Rapid Transport within Cerebral Perivascular Spaces Underlies Widespread

Tracer Distribution in the Brain after Intranasal Administration', Journal of

Cerebral Blood Flow & Metabolism, 35: 371-81.

Lochhead, Jeffrey J., and Robert G. Thorne. 2012. 'Intranasal delivery of biologics to

the central nervous system', Advanced Drug Delivery Reviews, 64: 614-28.

Locht, Camille, Loic Coutte, and Nathalie Mielcarek. 2011. 'The ins and outs of

pertussis toxin', The FEBS Journal, 278: 4668-82.

Lodish, Berk, Zipursky, et al. 2000. 'Receptor-Mediated Endocytosis and the Sorting

of Internalized Proteins.' in, Molecular Cell Biology (W. H. Freeman: New York).

López-Espíndola, Daniela, Carmen Morales-Bastos, Carmen Grijota-Martínez, Xiao-

Hui Liao, Dorit Lev, Ella Sugo, Charles F Verge, Samuel Refetoff, Juan Bernal,

and Ana Guadaño-Ferraz. 2014. 'Mutations of the thyroid hormone transporter

MCT8 cause prenatal brain damage and persistent hypomyelination', The

Journal of Clinical Endocrinology & Metabolism, 99: E2799-E804.

Loup, F., E. Tribollet, M. Dubois-Dauphin, and J. J. Dreifuss. 1991. 'Localization of

high-affinity binding sites for oxytocin and vasopressin in the human brain. An

autoradiographic study', Brain research, 555: 220-32.

Lyga, Sandra, Silvia Volpe, Ruth C Werthmann, Konrad Gotz, Titiwat Sungkaworn,

Martin J Lohse, and Davide Calebiro. 2016. 'Persistent cAMP signaling by

internalized LH receptors in ovarian follicles', Endocrinology, 2016: 63-71.

Macchia, Paolo Emidio, Paola Lapi, Heiko Krude, Maria Teresa Pirro, Caterina

Missero, Luca Chiovato, Abdallah Souabni, Mariangiola Baserga, Vittorio Tassi,

and Aldo Pinchera. 1998. 'PAX8 mutations associated with congenital

hypothyroidism caused by thyroid dysgenesis', Nature genetics, 19: 83-86.

Machleidt, Thomas, Carolyn C Woodroofe, Marie K Schwinn, Jacqui Méndez, Matthew

B Robers, Kris Zimmerman, Paul Otto, Danette L Daniels, Thomas A Kirkland,

and Keith V Wood. 2015. 'NanoBRET・ A Novel BRET Platform for the Analysis

of Protein–Protein Interactions', ACS chemical biology, 10: 1797-804.

Makino, Ayako, Jorge Suarez, Hong Wang, Darrell D Belke, Brian T Scott, and

Wolfgang H Dillmann. 2008. 'Thyroid hormone receptor-β is associated with

coronary angiogenesis during pathological cardiac hypertrophy', Endocrinology,

150: 2008-15.

Martin, Negin P, Ezequiel Marron Fernandez de Velasco, Fengxia Mizuno, Erica L

Scappini, Bernd Gloss, Christian Erxleben, Jason G Williams, Heather M

Stapleton, Saverio Gentile, and David L Armstrong. 2014. 'A rapid cytoplasmic

mechanism for PI3 kinase regulation by the nuclear thyroid hormone receptor,

TRβ, and genetic evidence for its role in the maturation of mouse hippocampal

synapses in vivo', Endocrinology, 155: 3713-24.

Mauvais-Jarvis, Franck. 2011. 'Estrogen and androgen receptors: regulators of fuel

homeostasis and emerging targets for diabetes and obesity', Trends in

Endocrinology & Metabolism, 22: 24-33.

Mayerl, Steffen, Julia Müller, Reinhard Bauer, Sarah Richert, Celia M Kassmann,

Veerle M Darras, Katrin Buder, Anita Boelen, Theo J Visser, and Heike Heuer.

2014. 'Transporters MCT8 and OATP1C1 maintain murine brain thyroid

hormone homeostasis', The Journal of clinical investigation, 124: 1987-99.

Meyer-Lindenberg, A., B. Kolachana, B. Gold, A. Olsh, K. K. Nicodemus, V. Mattay,

M. Dean, and D. R. Weinberger. 2009. 'Genetic variants in AVPR1A linked to

autism predict amygdala activation and personality traits in healthy humans',

Molecular psychiatry, 14: 968-75.

Meyer-Lindenberg, Andreas, Gregor Domes, Peter Kirsch, and Markus Heinrichs.

2011. 'Oxytocin and vasopressin in the human brain: social neuropeptides for

translational medicine', Nature reviews. Neuroscience, 12: 524.

Molnár, Miklós, and Frank Hertelendy. 1990. 'Regulation of intracellular free calcium in

human myometrial cells by prostaglandin F2 α: comparison with oxytocin', The

Journal of Clinical Endocrinology & Metabolism, 71: 1243-50.

Montrose-Rafizadeh, Chahrzad, Pavel Avdonin, Michael J Garant, Buel D Rodgers,

Sutapa Kole, Huan Yang, Michael A Levine, William Schwindinger, and Michel

Bernier. 1999. 'Pancreatic glucagon-like peptide-1 receptor couples to multiple

G proteins and activates mitogen-activated protein kinase pathways in Chinese

hamster ovary cells', Endocrinology, 140: 1132-40.

Moog, Nora K, Sonja Entringer, Christine Heim, Pathik D Wadhwa, Norbert Kathmann,

and Claudia Buss. 2017. 'Influence of maternal thyroid hormones during

gestation on fetal brain development', Neuroscience, 342: 68-100.

Moolman, JA. 2002. 'Thyroid hormone and the heart: review article', Cardiovascular

journal of south África, 13: 159-63.

Moore, Catherine AC, Shawn K Milano, and Jeffrey L Benovic. 2007. 'Regulation of

receptor trafficking by GRKs and arrestins', Annu. Rev. Physiol., 69: 451-82.

Moran, Carla, Maura Agostini, W Edward Visser, Erik Schoenmakers, Nadia

Schoenmakers, Amaka C Offiah, Ken Poole, Odelia Rajanayagam, Greta

Lyons, and David Halsall. 2014. 'Resistance to thyroid hormone caused by a

mutation in thyroid hormone receptor (TR) α1 and TRα2: clinical, biochemical,

and genetic analyses of three related patients', The lancet Diabetes &

endocrinology, 2: 619-26.

Moran, Carla, Nadia Schoenmakers, Maura Agostini, Erik Schoenmakers, Amaka

Offiah, Anna Kydd, George Kahaly, Susan Mohr-Kahaly, Odelia Rajanayagam,

and Greta Lyons. 2013. 'An adult female with resistance to thyroid hormone

mediated by defective thyroid hormone receptor α', The Journal of Clinical

Endocrinology & Metabolism, 98: 4254-61.

Moreno, José C., Hennie Bikker, Marlies J.E. Kempers, A.S. Paul van Trotsenburg,

Frank Baas, Jan J.M. de Vijlder, Thomas Vulsma, and C. Ris-Stalpers. 2002.

'Inactivating Mutations in the Gene for Thyroid Oxidase 2 (THOX2) and

Congenital Hypothyroidism', New England Journal of Medicine, 347: 95-102.

Moreno, Maria, Pieter de Lange, Assunta Lombardi, Elena Silvestri, Antonia Lanni, and

Fernando Goglia. 2008. 'Metabolic effects of thyroid hormone derivatives',

Thyroid, 18: 239-53.

Mullur, Rashmi, Yan-Yun Liu, and Gregory A Brent. 2014. 'Thyroid hormone regulation

of metabolism', Physiological reviews, 94: 355-82.

Mundy, Gregory R, and Theresa A Guise. 1999. 'Hormonal control of calcium

homeostasis', Clinical Chemistry, 45: 1347-52.

Naganawa, Shinji, Toshiki Nakane, Hisashi Kawai, and Toshiaki Taoka. 2017. 'Lack of

contrast enhancement in a giant perivascular space of the basal ganglion on

delayed FLAIR images: implications for the glymphatic system', Magnetic

Resonance in Medical Sciences, 16: 89-90.

New, David C, and Yung H Wong. 2003. 'CC chemokine receptor-coupled signalling

pathways', Sheng wu hua xue yu sheng wu wu li xue bao Acta biochimica et

biophysica Sinica, 35: 779-88.

Nicholson, Jean L, and Joseph Altman. 1972. 'The effects of early hypo-and

hyperthyroidism on the development of rat cerebellar cortex. I. Cell proliferation

and differentiation', Brain research, 44: 13-23.

Nobori, Masakazu, Shigeki Saiki, Nobutaka Tanaka, Yasushi Harihara, Shunya

Shindo, and Yoshihide Fujimoto. 1994. 'Blood supply of the parathyroid gland

from the superior thyroid artery', Surgery, 115: 417-23.

Oakley, Robert H, Stéphane A Laporte, Jason A Holt, Larry S Barak, and Marc G

Caron. 2001. 'Molecular determinants underlying the formation of stable

intracellular G protein-coupled receptor-β-arrestin complexes after receptor

endocytosis', Journal of Biological Chemistry, 276: 19452-60.

Ohmichi, M., K. Koike, A. Nohara, Y. Kanda, Y. Sakamoto, Z. X. Zhang, K. Hirota, and

A. Miyake. 1995. 'Oxytocin stimulates mitogen-activated protein kinase activity

in cultured human puerperal uterine myometrial cells', Endocrinology, 136:

2082-87.

Palacín, Manuel, Ekaitz Errasti-Murugarren, and Albert Rosell. 2016. 'Heteromeric

amino acid transporters. In search of the molecular bases of transport cycle

mechanisms', Biochemical Society Transactions, 44: 745-52.

Papadimitriou, Anastasios, Alexandra Mihaela Dumitrescu, Antigone Papavasiliou,

Andreas Fretzayas, Polyxeni Nicolaidou, and Samuel Refetoff. 2008. 'A novel

monocarboxylate transporter 8 gene mutation as a cause of severe neonatal

hypotonia and developmental delay', Pediatrics, 121: e199-e202.

Pappa, Theodora, Alfonso Massimiliano Ferrara, and Samuel Refetoff. 2015. 'Inherited

defects of thyroxine-binding proteins', Best practice & research Clinical

endocrinology & metabolism, 29: 735-47.

Paruthiyil, Sreenivasan, Hema Parmar, Vaishali Kerekatte, Gerald R Cunha, Gary L

Firestone, and Dale C Leitman. 2004. 'Estrogen receptor β inhibits human

breast cancer cell proliferation and tumor formation by causing a G2 cell cycle

arrest', Cancer research, 64: 423-28.

Passoni, I., M. Leonzino, V. Gigliucci, B. Chini, and M. Busnelli. 2016. 'Carbetocin is a

Functional Selective Gq Agonist That Does Not Promote Oxytocin Receptor

Recycling After Inducing β-Arrestin-Independent Internalisation', Journal of

neuroendocrinology, 28.

Passos‐Bueno, Maria Rita, Barbara C Byth, Sergio Rosenberg, Reinaldo I Takata,

Egbert Bakker, Alan H Beggs, Rita C Pavanello, Mariz Vainzof, Kay E Davies,

and Mayana Zatz. 1993. 'Severe nonspecific X‐linked mental retardation

caused by a proximally Xp located gene: Intragenic heterogeneity or a new form

of X‐linked mental retardation?', American Journal of Medical Genetics Part A,

46: 172-75.

Patten, JL, PM Smallwood, and C Eil. 1989. 'An initiator codon mutation in the gene

encoding the alpha subunit of Gs in pseudohypoparathyroidism type IA (PHP

IA)', Am J Hum Genet, 45: A212.

Pavlos, N. J., and P. A. Friedman. 2017. 'GPCR Signaling and Trafficking: The Long

and Short of It', Trends Endocrinol Metab, 28: 213-26.

Pedersen, Cort A., Clare M. Gibson, Shane W. Rau, Kayvon Salimi, Kelly L. Smedley,

Robin L. Casey, Jane Leserman, L. Fredrik Jarskog, and David L. Penn. 2011.

'Intranasal oxytocin reduces psychotic symptoms and improves Theory of Mind

and social perception in schizophrenia', Schizophrenia Research, 132: 50-53.

Pennock, GREGORY D, THOMAS E Raya, JOSEPH J Bahl, STEVEN Goldman, and

EUGENE Morkin. 1992. 'Cardiac effects of 3, 5-diiodothyropropionic acid, a

thyroid hormone analog with inotropic selectivity', Journal of Pharmacology and

Experimental Therapeutics, 263: 163-69.

Phaneuf, S., G. N. Europe-Finner, M. Varney, I. Z. MacKenzie, S. P. Watson, and A.

Lopez Bernal. 1993. 'Oxytocin-stimulated phosphoinositide hydrolysis in human

myometrial cells: involvement of pertussis toxin-sensitive and -insensitive G-

proteins', J Endocrinol, 136: 497-509.

Pierce, Kristen L, and Robert J Lefkowitz. 2001. 'Classical and new roles of β-arrestins

in the regulation of G-protein-coupled receptors', Nature reviews neuroscience,

2: 727.

Pierce, Kristen L, Richard T Premont, and Robert J Lefkowitz. 2002. 'Seven-

transmembrane receptors', Nature Reviews Molecular Cell Biology, 3: 639-50.

Pietzner, Maik, Ina Lehmphul, Nele Friedrich, Claudia Schurmann, Till Ittermann,

Marcus Dörr, Matthias Nauck, René Laqua, Uwe Völker, and Georg Brabant.

2015. 'Translating pharmacological findings from hypothyroid rodents to

euthyroid humans: is there a functional role of endogenous 3, 5-T2?', Thyroid,

25: 188-97.

Plateroti, Michelina, Karine Gauthier, Claire Domon-Dell, Jean-Noël Freund, Jacques

Samarut, and Olivier Chassande. 2001. 'Functional interference between

thyroid hormone receptor α (TRα) and natural truncated TR∆α isoforms in the

control of intestine development', Molecular and cellular biology, 21: 4761-72.

Pritchard, John B, Douglas H Sweet, David S Miller, and Ramsey Walden. 1999.

'Mechanism of organic anion transport across the apical membrane of choroid

plexus', Journal of Biological Chemistry, 274: 33382-87.

Refetoff, Samuel, Loren T DeWind, and Leslie J DeGroot. 1967. 'Familial syndrome

combining Deaf-Mutism, stippled epiphyses, goiter and abnormally high PBI:

possible target organ refractoriness to thyroid hormone 1 2', The Journal of

Clinical Endocrinology & Metabolism, 27: 279-94.

Refetoff, Samuel, and Alexandra M Dumitrescu. 2007. 'Syndromes of reduced

sensitivity to thyroid hormone: genetic defects in hormone receptors, cell

transporters and deiodination', Best practice & research Clinical endocrinology

& metabolism, 21: 277-305.

Roberts, Lori M, Kathleen Woodford, Mei Zhou, Deborah S Black, Jill E Haggerty,

Emily H Tate, Kent K Grindstaff, Wondwessen Mengesha, Chandrasekaran

Raman, and Noa Zerangue. 2008. 'Expression of the thyroid hormone

transporters monocarboxylate transporter-8 (SLC16A2) and organic ion

transporter-14 (SLCO1C1) at the blood-brain barrier', Endocrinology, 149:

6251-61.

Roed, Sarah Noerklit, Anne Cathrine Nøhr, Pernille Wismann, Helle Iversen, Hans

Bräuner-Osborne, Sanne Moeller Knudsen, and Maria Waldhoer. 2015.

'Functional consequences of glucagon-like peptide-1 receptor cross-talk and

trafficking', Journal of Biological Chemistry, 290: 1233-43.

Roed, Sarah Noerklit, Pernille Wismann, Christina Rye Underwood, Nikolaj Kulahin,

Helle Iversen, Karen Arevad Cappelen, Lauge Schäffer, Janne Lehtonen, Jacob

Hecksher-Soerensen, Anna Secher, Jesper Mosolff Mathiesen, Hans Bräuner-

Osborne, Jennifer L. Whistler, Sanne Moeller Knudsen, and Maria Waldhoer.

2014. 'Real-time trafficking and signaling of the glucagon-like peptide-1

receptor', Molecular and Cellular Endocrinology, 382: 938-49.

Rosenbaum, Daniel M, Søren GF Rasmussen, and Brian K Kobilka. 2009. 'The

structure and function of G-protein-coupled receptors', Nature, 459: 356-63.

Roth, Stephan, Anita Kinne, and Ulrich Schweizer. 2010. 'The tricyclic antidepressant

desipramine inhibits T3 import into primary neurons', Neuroscience Letters,

478: 5-8.

Runge, Steffen, Henning Thøgersen, Kjeld Madsen, Jesper Lau, and Rainer Rudolph.

2008. 'Crystal structure of the ligand-bound glucagon-like peptide-1 receptor

extracellular domain', Journal of Biological Chemistry, 283: 11340-47.

Sakaguchi, YOSHIHIDE, GUANGGEN Cui, and LUYI Sen. 1996. 'Acute effects of

thyroid hormone on inward rectifier potassium channel currents in guinea pig

ventricular myocytes', Endocrinology, 137: 4744-51.

Sakurai, Akihiro, Kyoko Takeda, Kenneth Ain, Paola Ceccarelli, Akira Nakai, Susumu

Seino, Graeme I Bell, Samuel Refetoff, and Leslie J DeGroot. 1989.

'Generalized resistance to thyroid hormone associated with a mutation in the

ligand-binding domain of the human thyroid hormone receptor beta',

Proceedings of the National Academy of Sciences, 86: 8977-81.

Sanger, F., S. Nicklen, and A. R. Coulson. 1977. 'DNA sequencing with chain-

terminating inhibitors', Proc Natl Acad Sci U S A, 74: 5463-7.

Scharbo-Dehaan, Marianne. 1996. 'Hormone replacement therapy', The Nurse

practitioner, 21: 1-13.

Schoenmakers, Nadia, Carla Moran, Robin P Peeters, Theo Visser, Mark Gurnell, and

Krishna Chatterjee. 2013. 'Resistance to thyroid hormone mediated by defective

thyroid hormone receptor alpha', Biochimica et Biophysica Acta (BBA)-General

Subjects, 1830: 4004-08.

Schöneberg, Torsten, Michael Hofreiter, Angela Schulz, and Holger Römpler. 2007.

'Learning from the past: evolution of GPCR functions', Trends in

pharmacological sciences, 28: 117-21.

Schöneberg, Torsten, Evi Kostenis, Jie Liu, Thomas Gudermann, and Jürgen Wess.

1998. 'Molecular aspects of vasopressin receptor function.' in, Vasopressin and

Oxytocin (Springer).

Schorderet, D. F. 1994. 'Polymerase chain reaction: basic principles', Schweiz

Rundsch Med Prax, 83: 588-94.

Schussler, George C. 2000. 'The thyroxine-binding proteins', Thyroid, 10: 141-49.

Schwartz, Charles E, Melanie M May, Nancy J Carpenter, R Curtis Rogers, Judith

Martin, Martin G Bialer, Jewell Ward, Javier Sanabria, Silvana Marsa, and

James A Lewis. 2005. 'Allan-Herndon-Dudley syndrome and the

monocarboxylate transporter 8 (MCT8) gene', The American Journal of Human

Genetics, 77: 41-53.

Schweizer, Ulrich, Jörg Johannes, Dorothea Bayer, and Doreen Braun. 2014.

'Structure and function of thyroid hormone plasma membrane transporters',

European thyroid journal, 3: 143-53.

Schwenk, Robert W, Christian Baumeier, Brian Finan, Oliver Kluth, Christine Brauer,

Hans-Georg Joost, Richard D DiMarchi, Matthias H Tschöp, and Annette

Schürmann. 2015. 'GLP-1–oestrogen attenuates hyperphagia and protects

from beta cell failure in diabetes-prone New Zealand obese (NZO) mice',

Diabetologia, 58: 604-14.

Segal, J, and SH Ingbar. 1990. '3, 5, 3′-Tri-iodothyronine enhances sugar transport in

rat thymocytes by increasing the intrinsic activity of the plasma membrane sugar

transporter', Journal of Endocrinology, 124: 133-40.

Sen, Luyi, Yoshihide Sakaguchi, and Guanggen Cui. 2002. 'G protein modulates

thyroid hormone-induced Na^+ channel activation in ventricular myocytes',

AMERICAN JOURNAL OF PHYSIOLOGY, 283: H2119-H29.

Shendure, J. A., G. J. Porreca, G. M. Church, A. F. Gardner, C. L. Hendrickson, J.

Kieleczawa, and B. E. Slatko. 2011. 'Overview of DNA sequencing strategies',

Curr Protoc Mol Biol, Chapter 7: Unit7 1.

Shenoy, S. K., M. T. Drake, C. D. Nelson, D. A. Houtz, K. Xiao, S. Madabushi, E.

Reiter, R. T. Premont, O. Lichtarge, and R. J. Lefkowitz. 2006. 'beta-arrestin-

dependent, G protein-independent ERK1/2 activation by the beta2 adrenergic

receptor', J Biol Chem, 281: 1261-73.

Shibusawa, Nobuyuki, Koshi Hashimoto, Amisra A Nikrodhanond, M Charles

Liberman, Meredithe L Applebury, Xiao Hui Liao, Janet T Robbins, Samuel

Refetoff, Ronald N Cohen, and Fredric E Wondisford. 2003. 'Thyroid hormone

action in the absence of thyroid hormone receptor DNA-binding in vivo', The

Journal of clinical investigation, 112: 588-97.

Shukla, A. K. 2014. 'Biasing GPCR Signaling from Inside', Science Signaling, 7: pe3-

pe3.

Shukla, Arun K, Garima Singh, and Eshan Ghosh. 2014. 'Emerging structural insights

into biased GPCR signaling', Trends in biochemical sciences, 39: 594-602.

Singleton, Paul. 2010. Dictionary of DNA and Genome Technology (Wiley-Blackwell).

Skoglund, Gunnar, Mehboob A Hussain, and George G Holz. 2000. 'Glucagon-like

peptide 1 stimulates insulin gene promoter activity by protein kinase A-

independent activation of the rat insulin I gene cAMP response element',

Diabetes, 49: 1156-64.

Smith, Jeffrey S, Robert J Lefkowitz, and Sudarshan Rajagopal. 2018. 'Biased

signalling: from simple switches to allosteric microprocessors', Nature Reviews

Drug Discovery, 17: 243.

Stiles, Joan, and Terry L Jernigan. 2010. 'The basics of brain development',

Neuropsychology review, 20: 327-48.

Stitzer, L KENT, and JOHN A Jacquez. 1975. 'Neutral amino acid transport pathways

in uptake of L-thyroxine by Ehrlich ascites cells', American Journal of

Physiology--Legacy Content, 229: 172-77.

Stoop, Ron. 2012. 'Neuromodulation by Oxytocin and Vasopressin', Neuron, 76: 142-

59.

Storey, Nina M, Saverio Gentile, Hemayet Ullah, Angela Russo, Michelle Muessel,

Christian Erxleben, and David L Armstrong. 2006. 'Rapid signaling at the

plasma membrane by a nuclear receptor for thyroid hormone', Proc Natl Acad

Sci U S A, 103: 5197-201.

Strakova, Z, and MELVYN S Soloff. 1997. 'Coupling of oxytocin receptor to G proteins

in rat myometrium during labor: Gi receptor interaction', American journal of

physiology-endocrinology and metabolism, 272: E870-E76.

Strakova, Zuzana, John A Copland, Stephen J Lolait, and Melvyn S Soloff. 1998.

'ERK2 mediates oxytocin-stimulated PGE 2 synthesis', American journal of

physiology-endocrinology and metabolism, 274: E634-E41.

Sun, Zhuo-Qian, Kaie Ojamaa, William A Coetzee, Michael Artman, and Irwin Klein.

2000. 'Effects of thyroid hormone on action potential and repolarizing currents

in rat ventricular myocytes', American journal of physiology-endocrinology and

metabolism, 278: E302-E07.

Szeto, Angela, Philip M McCabe, Daniel A Nation, Benjamin A Tabak, Maria A

Rossetti, Michael E McCullough, Neil Schneiderman, and Armando J Mendez.

2011. 'Evaluation of enzyme immunoassay and radioimmunoassay methods for

the measurement of plasma oxytocin', Psychosomatic medicine, 73: 393.

Tahara, A, M Saito, T Sugimoto, Y Tomura, K Wada, T Kusayama, J Tsukada, N Ishii,

T Yatsu, and W Uchida. 1997. 'Pharmacological characterization of YM087, a

potent, nonpeptide human vasopressin V1A and V2 receptor antagonist',

Naunyn-Schmiedeberg's archives of pharmacology, 357: 63-69.

Tahara, Atsuo, Masayuki Saito, Toru Sugimoto, Yuichi Tomura, Koh-ichi Wada,

Toshiyuki Kusayama, Junko Tsukada, Noe Ishii, Takeyuki Yatsu, and Wataru

Uchida. 1999. 'AVP-induced mitogenic responses of Chinese hamster ovary

cells expressing human V1A or V1B receptors', Pflügers Archiv, 437: 219-26.

Tahara, Atsuo, Masayuki Saito, Toru Sugimoto, Yuichi Tomura, Koh‐ichi Wada,

Toshiyuki Kusayama, Junko Tsukada, Noe Ishii, Takeyuki Yatsu, and Wataru

Uchida. 1998. 'Pharmacological characterization of the human vasopressin

receptor subtypes stably expressed in Chinese hamster ovary cells', British

journal of pharmacology, 125: 1463-70.

Tam, Charmaine S, Virgile Lecoultre, and Eric Ravussin. 2012. 'Brown Adipose

Tissue', Circulation, 125: 2782-91.

Taurog, Alvin, Martha L Dorris, and Daniel R Doerge. 1996. 'Mechanism of

simultaneous iodination and coupling catalyzed by thyroid peroxidase', Archives

of biochemistry and biophysics, 330: 24-32.

Taylor, Peter M, and James WA Ritchie. 2007. 'Tissue uptake of thyroid hormone by

amino acid transporters', Best practice & research Clinical endocrinology &

metabolism, 21: 237-51.

Terrillon, Sonia, Claude Barberis, and Michel Bouvier. 2004. 'Heterodimerization of

V1a and V2 vasopressin receptors determines the interaction with β-arrestin

and their trafficking patterns', Proc Natl Acad Sci U S A, 101: 1548-53.

Thibonnier, Marc, Colette Auzan, Zuhayr Madhun, Pamela Wilkins, Liliana Berti-

Mattera, and Eric Clauser. 1994. 'Molecular cloning, sequencing, and functional

expression of a cDNA encoding the human V1a vasopressin receptor', Journal

of Biological Chemistry, 269: 3304-10.

Thompson, Aiysha, and Venkateswarlu Kanamarlapudi. 2015. 'Agonist-induced

internalisation of the glucagon-like peptide-1 receptor is mediated by the Gαq

pathway', Biochemical pharmacology, 93: 72-84.

Thorne, R. G., G. J. Pronk, V. Padmanabhan, and W. H. Frey. 2004. 'Delivery of

insulin-like growth factor-I to the rat brain and spinal cord along olfactory and

trigeminal pathways following intranasal administration', Neuroscience, 127:

481-96.

Tiano, Joseph P, Chandra R Tate, Bin S Yang, Richard DiMarchi, and Franck Mauvais-

Jarvis. 2015. 'Effect of targeted estrogen delivery using glucagon-like peptide-1

on insulin secretion, insulin sensitivity and glucose homeostasis', Scientific

reports, 5.

Tobin, Andrew B, Adrian J Butcher, and Kok Choi Kong. 2008. 'Location, location,

location… site-specific GPCR phosphorylation offers a mechanism for cell-type-

specific signalling', Trends in pharmacological sciences, 29: 413-20.

Trajkovic, Marija, Theo J Visser, Jens Mittag, Sigrun Horn, Jan Lukas, Veerle M

Darras, Genadij Raivich, Karl Bauer, and Heike Heuer. 2007. 'Abnormal thyroid

hormone metabolism in mice lacking the monocarboxylate transporter 8', The

Journal of clinical investigation, 117: 627-35.

Tylki-Szymanska, A., R. Acuna-Hidalgo, M. Krajewska-Walasek, A. Lecka-Ambroziak,

M. Steehouwer, C. Gilissen, H. G. Brunner, A. Jurecka, A. Rozdzynska-

Swiatkowska, A. Hoischen, and K. H. Chrzanowska. 2015. 'Thyroid hormone

resistance syndrome due to mutations in the thyroid hormone receptor a gene

(THRA)', Journal of medical genetics, 52: 312-16.

Tzabazis, Alexander, Jordan Mechanic, James Miller, Michael Klukinov, Conrado

Pascual, Neil Manering, Dean S. Carson, Allon Jacobs, Yanli Qiao, Jason

Cuellar, William H. Frey, Daniel Jacobs, Martin Angst, and David C. Yeomans.

2015. 'Oxytocin receptor: Expression in the trigeminal nociceptive system and

potential role in the treatment of headache disorders', Cephalalgia, 36: 943-50.

Underwood, Christina Rye, Patrick Garibay, Lotte Bjerre Knudsen, Sven Hastrup,

Günther H Peters, Rainer Rudolph, and Steffen Reedtz-Runge. 2010. 'Crystal

structure of glucagon-like peptide-1 in complex with the extracellular domain of

the glucagon-like peptide-1 receptor', Journal of Biological Chemistry, 285: 723-

30.

van Gucht, Anja L. M., Carla Moran, Marcel E. Meima, W. Edward Visser, Krishna

Chatterjee, Theo J. Visser, and Robin P. Peeters. 2017. 'Chapter Twelve -

Resistance to Thyroid Hormone due to Heterozygous Mutations in Thyroid

Hormone Receptor Alpha.' in Douglas Forrest and Sophia Tsai (eds.), Current

Topics in Developmental Biology (Academic Press).

Vargas-Uricoechea, Hernando, Anilsa Bonelo-Perdomo, and Carlos Hernán Sierra-

Torres. 2014. 'Effects of thyroid hormones on the heart', Clínica e Investigación

en Arteriosclerosis, 26: 296-309.

Vatine, Gad D, Abraham Al-Ahmad, Bianca K Barriga, Soshana Svendsen, Ariel Salim,

Leslie Garcia, Veronica J Garcia, Ritchie Ho, Nur Yucer, and Tongcheng Qian.

2017. 'Modeling psychomotor retardation using iPSCs from MCT8-deficient

patients indicates a prominent role for the blood-brain barrier', Cell Stem Cell,

20: 831-43. e5.

Vatine, Gad David, David Zada, Tali Lerer-Goldshtein, Adi Tovin, Guy Malkinson,

Karina Yaniv, and Lior Appelbaum. 2013. 'Zebrafish as a model for

monocarboxyl transporter 8-deficiency', Journal of Biological Chemistry, 288:

169-80.

Venero, César, Ana Guadaño-Ferraz, Ana Isabel Herrero, Kristina Nordström, Jimena

Manzano, Gabriella Moreale de Escobar, Juan Bernal, and Björn Vennström.

2005. 'Anxiety, memory impairment, and locomotor dysfunction caused by a

mutant thyroid hormone receptor α1 can be ameliorated by T3 treatment',

Genes & development, 19: 2152-63.

Vennström, Björn, Jens Mittag, and Karin Wallis. 2008. 'Severe psychomotor and

metabolic damages caused by a mutant thyroid hormone receptor alpha 1 in

mice: can patients with a similar mutation be found and treated?', Acta

Paediatrica, 97: 1605-10.

Verge, Charles F, Daniel Konrad, Michal Cohen, Caterina Di Cosmo, Alexandra M

Dumitrescu, Teresa Marcinkowski, Shihab Hameed, Jill Hamilton, Roy E Weiss,

and Samuel Refetoff. 2012. 'Diiodothyropropionic acid (DITPA) in the treatment

of MCT8 deficiency', The Journal of Clinical Endocrinology & Metabolism, 97:

4515-23.

Vie, Marie-Pierre, Claudine Evrard, Jeannine Osty, Aline Breton-Gilet, Pascal

Blanchet, Martine Pomerance, Pierre Rouget, Jacques Francon, and Jean-Paul

Blondeau. 1997. 'Purification, molecular cloning, and functional expression of

the human nicodinamide-adenine dinucleotide phosphate-regulated thyroid

hormone-binding protein', Molecular endocrinology, 11: 1728-36.

Vilardaga, Jean-Pierre, Thomas J Gardella, Vanessa L Wehbi, and Timothy N

Feinstein. 2012. 'Non-canonical signaling of the PTH receptor', Trends in

pharmacological sciences, 33: 423-31.

Vilardaga, Jean-Pierre, Frederic G Jean-Alphonse, and Thomas J Gardella. 2014.

'Endosomal generation of cAMP in GPCR signaling', Nature chemical biology,

10: 700-06.

Visser, W Edward, Edith CH Friesema, and Theo J Visser. 2011. 'Minireview: thyroid

hormone transporters: the knowns and the unknowns', Molecular

endocrinology, 25: 1-14.

Visser, W Edward, Jurgen Jansen, Edith CH Friesema, Monique HA Kester, Edna

Mancilla, Johan Lundgren, Marjo S van der Knaap, Roelineke J Lunsing,

Oebele F Brouwer, and Theo J Visser. 2009. 'Novel pathogenic mechanism

suggested by ex vivo analysis of MCT8 (SLC16A2) mutations', Human

mutation, 30: 29-38.

Visser, W Edward, Nancy J Philp, Thamar B van Dijk, Wim Klootwijk, Edith CH

Friesema, Jurgen Jansen, Philip W Beesley, Alexandra G Ianculescu, and Theo

J Visser. 2009. 'Evidence for a homodimeric structure of human

monocarboxylate transporter 8', Endocrinology, 150: 5163-70.

Visser, W Edward, Paul Vrijmoeth, Frank E Visser, Willem Frans M Arts, Hans Toor,

and Theo J Visser. 2013. 'Identification, functional analysis, prevalence and

treatment of monocarboxylate transporter 8 (MCT8) mutations in a cohort of

adult patients with mental retardation', Clinical endocrinology, 78: 310-15.

Voet, Donald, and Judith G Voet. 2011. 'Biochemistry, 4-Th Edition', NewYork: John

Wiley& SonsInc: 492-96.

Vogel, Heike, Stefanie Wolf, Cristina Rabasa, Francisca Rodriguez-Pacheco, Carina

S Babaei, Franziska Stöber, Jürgen Goldschmidt, Richard D DiMarchi, Brian

Finan, and Matthias H Tschöp. 2016. 'GLP-1 and estrogen conjugate acts in the

supramammillary nucleus to reduce food-reward and body weight',

Neuropharmacology, 110: 396-406.

Wagner, Carsten A, Florian Lang, and Stefan Bröer. 2001. 'Function and structure of

heterodimeric amino acid transporters', American Journal of Physiology-Cell

Physiology, 281: C1077-C93.

Wallis, Karin, Maria Sjögren, Max Van Hogerlinden, Gilad Silberberg, André Fisahn,

Kristina Nordström, Lars Larsson, Håkan Westerblad, Gabriela Morreale De

Escobar, and Oleg Shupliakov. 2008. 'Locomotor deficiencies and aberrant

development of subtype-specific GABAergic interneurons caused by an

unliganded thyroid hormone receptor α1', Journal of Neuroscience, 28: 1904-

15.

Weetman, Anthony P. 2000. 'Graves' disease', New England Journal of Medicine, 343:

1236-48.

Wheeler, MICHAEL B, MING Lu, JOSEPH S Dillon, XING-HONG Leng, CHUAN Chen,

and AE Boyd 3rd. 1993. 'Functional expression of the rat glucagon-like peptide-

I receptor, evidence for coupling to both adenylyl cyclase and phospholipase-

C', Endocrinology, 133: 57-62.

Wikström, Lilian, Catarina Johansson, Carmen Saltó, Carrolee Barlow, Angel Campos

Barros, Frank Baas, Douglas Forrest, Peter Thorén, and Björn Vennström.

1998. 'Abnormal heart rate and body temperature in mice lacking thyroid

hormone receptor α1', The EMBO journal, 17: 455-61.

Williams, GR, Helen Robson, and Stephen M Shalet. 1998. 'Thyroid hormone actions

on cartilage and bone: interactions with other hormones at the epiphyseal plate

and effects on linear growth', Journal of Endocrinology, 157: 391-403.

Williams, Graham R. 2000. 'Cloning and characterization of two novel thyroid hormone

receptor β isoforms', Molecular and cellular biology, 20: 8329-42.

Wirth, E. K., and F. Meyer. 2017. 'Neuronal effects of thyroid hormone metabolites',

Molecular and Cellular Endocrinology, 458: 136-42.

Wirth, Eva K, Stephan Roth, Cristiane Blechschmidt, Sabine M Hölter, Lore Becker,

Ildiko Racz, Andreas Zimmer, Thomas Klopstock, Valerie Gailus-Durner, and

Helmut Fuchs. 2009. 'Neuronal 3′, 3, 5-triiodothyronine (T3) uptake and

behavioral phenotype of mice deficient in Mct8, the neuronal T3 transporter

mutated in Allan–Herndon–Dudley syndrome', Journal of Neuroscience, 29:

9439-49.

Wirth, Eva K., Stephan Roth, Cristiane Blechschmidt, Sabine M. Hölter, Lore Becker,

Ildiko Racz, Andreas Zimmer, Thomas Klopstock, Valerie Gailus-Durner,

Helmut Fuchs, Wolfgang Wurst, Thomas Naumann, Anja Bräuer, Martin Hrabé

de Angelis, Josef Köhrle, Annette Grüters, and Ulrich Schweizer. 2009.

'Neuronal 3′,3,5-Triiodothyronine (T3) Uptake and Behavioral Phenotype of

Mice Deficient in Mct8, the Neuronal T3 Transporter Mutated in Allan–Herndon–

Dudley Syndrome', The Journal of Neuroscience, 29: 9439-49.

Xu, Yong, Thekkethil P Nedungadi, Liangru Zhu, Nasim Sobhani, Boman G Irani,

Kathryn E Davis, Xiaorui Zhang, Fang Zou, Lana M Gent, and Lisa D Hahner.

2011. 'Distinct hypothalamic neurons mediate estrogenic effects on energy

homeostasis and reproduction', Cell metabolism, 14: 453-65.

Yoshida, Akio, Shinichi Taniguchi, Ichiro Hisatome, Ines E Royaux, Eric D Green,

Leonard D Kohn, and Koichi Suzuki. 2002. 'Pendrin is an iodide-specific apical

porter responsible for iodide efflux from thyroid cells', The Journal of Clinical

Endocrinology & Metabolism, 87: 3356-61.

Zada, David, Adi Tovin, Tali Lerer-Goldshtein, Gad David Vatine, and Lior Appelbaum.

2014. 'Altered behavioral performance and live imaging of circuit-specific neural

deficiencies in a zebrafish model for psychomotor retardation', PLoS Genet, 10:

e1004615.

Zevenbergen, C. 2015. 'Cellular mechanisms of thyroid hormone homeostasis',

Erasmus University Rotterdam.

Zhang, Guodong, Yizhong Zhang, Douglas M. Fast, Zhaosheng Lin, and Rick

Steenwyk. 2011. 'Ultra sensitive quantitation of endogenous oxytocin in rat and

human plasma using a two-dimensional liquid chromatography–tandem mass

spectrometry assay', Analytical Biochemistry, 416: 45-52.

Zhang, Hao, Yu-Ning Song, Wei-Guo Liu, Xiu-Li Guo, and Lu-Gang Yu. 2010.

'Regulation and role of organic anion-transporting polypeptides (OATPs) in drug

delivery at the choroid plexus', Journal of Clinical Neuroscience, 17: 679-84.

Zhang, Jibo, Vincent R. Roggero, and Lizabeth A. Allison. 2018. 'Chapter Three -

Nuclear Import and Export of the Thyroid Hormone Receptor.' in Gerald Litwack

(ed.), Vitamins and Hormones (Academic Press).

Zhang, Qi-Zhi, Liu-Sheng Zha, Yan Zhang, Wen-Ming Jiang, Wei Lu, Zhen-Qi Shi, Xin-

Guo Jiang, and Shou-Kuan Fu. 2006. 'The brain targeting efficiency following

nasally applied MPEG-PLA nanoparticles in rats', Journal of Drug Targeting, 14:

281-90.

Zhong, Miao, Ming Yang, and Barbara M. Sanborn. 2003. 'Extracellular Signal-

Regulated Kinase 1/2 Activation by Myometrial Oxytocin Receptor Involves

GαqGβγ and Epidermal Growth Factor Receptor Tyrosine Kinase Activation',

Endocrinology, 144: 2947-56.

Zingg, Hans H, Charles W Bourque, and Daniel G Bichet. 2012. Vasopressin and

oxytocin: molecular, cellular, and clinical advances (Springer Science &

Business Media).

Zorick, Todd S, Suzana Kleimann, A Sertie, Mayana Zatz, Sérgio Rosenberg, and

Maria Rita Passos‐Bueno. 2004. 'Fine mapping and clinical reevaluation of a

Brazilian pedigree with a severe form of X‐linked mental retardation associated

with other neurological dysfunction', American Journal of Medical Genetics Part

A, 127: 321-23.

Zwanziger, Denise, Mathias Schmidt, Jana Fischer, Gunnar Kleinau, Doreen Braun,

Ulrich Schweizer, Lars Christian Moeller, Heike Biebermann, and Dagmar

Fuehrer. 2016. 'The long N-terminus of the human monocarboxylate transporter

8 is a target of ubiquitin-dependent proteasomal degradation which regulates

protein expression and oligomerization capacity', Molecular and Cellular

Endocrinology, 434: 278-87.

7 Supplements

7.1 Cloning primers

Table S 1 List of all cloning primers used for this project

Gene Recipient

plasmid

Primer

F/R Sequence

rARRB2 pFC32K F AACTGCGATCGCCATGGGGGAGAAACCCGGGACC

R ACGCGTTTAAACGCAGAGTTGATCATCATAGTCGTCATCCTTCATCC

GLP1R pcDps F CTGAATTCGCCACCATGGCCGGCGCCCCCGGCCCGCTGC

R CCACTAGTTCAGCTGCAGGAATTTTGGCAGGTGGC

pSNAPf F GTCAGCTAGCATGGCCGGCGCCCCC (signal peptide)

R GTCAGAATCCGGGGCCGGCCCTGCC (signal peptide)

F AAGACTCGAGCGCCCCCAGGGTGCCACT (mature peptide)

R TAGATTAATTAATCAGCTGCAGGAGGCCTGGCA (mature peptide)

pFC14A F GTCGGCGATCGCCATGGCCGGCGCCCCCGGCCC

R TGTTGTTTAAACGCTGCAGGAGGCCTGGCAAGTGGCTGT

pBiT3.1-

secN

F TATATCTAGACGCCCCCAGGGTGCCACTGTGTCC

R TCGAAAGCTTTCAGCTGCAGGAGGCCTGGCAAGTGGC

OTR pcDps F ATATAGAATTCGCCACCATGGAGGGCGCGCTCGCAG

R CTGCACTAGTTCACGCCGTGGATGGCTGGGAGCA

PFC14A F GACCGCGATCGCCATGGAGGGCGCGCTCGCAGCCAA

R CGCGGTTTAAACCGCCGTGGATGGCTGGGAGCA

V1AR pcDps F ATATGAATTCGCCACCATGCGTCTCTCCGCC

R GAGCACTAGTTCAAGTTGAAACAGGAATGAATTTGATGGACTTGGA

PFC14A F GACCGCGATCGCCATGCGTCTCTCCGCCGGTCCC

R TGTCGTTTAAACAGTTGAAACAGGAATGAATTTGATGGACTTGGA

7.2 Genes of interest (cDNA)

All cloned constructs were sequenced using the Sanger-method and the sequence

were aligned with the known cDNA from databases such as NCBI (National Center for

Biotechnology Information) or Ensembl gene browser using the BLAST (basic local

alignment search tool).

ARRB2 (rattus norvegicus) NM_012911, 1232 bp, 410 aa

ATGGGTGAAAAACCCGGGACCAGGGTCTTCAAGAAGTCGAGCCCTAACTGCAAGCTCACCGTGT

ACTTGGGCAAGCGTGACTTTGTGGATCACTTGGACAAAGTGGATCCTGTCGATGGTGTGGTGCTT

GTGGATCCTGACTACTTGAAGGACCGGAAAGTGTTTGTGACCCTCACCTGTGCCTTCCGCTATGG

CCGAGAAGACCTGGATGTACTGGGCCTGTCTTTCCGCAAAGATCTGTTCATCGCCACCTACCAG

GCCTTCCCCCCCATGCCCAACCCACCTCGGCCCCCCACCCGCCTACAGGACCGACTGCTGAAG

AAGTTGGGCCAGCATGCCCACCCCTTTTTTTTCACAATACCCCAGAATTTGCCTTGCTCCGTCACA

CTGCAGCCAGGACCGGAGGACACAGGGAAGGCCTGTGGAGTAGACTTTGAGATTCGAGCCTTCT

GTGCCAAATCTATAGAAGAAAAAAGCCACAAAAGGAACTCCGTGCGGCTTATCATCAGAAAGGTA

CAGTTTGCTCCTGAGACACCCGGCCCCCAGCCATCAGCTGAAACCACACGCCACTTCCTCATGT

CTGACCGGAGGTCCCTGCACCTAGAGGCTTCCCTGGACAAAGAGCTGTACTACCATGGGGAACC

CCTCAATGTCAACGTCCACGTCACCAACAATTCTGCCAAGACCGTCAAGAAGATCAGAGTGTCTG

TGAGACAGTATGCCGACATTTGCCTCTTCAGCACCGCGCAGTACAAGTGTCCTGTGGCTCAGCTT

GAACAAGATGACCAGGTGTCTCCCAGTTCCACATTCTGCAAGGTGTACACCATAACCCCGCTGCT

CAGTGACAACCGAGAGAAGCGTGGCCTTGCCCTTGATGGGCAACTCAAGCACGAAGACACCAAC

CTGGCTTCCAGCACCATTGTGAAGGAGGGAGCCAACAAGGAGGTGCTGGGAATCCTAGTATCCT

ACAGGGTCAAGGTGAAGCTGGTGGTGTCTCGAGGCGGGGATGTCTCCGTGGAGCTACCTTTCGT

CCTAATGCACCCCAAGCCCCACGACCACATCACCCTTCCCCGACCCCAGTCAGCCCCCCGGGAA

ATAGACATCCCTGTGGATACCAACCTCATTGAATTCGATACCAACTATGCCACAGACGACGACAT

CGTGTTTGAGGACTTTGCGAGGCTTCGGCTGAAGGGGATGAAGGATGACGACTGTGATGACCAG

TTCTGCTAG

GLP1R (homo sapiens) NM_002062.3, 1391 bp, 463 aa

ATGGCCGGCGCCCCCGGCCCGCTGCGCCTTGCGCTGCTGCTGCTCGGGATGGTGGGCAGGGC

CGGCCCCCGCCCCCAGGGTGCCACTGTGTCCCTCTGGGAGACGGTGCAGAAATGGCGAGAATA

CCGACGCCAGTGCCAGCGCTCCCTGACTGAGGATCCACCTCCTGCCACAGACTTGTTCTGCAAC

CGGACCTTCGATGAATACGCCTGCTGGCCAGATGGGGAGCCAGGCTCGTTCGTGAATGTCAGCT

GCCCCTGGTACCTGCCCTGGGCCAGCAGTGTGCCGCAGGGCCACGTGTACCGGTTCTGCACAG

CTGAAGGCCTCTGGCTGCAGAAGGACAACTCCAGCCTGCCCTGGAGGGACTTGTCGGAGTGCG

AGGAGTCCAAGCGAGGGGAAAGAAGCTCCCCGGAGGAGCAGCTCCTGTTCCTCTACATCATCTA

CACGGTGGGCTACGCACTCTCCTTCTCTGCTCTGGTTATCGCCTCTGCGATCCTCCTCGGCTTCA

GACACCTGCACTGCACCAGGAACTACATCCACCTGAACCTGTTTGCATCCTTCATCCTGCGAGCA

TTGTCCGTCTTCATCAAGGACGCAGCCCTGAAGTGGATGTATAGCACAGCCGCCCAGCAGCACC

AGTGGGATGGGCTCCTCTCCTACCAGGACTCTCTGAGCTGCCGCCTGGTGTTTCTGCTCATGCA

GTACTGTGTGGCGGCCAATTACTACTGGCTCTTGGTGGAGGGCGTGTACCTGTACACACTGCTG

GCCTTCTCGGTCTTATCTGAGCAATGGATCTTCAGGCTCTACGTGAGCATAGGCTGGGGTGTTCC

CCTGCTGTTTGTTGTCCCCTGGGGCATTGTCAAGTACCTCTATGAGGACGAGGGCTGCTGGACC

AGGAACTCCAACATGAACTACTGGCTCATTATCCGGCTGCCCATTCTCTTTGCCATTGGGGTGAA

CTTCCTCATCTTTGTTCGGGTCATCTGCATCGTGGTATCCAAACTGAAGGCCAATCTCATGTGCAA

GACAGACATCAAATGCAGACTTGCCAAGTCCACGCTGACACTCATCCCCCTGCTGGGGACTCAT

GAGGTCATCTTTGCCTTTGTGATGGACGAGCACGCCCGGGGGACCCTGCGCTTCATCAAGCTGT

TTACAGAGCTCTCCTTCACCTCCTTCCAGGGGCTGATGGTGGCCATATTATACTGCTTTGTCAACA

ATGAGGTCCAGCTGGAATTTCGGAAGAGCTGGGAGCGCTGGCGGCTTGAGCACTTGCACATCCA

GAGGGACAGCAGCATGAAGCCCCTCAAGTGTCCCACCAGCAGCCTGAGCAGTGGAGCCACGGC

GGGCAGCAGCATGTACACAGCCACTTGCCAGGCCTCCTGCAGCTGA

100

OTR (homo sapiens) NM_000916.3, 1169 bp, 389 aa

ATGGAGGGCGCGCTCGCAGCCAACTGGAGCGCCGAGGCAGCCAACGCCAGCGCCGCGCCGCC

GGGGGCCGAGGGCAACCGCACCGCCGGACCCCCGCGGCGCAACGAGGCCCTGGCGCGCGTG

GAGGTGGCGGTGCTGTGTCTCATCCTGCTCCTGGCGCTGAGCGGGAACGCGTGTGTGCTGCTG

GCGCTGCGCACCACACGCCAGAAGCACTCGCGCCTCTTCTTCTTCATGAAGCACCTAAGCATCG

CCGACCTGGTGGTGGCAGTGTTTCAGGTGCTGCCGCAGTTGCTGTGGGACATCACCTTCCGCTT

CTACGGGCCCGACCTGCTGTGCCGCCTGGTCAAGTACTTGCAGGTGGTGGGCATGTTCGCCTCC

ACCTACCTGCTGCTGCTCATGTCCCTGGACCGCTGCCTGGCCATCTGCCAGCCGCTGCGCTCGC

TGCGCCGCCGCACCGACCGCCTGGCAGTGCTCGCCACGTGGCTCGGCTGCCTGGTGGCCAGC

GCGCCGCAGGTGCACATCTTCTCTCTGCGCGAGGTGGCTGACGGCGTCTTCGACTGCTGGGCC

GTCTTCATCCAGCCCTGGGGACCCAAGGCCTACATCACATGGATCACGCTAGCTGTCTACATCGT

GCCGGTCATCGTGCTCGCTGCCTGCTACGGCCTTATCAGCTTCAAGATCTGGCAGAACTTGCGG

CTCAAGACCGCTGCAGCGGCGGCGGCCGAGGCGCCAGAGGGCGCGGCGGCTGGCGATGGGG

GGCGCGTGGCCCTGGCGCGTGTCAGCAGCGTCAAGCTCATCTCCAAGGCCAAGATCCGCACGG

TCAAGATGACTTTCATCATCGTGCTGGCCTTCATCGTGTGCTGGACGCCTTTCTTCTTCGTGCAG

ATGTGGAGCGTCTGGGATGCCAACGCGCCCAAGGAAGCCTCGGCCTTCATCATCGTCATGCTCC

TGGCCAGCCTCAACAGCTGCTGCAACCCCTGGATCTACATGCTGTTCACGGGCCACCTCTTCCA

CGAACTCGTGCAGCGCTTCCTGTGCTGCTCCGCCAGCTACCTGAAGGGCAGACGCCTGGGAGA

GACGAGTGCCAGCAAAAAGAGCAACTCGTCCTCCTTTGTCCTGAGCCATCGCAGCTCCAGCCAG

AGGAGCTGCTCCCAGCCATCCACGGCGTGA

V1AR (homo sapiens) NM_000706.4, 1265 bp, 418 aa

ATGCGTCTCTCCGCCGGTCCCGACGCGGGGCCCTCGGGCAACTCCAGCCCATGGTGGCCTCTG

GCCACCGGCGCTGGCAACACAAGCCGGGAGGCCGAAGCCCTCGGGGAGGGCAACGGCCCACC

GAGGGACGTGCGCAACGAGGAGCTGGCCAAACTGGAGATCGCCGTGCTGGCGGTGACTTTCGC

GGTGGCCGTGCTGGGCAACAGCAGCGTACTGCTGGCTCTGCACCGGACGCCGCGCAAGACGTC

CCGCATGCACCTCTTCATCCGACACCTCAGCCTGGCCGACCTGGCCGTGGCATTCTTCCAGGTG

CTGCCGCAAATGTGCTGGGACATCACCTACCGCTTCCGCGGCCCCGACTGGCTGTGCCGCGTG

GTGAAGCACCTGCAGGTGTTCGGCATGTTTGCGTCGGCCTACATGCTGGTAGTCATGACAGCCG

ACCGCTACATCGCGGTGTGCCACCCGCTCAAGACTCTGCAACAGCCCGCGCGCCGCTCGCGCC

TCATGATCGCGGCCGCCTGGGTGCTGAGCTTCGTGCTGAGCACGCCGCAGTACTTCGTCTTCTC

CATGATCGAGGTGAACAATGTCACCAAGGCCCGCGACTGCTGGGCCACCTTCATCCAGCCCTGG

GGTTCTCGTGCCTACGTGACCTGGATGACGGGCGGCATCTTTGTGGCGCCCGTGGTCATCTTGG

GTACCTGCTACGGCTTCATCTGCTACAACATCTGGTGCAACGTCCGCGGGAAGACGGCGTCGCG

CCAGAGCAAGGGTGCAGAGCAAGCGGGTGTGGCCTTCCAAAAGGGGTTCCTGCTCGCACCCTG

TGTCAGCAGCGTGAAGTCCATTTCCCGGGCCAAGATCCGCACGGTGAAGATGACTTTTGTGATC

GTGACGGCTTACATCGTCTGCTGGGCGCCTTTCTTCATCATCCAGATGTGGTCTGTCTGGGATCC

CATGTCCGTCTGGACCGAATCGGAAAACCCTACCATCACCATCACTGCATTACTGGGTTCCTTGA

ATAGCTGCTGTAATCCCTGGATATACATGTTTTTTAGTGGCCATCTCCTTCAAGACTGTGTTCAAA

GCTTCCCATGCTGCCAAAACATGAAGGAAAAATTCAACAAAGAAGATACTGACAGTATGAGCAGA

AGACAGACTTTTTATTCTAACAATCGAAGCCCAACAAACAGTACGGGTATGTGGAAGGACTCGCC

TAAATCTTCCAAGTCCATCAAATTCATTCCTGTTTCAACTTGA

101

7.3 Plasmid maps

Figure S 1: Overview of all expression vectors that result in a fusion protein. A: SNAP-

GLP1R includes the signal peptide in front of the SNAP-tag. The mature peptide follows

the tag. B: rARRB2 NanoLuc (C-terminally tagged) was already cloned during my

diploma thesis and used for the BRET recruitment assays. C: GLP1 Halotag® (C-

terminally tagged), D: V1AR HaloTag® (C-terminally tagged) and E: OTR HaloTag® (C-

terminally tagged) were also used for the BRET assays. F: HiBiT-GLP1 was cloned into

the pBiT3.1-secN and a 24 aa long linker was kept between the tag and the cDNA of the

receptor.

102

7.4 Establishment of the β-arrestin2 recruitment BRET assay

Figure S 2: Establishment for the β-arrestin2 recruitment assay.

For every tested GPCR, the optimal BRET pairing was investigated by preliminary experiments.

Here, all available BRET pairings were tested to determine the pairing with the highest ligand

promoted BRET change (NET BRET). For all tested receptors, the combination of ARRB2 NL and

receptor-HT was showing the highest NET in comparison to the other combinations. The values

ranged from 15 mBU for GLP1R (A and B), to 5 mBU for OTR (C and D) and

1AR (E and F). These

combinations were then used for concentration-response assays to compare the β-arrestin2

recruitment between peptide and conjugate. Data represent one experiment for GLP1R, two

independent experiments for V1AR and three independent experiment, performed in triplicates.

Value represent mean ± SEM. NL: NanoLuc, energy donor; HT: HaloTag®, epitope tag that binds

the energy acceptor, N-terminal indicated with N-HT and C-terminal indicated with C-HT; mBU:

milliBRETUnits

103

8 Abbreviations

aa Amino acid

AC adenylate cyclase

ADHS Attention-deficit hyperactivity disorder

AHDS Allan-Herndon-Dudley syndrome

ANOVA analysis of variance

ASRT endosomal sorting actin-sorting nexin 27 (SNX)-

retromer tubule

ATP adenosine triphosphate

AVP arginine vasopressin

BAT brown adipose tissue

BBB blood-brain-barrier

BCH 2-aminobicyclo-(2,2,1)-heptane-2-carboxylic acid

BLAST basic local alignment search tool

bp base pair

BRET bioluminescence resonance energy transfer

BSA bovine serum albumin

CALB calbindin-D28k

cAMP cyclic adenosine monophosphate

cDNA complementary DNA

CH congenital hypothyroidism

CIP calf intestine phosphatase

CNS central nervous system

CO2 carbon dioxide

COS7 Cercopithecus aethiops cell line 7

CRISPR/Cas9 Clustered Regularly Interspaced Short Palindromic

Repeats/CRISPR-associated protein 9

CRYM µ-crystallin

CSF Cerebrospinal-fluid

Da Dalton

DAG diacylglycerol

DAPI 4′,6-diamidino-2-phenylindole

ddNTP dideoxynucleotide

DE Deutschland, Germany

DIO deiodinase

DIT diiodothyrosine

DITPA 3,5-diiodothyropropionic acid

DKO double knock out

DMEM dulbecco's MEM

DMI desipramine hydrochloride

DMSO dimethylsufoxide

DNA deoxyribonucleic acid

dNTP deoxynucleotide

DR dopamine receptor

DR4 direct repeat separated by four nucleotides

104

EC50 Half maximal effective concentration

ECL extracellular loops

EDTA Disodium ethylenediaminetetraacetate dihydrate

ELISA Enzyme-linked immunosorbent assay

ER estrogen receptor

ER endoplasmic reticulum

ERK1/2 extracellular signal-regulated kinase 1/2

FBS fetal bovine serum

FOXE1 forkhead-box protein E1

GAPDH Glycerinaldehyd-3-phosphat-Dehydrogenase

GCGR glucagon receptor

GDP guanine diphosphate

GH growth hormone

GLP1 Glucagon-like peptide 1

GLP1R Glucagon-like peptide 1 receptor

GNAS guanine nucleotide binding protein α subunit

GPCR G protein coupled receptor

GRK GPCR kinase

GTP guanine triphosphate

h hours

HEK293 human embryonal kidney cell line 293

HPLC high pressure liquid chromatography

HPT hypothalamus-pituitary-thyroid

HT HaloTag®

iBMEC induced brain microvascular endothelia cell

IBMX 3-Isobutyl-1-methylaxanthine

IC50 Half maximal inhibitory concentration

ICL intracellular loops

IN intranasal

IP3 inositol triphosphate

iPSC induced pluripotent stem cell

IV intravenous

KO knock out

LAT L-type amino acid transporters

LB lysogeny broth

LC/MS Liquid chromatography/mass spectrometry

LDL low-density lipoprotein

LHR luteinizing hormone receptor

M molar

MAPK mitogen-activated protein kinase

mBU milliBRET units

MCS mutiple cloning site

MCT monocarboxylate transporter

MEM minimal essential media

MFS major facilitator superfamily

min minute

MIT monoiodotyrosine

105

MPEG-PLA methoxy poly(ethylene glycol)-

poly(lactic acid)

NCBI National Center for Biotechnology Information

NEA non-essential amino acids

NEFL light neurofilament subunit

NFAT nuclear factor of activated T-cells

NIS sodium-iodine-symporter

NKX2.1 NK homeobox 1

NL NanoLuc

nm nanometer

OATP organic anion transporter polypeptide

OT oxytocin

OTR oxytocin receptor

PAX8 paired-box-protein 8

PBS phosphate buffered saline

PCR polymerase chain reaction

PDE phosphodiesterase

PEG Poly ethylene glycol

PFA paraformaldehyde

PI3K phosphatidylinositol 3-kinase

PIP2 phosphatidylinositol-4,5 bisphosphate

PKA protein kinase A

PKC protein kinase C

PLA Poly lactic acid

PLC protein lipase C

PTHR parathyroid receptor

PTX pertussis toxin

PVALB Ca2+-binding proteins parvalbumin

PVN paraventricular nucleus

RhoGEF Rho guanine-nucleotide exchange factors

RLU relative light unit

RNA ribonucleic acid

RT reverse transcriptase

rT3 reverse triiodothyronine

RXR retinoid X receptor

SLC solute carrier family

SNX sorting actin-sorting nexin 27

SON supraoptic nucleus

SRE serum response element

STAT signal transducer and activator of transcription

T2 diiodothyronine

T3 triiodothyronine

T4 thyroxine, 3,3',5,5'-Tetraiod-L-thyronin

TGB thyroxine-binding globulin

TH Thyroid hormone

THOX thyroid oxidase

TMH transmembrane helices

106

TPO thyroperoxidase

TR thyroid hormone receptor

TRE thyroid hormone response elements

TRH thyrotropin-releasing hormone

TRHR thyrotropin-releasing hormone receptor

Triac 3,5,3’-triiodothyroacetic acid

TSH thyroid-stimulating hormone

UCP1 uncoupling protein 1

USA United States of America

V1AR vasopressin 1A receptor

V1BR vasopressin 1B receptor

V2R vasopressin 2 receptor

WAT white adipose tissue

XPCT X-linked PEST-containing transporter

β2AR β2 adrenergic receptor

µl Micro liter

µM Micro molar

107

9 List of figures

Figure 1: The HPT axis is the control system of TH synthesis 2

Figure 2: Canonical and non-canonical pathways of TH 9

Figure 3: Overview of altered thyroid state of different tissues in AHDS patients 13

Figure 4: Graphical overview of GPCR signaling cascades and

internalization mechanism 19

Figure 5: MCT8-deficiency and the “Trojan Horse”-like mechanism 22

Figure 6: Schematical depiction of the workflow for this project 31

Figure 7: Overview of in vitro assay and the expression vectors used for

cloning for the specific assays 32

Figure 8: Schematical depiction of the AlphaScreen™ assay used for

determination of intracellular cAMP concentration 40

Figure 9: Concept of a bioluminescence resonance energy transfer (BRET)

assay 42

Figure 10: Graphical depiction of the HiBiT Cell surface expression assay 44

Figure 11: The general structure of the results part 46

Figure 12: GLP1-T3 is able to activate GLP1R in the same manner as GLP1 47

Figure 13: GLP1-T3 is able to initiate GLP1R internalization in a similar

fashion to GLP1 48

Figure 14: Functional studies on the OTR showed biased agonism of OT-T3 50

Figure 15: Functional characterization on the V1AR showed biased agonism

of OT-T3 50

Figure 16: Stimulation of OTR and V1AR with OT and OT-T3 resulted in

different β-arrestin2 recruitment pattern 51

Figure 17: Preliminary assays to determine a suitable blocker for endogenous

expressed TH transporter 54

Figure 18: Luciferase based reporter gene studies of nuclear response

to TH (canonical pathway) in a time- and concentration-dependent manner 55

Figure 19: Luciferase based reporter gene studies of non-canonical response

to TH in a time -dependent manner 56

Figure 20: Luciferase based reporter gene studies of canonical response

to TH in a time -dependent manner 57

Figure 21: Graphical depiction of the results for canonical and

non-canonical TH signaling assays of GLP1-T3 58

Figure 22: Graphical depiction of the results for canonical TH signaling of

OT-T3 58

Figure 23: Graphical overview of the considerations regarding the perfect

receptor-ligand pair as well as an outline of the aspects investigated in the course of

this project 59

Figure 24: Depiction of the signaling pathways at the OTR and the biased

signaling caused by PTX and OT-T3 63

Figure 25: Graphical abstract for the first investigated conjugate GLP1-T3 69

Figure 26: Graphical abstract for the second investigated conjugate OT-T3 70

Figure 27: Structural homology model of OTR 72

Figure S 1: Overview of all expression vectors that result in a fusion protein 101

Figure S 2: Establishment for the β-arrestin2 recruitment assay 102

108

10 List of tables

Table 1: Summarizes effects of THs on their target tissues 4

Table 2: Overview of symptoms of TR deficiencies 7

Table 3: Substrate specificity of TH transporters, numbers of + are indicating

the degree of affinity 11

Table 4: Machines used and their corresponding supplier company 24

Table 5: Supplier companies for consumables (A) and chemicals (B) 25

Table 6: Overview of the commercially available kits that were used in the course

of this project 25

Table 7: Buffers and reagents commercially purchased or prepared in the lab 26

Table 8: Stimulation agents used in this project 27

Table 9: Commercially available enzymes used for this project. 27

Table 10: Expression vectors utilized in the course of this project; * marks 28

Table 11: Sources for cDNA of proteins of interest for this project 28

Table 12: Fluorescent dyes and DNA markers implemented for this project 28

Table 13: Overview of cell cultured related reagents and media for this project 29

Table 14: Composition for complete media for the different cell lines used in

the course of this project 29

Table 15: Composition for bacterial culture reagents and media implemented for

this project 30

Table 16: Bacterial strains and eukaryotic cell lines used in the course of this

project 30

Table 17: Computer software used for this project 30

Table 18: PCR preparation and cycler program 34

Table 19: Site-directed mutagenesis PCR preparation and cycler program 34

Table 20: Sequencing PCR preparation and PCR program 35

Table 21: Restriction digestion preparation for expression vector and insert 36

Table 22: Ligation preparation 36

Table 23: Overview of the pathways investigated with luciferase-based reporter

gene assays 41

Table 24: Summary of the results for the functional characterization of GLP1-T3

at the GLP1R 52

Table 25: Summary of the results for the functional characterization of OT-T3 at

the OTR and V1AR 52

Table S 1 List of all cloning primers used for this project 98