The calcineurin inhibitor cyclosporine A activates the renal Na-(K)-Cl cotransporters via local and systemic mechanisms [original]

The Calcineurin Inhibitor Cyclosporine A Activates the Renal Na-(K)-Cl

Cotransporters via Local and Systemic Mechanisms

vorgelegt von

Dipl.-Ing.

Katharina Ilse Blankenstein

von der Fakultät III – Prozesswissenschaften

der Technischen Universität Berlin

zur Erlangung des akademischen Grades

Doktor der Ingenieurwissenschaften

– Dr.-Ing. –

genehmigte Dissertation

Promotionsausschuss:

Vorsitzender: Prof. Dr. Lorenz Adrian

Erster Gutachter: Prof. Dr. Sebastian Bachmann

Zweiter Gutachter: Prof. Dr. Roland Lauster

Dritte Gutachterin: Prof. Dr. Vera Meyer

Tag der wissenschaftlichen Aussprache: 12. Juli 2019

Berlin 2019

Statement on the originality of the data

Herewith I declare that I prepared the Ph.D. thesis ‘The Calcineurin Inhibitor Cy-

closporine A Activates the Renal Na-(K)-Cl Cotransporters via Local and

Systemic Mechanisms’ on my own and with no other sources and aids than quoted.

Abstract

The Calcineurin Inhibitor Cyclosporine A Activates the Renal

Na-(K)-Cl Cotransporters via Local and Systemic Mechanisms

The two calcineurin inhibitors (CNIs) cyclosporine A (CsA) and tacrolimus are potent

immunosuppressive drugs that have become indispensable in the post-transplantational

therapy and in the treatment of autoimmune diseases. Immunosuppression via CNIs is

achieved by inhibition of the calcineurin phosphatase and thus blocking the signaling

pathway of NFAT (nuclear factor of activated T cells), thereby inhibiting T cell immune

response. However, chronic use of CNIs is complicated by serious side effects which par-

ticularly affect the kidney, reflected by a large number of patients who develop electrolyte

disorders, hypertension and renal failure under CNI treatment. The mechanisms of CNI-

induced salt retention and hypertension are complex, involving not only intraepithelial

but also systemic effects. Since calcineurin inhibition is associated with activation of the

renal Na+-K+-2Cl-cotransporter (NKCC2) and the Na+-Cl-cotransporter (NCC) that

are expressed in the thick ascending limb (TAL) and the distal convoluted tubule (DCT)

of the nephron, respectively, it has been suggested that the calcineurin phosphatase is

involved in the regulation of the salt transporters. Interestingly, previous studies have

demonstrated that CsA increases both NKCC2 and NCC activity, whereas tacrolimus

merely stimulates NCC function. This study aimed to unravel the local and systemic

mechanisms underlying calcineurin inhibition that lead to salt retention and hypertension.

We found that key proteins involved in calcineurin inhibition and the activation of NKCC2

and NCC are ubiquitously expressed along TAL and DCT. These results indicate that

the distinct effects of CsA and tacrolimus on the activation of the salt transporters do

not arise from differential expression of these key proteins, so that systemic mechanims

might play a role. Thus, we sought to differentiate between local and systemic effects

of CsA treatment by applying short- and long-term treatment protocols in vivo and in

vitro. We analyzed physiological changes induced by CsA treatment and studied the

role of endocrine factors such as the hormone arginine vasopressin (AVP). We show that

CsA-induced activation of NKCC2 and NCC and their activating kinases chiefly occurs

via post-translational modification by phosphorylation/dephosphorylation reactions. We

further found that CsA-induced activation of NKCC2, but not NCC, requires AVP sig-

naling. In a pilot study on NCC knockout mice we show that CsA treatment, unlike

tacrolimus, might induce high blood pressure regardless of NCC activity. These findings

provide evidence for the major role of NKCC2 in CsA-induced hypertension.

In summary, the results of this study demonstrate that CsA-induced activation of NKCC2

and NCC and their activating kinases chiefly occurs at the post-translational level via in-

creased phosphorylation. In DCT cells, local calcineurin inhibition is sufficient to induce

NCC activation, whereas NKCC2 activation appears to require additional stimulation by

AVP. Our data further suggest a pivotal role of NKCC2 in CsA-induced hypertension.

Altogether, this study contributes to a better understanding of CNI-induced salt retention

and hypertension and can help improve blood pressure control.

Zusammenfassung

Der Calcineurin-Inhibitor Cyclosporin A Aktiviert die Renalen

Na-(K)-Cl-Kotransporter mittels Lokaler und Systemischer

Mechanismen

Die Calcineurin-Inhibitoren (CNI) Cyclosporin A (CsA) und Tacrolimus sind hochwirk-

same immunsuppressive Substanzen, die in der Posttransplantationstherapie und in der

Behandlung verschiedener Autoimmunerkrankungen unverzichtbar geworden sind. Die

durch CNI ausgel¨oste Immunsuppression wird ¨uber die Inhibition der Calcineurinphos-

phatase vermittelt. Dies f¨uhrt zur Blockade des NFAT (nuclear factor of activated T

cells)-Signalwegs und somit zur Inhibierung der T-Zell-Immunantwort. Allerdings wird

die chronische CNI-Gabe durch ernsthafte Nebenwirkungen erschwert, die insbesondere

die Niere betreffen. Dies spiegelt sich in der großen Anzahl an Patienten wieder, die unter

CNI-Behandlung St¨orungen im Elektrolythaushalt, Bluthochdruck und Nierenversagen

entwickeln. Die Mechanismen, die zu CNI-induzierter Salzretention und Bluthochdruck

f¨uhren, sind komplex und umfassen neben intraepithelialen auch systemische Effekte.

Weil die Calcineurin-Inhibition mit einer Aktivierung der renalen Na+-K+-2Cl-- (NKCC2)

und Na+-Cl--Kotransporter (NCC) in Zusammenhang steht, die jeweils in der dicken

aufsteigenden Henle-Schleife (TAL) und dem distalen Konvolut (DCT) des Nephrons

exprimiert sind, vermutet man, dass die Calcineurinphosphatase in die Regulation der

Salztransporter involviert ist. Interessanterweise haben fr¨uhere Studien gezeigt, dass

CsA die Aktivit¨at von NKCC2 und NCC steigert, w¨ahrend Tacrolimus nur die NCC-

Funktion stimuliert. Die vorliegende Arbeit hatte daher zum Ziel, die lokalen und sys-

temischen Mechanismen der Calcineurin-Inhibition aufzudecken, die zu Salzretention und

Bluthochdruck f¨uhren.

Die von uns untersuchten Schl¨usselproteine, die in die Calcineurin-Inhibition und die Ak-

tivierung von NKCC2 und NCC involviert sind, zeigten sich entlang dem TAL und dem

DCT ubiquit¨ar exprimiert. Diese Ergebnisse wiesen darauf hin, dass die unterschiedliche

Wirkung von CsA und Tacrolimus hinsichtlich der Aktivierung der Salztransporter nicht

auf einer differentiellen Expression dieser Schl¨usselproteine beruht. Stattdessen k¨onnten

systemische Mechanismen eine Rolle spielen. Deshalb untersuchten wir im n¨achsten

Schritt lokale und systemische Effekte der CsA-Behandlung mittels Kurz- und Langzeit-

behandlungen am in vivo- und in vitro-Modell. Hierbei ermittelten wir die durch CsA

induzierten physiologischen ¨

Anderungen und untersuchten die Rolle endokriner Faktoren

wie dem Hormon Arginin-Vasopressin (AVP). Wir konnten zeigen, dass die CsA-induzierte

Aktivierung von NKCC2 und NCC sowie ihrer regulierenden Kinasen haupts¨achlich auf

Zusammenfassung vi

posttranslationaler Modifizierung via Phosphorylierungs-/Dephosphorylierungsreaktionen

beruht. Außerdem zeigen wir, dass f¨ur die CsA-induzierte Aktivierung von NKCC2 die

Stimulation durch AVP notwendig ist, w¨ahrend f¨ur die Aktivierung von NCC die lokale

Calcineurin-Inhibierung ausreichend ist. In einer Pilotstudie an NCC-Knockout-M¨ausen

liefern wir zudem Hinweise daf¨ur, dass CsA, im Gegensatz zu Tacrolimus, trotz Abwesen-

heit von NCC zu Bluthochdruck f¨uhrt. Diese Ergebnisse deuten auf eine entscheidende

Rolle von NKCC2 in der Entstehung von CsA-induziertem Bluthochdruck hin.

Zusammengefasst demonstriert die vorliegende Arbeit, dass die CsA-induzierte Aktivierung

von NKCC2 und NCC sowie ihrer regulierenden Kinasen haupts¨achlich auf der post-

translationalen Ebene ¨uber eine Hochregulation der Phosphorylierung abl¨auft. In DCT-

Zellen gen¨ugt f¨ur die Aktivierung von NCC die lokale Calcineurin-Inhibition, w¨ahrend

die NKCC2-Aktivierung in TAL-Zellen einer zus¨atzlichen Stimulation durch AVP be-

darf. Unsere Ergebnisse lassen auf eine Schl¨usselrolle von NKCC2 bei CsA-induziertem

Bluthochdruck schließen. Insgesamt tragen diese Daten zu einem besseren Verst¨andnis

von CsA-induzierten St¨orungen des Elektrolythaushalts und Hypertonie bei und k¨onnen

einen Beitrag zur Verbesserung der Behandlung von Bluthochdruck liefern.

Acknowledgements

I would like to thank Prof. Dr. Sebastian Bachmann for giving me the opportunity to

work in his department and for supporting me at all stages of this thesis.

I am grateful to PD Dr. Kerim Mutig for his support, his feedback and inspiring ideas on

our work.

I am also very thankful to Prof. Dr. Roland Lauster, my second supervisor, for his

mentoring throughout my academic career and for broadening the research perspective.

My sincere thanks goes to Dr. David Ellison for the time working in his group, a fruitful

collaboration and always exciting discussions.

Additionally, I want to thank all the members of the Sebastian Bachmann laboratory for

their support and the positive working environment but also for the great time we had

outside the lab.

Finally, my profound gratitude goes to my family and friends who constantly support and

encourage me.

vii

Contents

Promotionsausschuss i

Statement on the originality of the data ii

Abstract iii

Zusammenfassung v

Acknowledgements vii

List of Figures xi

List of Tables xii

Abbreviations xiii

1 Introduction 1

1.1 Calcineurin Inhibitors - Dealing with Hypertension . . . . . . . . . . . . . 1

1.2 Kidney Morphology and Functions . . . . . . . . . . . . . . . . . . . . . . 4

1.2.1 The Distal Tubule – Key Role in Volume Regulation . . . . . . . . 6

1.3 Expression and Regulation of NKCC2 and NCC . . . . . . . . . . . . . . 8

1.3.1 Gene and Protein Expression . . . . . . . . . . . . . . . . . . . . . 8

1.3.2 Trafficking and Phosphorylation . . . . . . . . . . . . . . . . . . . 10

1.3.3 Activation by the WNK-SPAK/OSR1 Kinase Cascade . . . . . . . 11

1.3.4 Endocrine Control of NKCC2 and NCC . . . . . . . . . . . . . . . 13

1.3.5 Pathophysiology of NKCC2 and NCC . . . . . . . . . . . . . . . . 14

1.4 Effects of Cyclosporine and Tacrolimus on NKCC2 and NCC Function . . 15

1.4.1 Calcineurin ............................... 16

1.4.2 Calcineurin Inhibition . . . . . . . . . . . . . . . . . . . . . . . . . 17

2 Hypotheses, Aims and Study Design 20

3 Material and Methods 22

3.1 Animals, Tissues, Treatments . . . . . . . . . . . . . . . . . . . . . . . . . 22

3.2 Perfusion Fixation and Tissue Embedding . . . . . . . . . . . . . . . . . . 24

3.3 CellCulture................................... 24

viii

3.4 Antibodies.................................... 25

3.5 Immunofluorescence .............................. 25

3.6 Immunoblotting................................. 26

3.7 QuantitativePCR ............................... 26

3.8 InSituHybridization ............................. 27

3.9 Ultrastructural Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

3.10Genotyping ................................... 27

3.11 Blood Pressure Measurements . . . . . . . . . . . . . . . . . . . . . . . . . 28

3.12Statistics .................................... 28

4 Results – Part I: Key Factors Involved in Local Calcineurin Inhibition 29

4.1 Localization of Calcineurin Isoforms . . . . . . . . . . . . . . . . . . . . . 30

4.2 Localization of Immunophilins . . . . . . . . . . . . . . . . . . . . . . . . . 31

4.3 Localization of KLHL3 and CUL3 . . . . . . . . . . . . . . . . . . . . . . 33

5 Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 35

5.1 CsA Treatment Activates NKCC2 and NCC on the Post-Translational Level 35

5.1.1 Acute CsA Treatment Induces NKCC2 and NCC Activation in vivo 36

5.1.2 Acute CsA Treatment Activates the WNK-SPAK/OSR1 Cascade

in vivo .................................. 39

5.1.3 Acute CsA Treatment Does not Affect mRNA of the WNK-SPAK/OSR1

Cascade................................. 42

5.1.4 Chronic CsA Treatment Induces NKCC2 and NCC Activation in vivo 43

5.1.5 Chronic CsA Treatment Activates SPAK/OSR1 in vivo ...... 45

5.1.6 Chronic CsA Treatment Does not Affect mRNA of the WNK-SPAK/OSR1

Cascade................................. 46

5.2 Effects of CsA on Endocrine and Paracrine Regulation of NKCC2 and NCC 47

5.3 CsA-Induced Salt Retention and Hypertension . . . . . . . . . . . . . . . 50

5.4 CsA-Induced Activation of NKCC2 Depends on AVP . . . . . . . . . . . . 52

5.4.1 CsA Stimulates NCC but not NKCC2 in AVP-Deficient Brattleboro

Rats................................... 52

5.4.2 AVP is Required for CsA-Induced Activation of NKCC2 in vitro . 52

6 Results – Part III: Key Role of NKCC2 in Blood Pressure Regulation 55

6.1 Effects of CsA Treatment in NCC Knockout Mice - A Pilot Study . . . . 56

6.2 Stimulated Salt Reabsorption Cascade in the TAL of NCC Knockout Mice 58

7 Discussion 63

8 Conclusion 71

9 Perspectives 72

A Supplementary Tables 74

A.1 PrimaryAntibodies .............................. 75

A.2 SecondaryAntibodies ............................. 76

A.3 qPCRPrimers ................................. 77

A.4 Primers for In Situ Hybridization . . . . . . . . . . . . . . . . . . . . . . . 77

References 78

List of Figures

1.1 Immunosuppression in T Cells . . . . . . . . . . . . . . . . . . . . . . . . . 3

1.2 Structure of the Nephron . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

1.3 Schematic of NKCC2 and NCC . . . . . . . . . . . . . . . . . . . . . . . . 9

1.4 Activation by the WNK-SPAK/OSR1 Kinase Cascade . . . . . . . . . . . 13

4.1 mRNA Expression of Calcineurin Isoforms . . . . . . . . . . . . . . . . . . 30

4.2 mRNA Expression of Immunophilins - qPCR . . . . . . . . . . . . . . . . 31

4.3 mRNA Expression of Immunophilins -In Situ Hybridization . . . . . . . . 32

4.4 mRNA Expression of KLHL3 and CUL3 . . . . . . . . . . . . . . . . . . . 33

4.5 Immunofluorescence of KLHL3 and CUL3 . . . . . . . . . . . . . . . . . . 34

5.1 Short-Term CsA Effects on Phosphorylation of Distal Salt Transporters . 37

5.2 Short-Term CsA Effects on Trafficking of NKCC2 and NCC . . . . . . . . 38

5.3 Short-Term CsA Effects on SPAK Phosphorylation . . . . . . . . . . . . . 40

5.4 Short-Term CsA Effects on Total SPAK/OSR1 Abundance . . . . . . . . 41

5.5 Short-Term CsA Effects on WNK1 Phosphorylation . . . . . . . . . . . . 42

5.6 Short-Term CsA Effects on WNK-SPAK/OSR1-NKCC2/NCC mRNA . . 43

5.7 Long-Term CsA Effects on NKCC2 and NCC Phosphorylation . . . . . . 44

5.8 Long-Term CsA Effects on SPAK Phosphorylation . . . . . . . . . . . . . 46

5.9 Long-Term CsA Effects on WNK-SPAK/OSR1-NKCC2/NCC mRNA . . 47

5.10 Short- and Long-Term CsA Effects on COX-2 mRNA Expression . . . . . 48

5.11 Regulation of Endocrine Factors by Long-Term CsA Treatment . . . . . . 49

5.12 Long-Term CsA Stimualtes Plasma Renin Activity . . . . . . . . . . . . . 50

5.13 CsA-Induced Salt Retention and Hypertension . . . . . . . . . . . . . . . 51

5.14FurosemideTest ................................ 51

5.15 Short-Term CsA Effects in Brattleboro Rats . . . . . . . . . . . . . . . . . 52

5.16 CsA Activates NCC but not NKCC2 in vitro ................ 53

5.17 DDAVP Stimulates NKCC2 and SPAK/OSR1 in Cultured TAL Cells . . 54

6.1 Blood Pressure in NCC KO Mice . . . . . . . . . . . . . . . . . . . . . . . 56

6.2 Effects of CsA Treatment on pNKCC2 in NCC KO Mice . . . . . . . . . . 58

6.3 NKCC2 Protein and Phosphorylation Levels in NCC KO Mice . . . . . . 59

6.4 SPAK Protein and Phosphorylation Levels in NCC KO Mice . . . . . . . 60

6.5 Expression Profile of Regulatory pS-SPAK/OSR1 in NCC KO Mice . . . 61

6.6 Expression Profile of Catalytic pT-SPAK/OSR1 in NCC KO Mice . . . . 62

List of Tables

A.1 PrimaryAntibodies .............................. 75

A.2 SecondaryAntibodies ............................. 76

A.3 qPCRPrimers ................................. 77

A.4 insituPrimers ................................. 77

xii

Abbreviations

ATPase Adenosine Triphosphatase

AVP Arginine VasoPressin

BW BodyWeight

cAMP Cyclic Adenosine Monophosphate

CCT Conserved C-Terminal

CD Collecting Duct

cDNA complementary Desoxyribonucleic Acid

CnA Calcineurin Catalytic Subunit A

CnB Calcineurin Regulatory Subunit B

CNI Calcineurin Inhibitor

CNT Connecting Tubule

COX-2 Cyclooxygenase 2

CsA Cyclosporine A

CUL3 Cullin-3

Cy Cyanine

CypA Cyclophilin A

CypB Cyclophilin B

DCT Distal Convoluted Tubule

DDAVP 1-Desamino-8-D-Arginine Vasopressin

DIG Digoxygenin

ENaC Epithelial Natrium Channel

ER Endoplasmatic Reticulum

FHHt Familial Hyperkalemic Hypertension

Fkbp12 FK506-binding protein-12

xiii

xiv

GFR Glomerular Filtration Rate

KLHL3 Kelch-Like Protein 3

KO Knockout

mDCT mouse Distal Convoluted Tubule

mRNA messenger Ribonucleic Acid

NaCl Natrium Chloride

NCC Natrium Chloride Cotransporter

NFAT Nuclear Factor of Activated TCells

NKCC2 Natrium Kalium Chloride Cotransporter 2

OSR1 Oxidative Stress-Response Kinase 1

PBS Phosphate Buffered Saline

PKA Protein Kinase A

qPCR Quantitative Polymerase Chain Reaction

RAAS Renin Angiotensin Aldosterone System

raTAL rat medullary Thick Ascending Limb

RNA Ribonucleic Acid

RT-PCR Reverse Transcription-Polymerase Chain Reaction

SLC Solute Carrier

SPAK Ste20-related Proline Alanine-rich Kinase

TAL Thick Ascending Limb

TCR T Cell Receptor

WNK With No Lysine Kinase

WT Wildtype

Chapter 1

Introduction

1.1 Calcineurin Inhibitors - Dealing with Hypertension

Immunosuppressants have become indispensable in the post-transplantational therapy to

prevent acute and chronic graft rejection and improve long-term survival of the patients.

They are also widely prescribed for the treatment of several autoimmune diseases such

as rheumatoid arthritis, psoriasis and atopic dermatitis. One of the most commonly used

class of immunosuppressive drugs are the two calcineurin inhibitors (CNIs) cyclosporine

A (CsA) and tacrolimus. Especially after solid organ transplantation, CNIs provide a

powerful tool for reducing morbidity and mortality ([1], [2], [3]).

The great treatment success achieved with CNIs since their market entry more than 20

years ago resulted in increasing and longer administration of the drugs. However, chronic

CNI treatment is complicated by serious side effects. One of the most severely affected or-

gans is the kidney; virtually all patients suffer from renal toxicity and some develop renal

failure ([4], [5], [6]). Elevated blood pressure and electrolyte disorders are typical early

symptoms after initiation of CNI therapy ([7], [8]). Since the pathogenetic mechanims of

CNI side effects are complex, search for optimal treatment options has been challenging

so far. Diuretics are commonly used to reduce sodium and water retention in order to

control hypertension, but diuretic resistance or adverse effects such as potassium wasting

complicate the anti-hypertensive treatment ([9], [10]).

Introduction 2

CNIs suppress the immune response of the body via inhibiting the calcineurin signal-

ing pathway inside T lymphocytes. Immune response in a T cell is induced by antigen

binding to its T cell receptor. This increases intracellular Ca2+ levels and activates the

Ca2+-calmodulin-dependent calcineurin phosphatase. Calcineurin subsequently dephos-

phorylates the transcriptional factor NFAT (nuclear factor of activated T cells) promoting

its nuclear translocation. Transcriptionally active NFAT then induces the expression of a

large number of immunomodulatory genes encoding cytokines such as interleukin 2 and

interleukin 4 (Figure 1.1). Inhibition of this pathway using CNIs successfully suppresses

the T cell immune response ([11], [12], [13], [14]).

Introduction 3

Figure 1.1. Mechanism of Immunosuppression in T Lymphocytes via Cyclosporine A and

Tacrolimus Activation of a T cell through antigen binding of its T cell receptor (TCR) leads to increased

intracellular Ca2+ levels and activation of the Ca2+-calmodulin-dependent calcineurin phosphatase. NFAT

(nuclear factor of activated T cells) is activated through dephosphorylation by calcineurin followed by

nuclear translocation and induction of target gene expression such as IL-2 and IL-4. Cyclosporine A (CsA)

and tacrolimus inhibit this pathway and thereby suppress T cell immune response. ER: Endoplasmatic

Reticulum.

Recent research has focused on the pathogenesis of the renal adverse effects of CNIs;

however, the precise molecular mechanisms remain to be completely understood. Nu-

merous studies indicate that both intrinsic and extrinsic signals are responsible for the

CNI-induced development of hypertension. Local mechanisms are reportedly based on

activation of the sodium chloride cotransporters in the kidney distal nephron ([15], [16],

[17], [18]). Additionally, endocrine signals such as arginine vasopressin (AVP) or the

Introduction 4

renin-angiotensin-aldosterone-system (RAAS) may exacerbate the side effects of CNIs

([19], [20], [21]).

As chronically elevated blood pressure is a severe risk factor for the development of sev-

eral cardiovascular conditions and ultimately leads to kidney failure, the clinical need for

effective antihypertensive therapy has immensely increased in the last decades. Thus,

in order to improve blood pressure control and prevent kidney damage, unraveling the

mechanisms of CNI-induced hypertension has enormous clinical value ([22], [23], [24]).

1.2 Kidney Morphology and Functions

The kidney plays a key role in blood pressure control. It is responsible for filtration of the

entire blood volume of the body. The filtration of blood in the kidney produces a large

amount of primary urine. The subsequent reabsorption process in the renal tubules serves

to preserve water, electrolytes and other useful components, whereas toxic substances are

excreted with the final urine ([25]). By precise regulation of water and electrolyte home-

ostasis, the kidney is able to keep the renal blood filtration rate constant over a wide

range of changes in perfusion pressure ([26], [27], [28]). This is achieved by its sophisti-

cated morphology ([29]).

The parenchyma of the kidney is divided into the renal cortex and the renal medulla. The

basic functional units of the kidney, the nephrons, span the cortex and the medulla. They

are composed of the renal corpuscle, which is the initial filtering portion, and a segmented

tubule ([30], [31], [32]). There are two sorts of a nephrons with different lenghts: the jux-

tamedullary and the cortical nephron with the latter possessing a longer loop of Henle and

the hairpin reaching to the inner zone of the medulla. A renal tubule is composed of the

following segments: The proximal tubule, the thin descending limb, the thin ascending

limb (only in the long looped nephron) and the thick ascending limb of the loop of Henle,

the distal convoluted tubule and the connecting tubule which empties into the collecting

duct together with approximately eleven other nephrons ([31], [33]) (Figure 1.2).

The gradual buildup of urine concentration is facilitated by the medullary architecture:

Descending and ascending tubules with varying water permeability are surrounded by

Introduction 5

straight arterioles. Substance specific transport across the single-layered epithelia of the

tubule system is realized by channels, transporters and pumps located at the luminal or

basolateral site of the polarized cells ([29]).

Figure 1.2. Structure of the Nephron Shown is the schematic of a juxtamedullary and a cortical

nephron. Juxtamedullary nephrons have a longer loop of Henle and the hairpin reaches to the inner zone

of the medulla. They are responsible for the development of the osmotic gradient that concentrates the

urine. 1: glomerulus, 2: proximal convoluted tubule, 3: proximal straight tubule, 4: thin descending

limb, 5: thin ascending limb, 6: the thick ascending limb of the loop of Henle, 7: macula densa, 8: distal

convoluted tubule, 9: connecting tubule, 10-12: collecting duct. Schematic from Kriz and Bankir ([29]).

The renal corpuscule consists of the glomerulus and the Bowman’s capsule. At the

glomerulus, blood enters via an afferent arteriole and leaves through an efferent arte-

riole ((Figure 1.2). Through the glomerular blood pressure in the capillaries within the

Introduction 6

glomerular network, water and solutes are filtered out of the blood. The renal filtra-

tion barrier is composed of the capillary endothelium, a basement membrane, and adja-

cent podocytes, which build a slit diaphragm that has the ability to discriminate among

molecules with different size and electrical charge ([25], [27], [29], [28]).

Solutes up to an effective radius of 1.8nm can freely pass the barrier while passage for

larger molecules becomes progressively difficult. Blood cells or the majority of proteins will

not enter the Bowman’s space but remain in the plasma. Filtration of anionic compounds

is more difficult due to the anionic components within the filtration barrier. Water and

small solutes such as glucose, amino acids, small proteins and electrolytes are filtered and

transferred into the renal tubule system ([25], [27]).

Around 180 lprimary urine per day are produced through glomerular filtration. 99 %

of the water and a large amount of the solutes are reabsorbed along the renal tubule,

creating a highly concentrated secondary urine. A final volume of approximately 1.5lare

excreted per day ([34], [35]).

1.2.1 The Distal Tubule – Key Role in Volume Regulation

To ensure renal excretion, glomerular filtration rate (GFR) has to be kept constant even

at fluctuations of systemic blood pressure. For this, intra- and extrarenal mechanisms

exist to effectively regulate fluid and electrolyte homeostasis. The distal tubule which

consists of the thick ascending limb (TAL) and the distal convoluted tubule (DCT) herein

plays a substantial role. Distal tubule cells possess a high density of Na+-K+-ATPase as

well as specific salt transporters which are target of widely used diuretic drugs ([36], [34]).

Another characteristic of TAL and DCT is their impermeability for water. Together, this

allows for effective separation of salt and water in the distal tubule ([37]).

The Thick Ascending Limb

The TAL reabsorbs around 25 % of the filtered sodium chloride (NaCl). The active NaCl

reabsorption in this water-impermeable segment results in a hypoosmolar urine. For this

reason, the TAL is also termed the ”diluting segment” ([37], [36]). The key component for

Introduction 7

sodium reabsorption in the TAL is the apical Na+-K+-2Cl-cotransporter (NKCC2). Fol-

lowing the electrochemical gradient that is generated by the basolateral Na+-K+-ATPase,

NKCC2 transports Na+, K+and 2Cl-electroneutrally into the cytoplasm ([36], [35]).

NaCl uptake via NKCC2 is particularly important in specialized macula densa cells which

are located in the late portion of the TAL. The macula densa is an area of closely packed

cells lying in direct contact with the vascular pole of the renal corpuscule ([38]). They

act as a NaCl sensor and are responsible for two mechanisms: the tubuloglomerular

feedback (TGF) and the activation of the renin-angiotensin-aldosterone system (RAAS).

As a consequence of these mechanisms, the tone of afferent and efferent arterioles can be

adjusted to keep renal filtration at a constant level ([39], [40]). In case of increased renal

filtration through increased systemic blood pressure, NaCl delivery to the macula densa

is enhanced. This stimulates macula densa cells to secrete adenosine, the mediator of the

TGF. Adenosine causes vasoconstriction of the afferent arteriole of the glomerulus which

finally reduces the GFR ([41], [42]).

In case of reduced GFR induced by low systemic blood pressure, reduced chloride delivery

to the macula densa stimulates the secretion of renin by specialized smooth muscle cells

in the wall of afferent arterioles, so called juxtaglomerular cells ([43], [44]). Renin secre-

tion initiates a cascade that ultimately forms angiotensin II. This hormone has strong

vasoconstrictive effects, systemically as well as intrinsically with a potential preference on

the efferent arterioles in the kidney, which consequently increases the GFR ([45]). Be-

sides, angiotensin II stimulates the secretion of aldosterone and AVP. Release of these

two hormones results in elevated sodium uptake with water following through the tubular

epithelium. This in turn increases blood volume and subsequently raises the systemic

blood pressure ([46], [47]). In contrast, when renal filtration is too high, the macula densa

downregulates the GFR by restricting the RAAS in order to avoid glomerular damage.

Together, the tubuloglomerular feedback and the RAAS are pivotal mechanisms regulated

directly and indirectly by the macula densa to maintain blood pressure and renal filtration

via intrinsic and extrinsic mechanisms ([48]).

Introduction 8

The Distal Convoluted Tubule

Approximately 5 - 10 % of the filtered sodium is reabsorbed in the DCT ([34]). The DCT

is functionally divided into an early (DCT1) and a late (DCT2) segment. Electroneutral

transport of sodium and chloride across the apical membrane of the DCT is driven by the

gradient created by the Na+-K+-ATPase. DCT cells are characterized by the Na+-Cl-

cotransporter (NCC) ([49], [50]). NCC is expressed in the entire DCT, but in the late

DCT2 its expression overlaps with the epithelial sodium channel ENaC. However, NCC

is chiefly responsible for the sodium uptake in the DCT ([51], [52]). The DCT2 builds a

portion of the aldosterone-sensitive distal nephron which includes the connecting tubule

and the collecting duct. Different studies have shown that angiotensin II and aldosterone

stimulate NCC activity ([53], [54]), and AVP also plays a role in stimulation of NCC

activity ([55]). Due to its ability to adapt to changes in hormonal stimuli and precisely

adjust sodium balance, the DCT is a critical nephron segment for the salt and water

homeostasis and volume regulation.

Since the distal tubule plays a key role in blood pressure regulation understanding the

mechanisms of how CNIs activate NKCC2 and NCC will help improve the treatment of

CNI-induced hypertension.

1.3 Expression and Regulation of NKCC2 and NCC

1.3.1 Gene and Protein Expression

NKCC2 and NCC belong to the group of solute carriers (SLC) and are members of the

SLC12 family of electroneutral cation-chloride-coupled cotransporters. The SLC12A1

gene encodes NKCC2 and the SLC12A3 gene encodes NCC ([56], [50]). The basic pro-

tein structure of NKCC2 and NCC is of high similarity, comprising 12 transmembrane-

spanning domains, flanked by a short cytoplasmic N-terminal and a large cytoplasmic

C-terminal domain, and segment 7 and 8 are connected by a long hydrophilic extracellu-

lar loop ([57]) (Figure 1.3).

Introduction 9

Figure 1.3. Schematic Structure of the Highly Homologous Distal Salt Transporters NKCC2

and NCC Basic structure of the NKCC2 and NCC consisting of 12 transmembrane-spanning domains,

a short cytoplasmic N-terminal (NH2) and a large cytoplasmic C-terminal (COOH) domain. Segment 7

and 8 are connected by a long hydrophilic extracellular loop. Most of the known conserved regulatory

phosphorylation (P) sites are located within the N-terminal domain. Schematic modified from Ares et al.

([57])

Both SLC12A1 and SLC12A3 give rise to different isoforms of NKCC2 and NCC, re-

spectively. Alternative splicing of exon 4 of the SLC12A1 gene yields three full length

transcripts of 1,099 amino acids (SLC12A1a, b and f) that differ in their distribution

along the TAL and transport affinity for Na+, K+and Cl-([58], [59]). In rodent kid-

ney, additional alternative splicing leads to the generation of three NKCC2 isoforms with

truncated C-terminal ends. Some studies suggest that the various NKCC2 isoforms coop-

erate in salt reabsorption in the TAL, since different salt concentrations can be detected

in this part of the nephron ([60], [61]). The predicted molecular mass of NKCC2 is 121

kDa; however, immunoblotting of NKCC2 shows an apparent molecular mass of 160 kDa,

conferred by two N-glycosylations at positions between transmembrane domain 7 and 8.

This glycosylation is likely involved in maturation and membrane targeting of NKCC2

([58], [62]).

The NCC gene gives rise to three alternative splice products, all of a size around 1,020

- 1,030 amino acids, of which isoform 3 is the dominant form in the human kidney and

Introduction 10

isoforms 1 and 2 are lacking in rodents ([63], [64]). Like NKCC2, NCC possesses two

N-glycosylations, located in the 4th extracellular loop, and they have been shown to be

essential for efficient function and surface expression. The non-glycosylated protein has a

molecular mass of 113 kDa, but the complexly glycosylated protein shows a broad band

between 125 and 160 kDa when detected by immunoblotting ([65], [66], [67]).

1.3.2 Trafficking and Phosphorylation

Function of most transmembrane proteins is regulated by trafficking. It is still unclear how

trafficking and phosphorylation increase function of NKCC2 and NCC. While increased

trafficking and phosphorylation usually correlate with activation, the underlying mecha-

nisms remain to be clarified. However, different studies have shown potential mechanistic

links and suggest that adequate regulation of NKCC2 phosphorylation and trafficking

facilitates its affinity to chloride and, thus, transport function ([68], [69], [70]).

In a dynamic process of endocytosis and exocytosis the transporters are balanced between

residing in the apical membrane and in subapical vesicles. Upon certain stimuli they un-

dergo exocytosis to increase surface expression which ultimately enhances Na+resorption

([71], [72]). The rate of exocytosis and the increase in apical abundance are most likely

mediated via the cAMP/PKA pathway ([73], [57], [74]). Endocytosis of NKCC2 and NCC

is reportedly regulated by the clathrin-mediated pathway ([75], [76]). Recycling of the

salt transporters via exocytosis of the intracellular pool presumably facilitates a faster

and more efficient NaCl reabsorption ([68]).

Like many proteins, NKCC2 and NCC are regulated by phosphorylation, being activated

or inhibited by this type of post-translational modification. Conserved serine and thre-

onine phosphorylation sites have been described for NKCC2 and NCC in both their N-

and C-terminal domains. Most of the regulatory phosphorylation sites are located on the

N-terminal domain ([57]) (Figure 1.3).

Regulation of NKCC2 by phosphorylation has first been described in 2003 ([77]). In

NKCC2, the three threonine sites T96, T101 and T114 (mouse and rat sequence, equiva-

lent to T100, T105 and T118 in human NKCC2) play a critical role. Individual mutations

of these threonine residues reduce NKCC2 activity; however, at least two of the sites have

to be mutated to completely abolish transport function ([70]). The serine residue S126

Introduction 11

(mouse, rat) is another important stimulatory site. With immunoblots, phosphoryla-

tion at this site is hardly detectable at baseline conditions but found phosphorylated

when NKCC2 activity is increased, and mutants lacking this site have strongly decreased

NKCC2 function ([69], [78]). It has been demonstrated that T101 and S126 play the most

important role for NKCC2 activity since combined mutation of the two residues abolishes

NKCC2 function ([69]). For NCC regulation, the critical phosphorylation sites are T53,

T58 and S71, localized within the N-terminal region of the protein. Activity of the trans-

porter has been demonstrated to correlate with phosphorylation of the threonine sites,

particularly with T58 ([79], [80], [81]). In contrast to NKCC2, an individual mutation

of T58 is sufficient to severely inhibit NCC function. However, mutations in S71 also

markedly reduce NCC function making it an important phospho-site for NCC regulation

([82], [83]).

Whether phosphorylation of NCC takes place at the apical membrane or within subapi-

cal vesicles is still under debate. For NKCC2, increased phosphorylation was shown to

coincide with its translocation to the apical membrane ([77]). For NCC, some studies

provide evidence that increased levels of the phosphorylated cotransporter are accompa-

nied by trafficking towards the plasma membrane; other studies report no correlation of

phosphorylation and trafficking processes ([84], [85]).

Since modulation of surface expression and malfunction of phosphorylation are associ-

ated with major defects in salt reabsorption, these mechanisms play a major role in the

regulation of NKCC2 and NCC activation.

1.3.3 Activation by the WNK-SPAK/OSR1 Kinase Cascade

NKCC2 and NCC are regulated by similar pathways that involve a kinase cascade of

WNK (with-no-lysine [K] kinase), SPAK (Ste20-related proline/alanine-rich kinase) and

OSR1 (oxidative stress-response kinase 1) (Figure 1.4).

The homologous SPAK and OSR1 are serine/threonine kinases with overlapping renal

expression along the distal nephron. SPAK (encoded by STK39) and OSR1 (encoded by

OXSR1) were found to interact with NKCC2 and NCC via binding directly to a docking

motif (RFXV for NKCC2 and RFXI for NCC) within the N-terminus of the transporters.

Introduction 12

A conserved docking site within the non-catalytical C-terminal of SPAK/OSR1 called

CCT (conserved C-terminal) was identified to be critical for this interaction ([86], [69],

[82], [23]). Studies on transgenic mouse models with modified SPAK expression revealed

that the T96 and T101 residues of NKCC2 are in fact the SPAK- and OSR1-dependent

phosphorylation sites and that these sites are associated with the regulation of blood

pressure ([23], [87]). For NCC, several studies illustrate the relevance of all three phos-

phorylation sites (T53, T58, S71) for transport activity, sodium reabsorption and blood

pressure regulation ([82], [80], [87]).

As members of the Ste20 kinase family, SPAK/OSR1 are able to coordinate downstream

as well as upstream molecules. In the kidney, known upstream regulators of SPAK/OSR1

are the WNK kinases. During the last decade, WNK kinases have been subject of intense

research since they were found to play a role in blood pressure regulation and activate

SPAK/OSR1 kinases by phosphorylation ([88], [89]). WNK (with-no-lysine [K]) kinases

are a unique subfamily of serine/threonine kinases that have their catalytic lysine residue

located in an unusual subdomain, distinguishing it from other kinase families. Four ho-

mologs of WNK kinases have been identified in human, of which at least WNK1 (encoded

by WNK1) and WNK4 (encoded by WNK4) are expressed in rats and mice ([90], [89],

[91]). WNK kinases are activated by osmotic stress and changes in chloride concentration

([92], [82]). They comprise a phosphorylation site at serine 382 (in rodents) which can

directly interact with chloride ions. Low chloride levels reportedly activate WNK1, ap-

parently by enabling autophosphorylation within the kinase domain ([93], [94]). The S382

residue is conserved in all WNK isoforms, suggesting that the autocatalytic mechanism

and its inhibition by chloride binding happens accordingly in all four WNK transcripts

([95]).

For the interaction with SPAK/OSR1, WNK kinases possess the same RFX[I/V]-

motif that is present on NKCC2 and NCC ([96], [97]). SPAK and OSR1 are regulated by

WNK1 and WNK4 through phosphorylation at threonine residues within their catalytic

domain of the T-loop (in rodents: T243 in SPAK, T185 in OSR1) and at serine residues

within a regulatory domain of the S-motif (in rodents: S383 in SPAK, S325 in OSR1)

([88], [98], [99]).

Introduction 13

Figure 1.4. Activation of NKCC2 and NCC by the WNK-SPAK/OSR1 Kinase Cascade

NKCC2 and NCC, the salt transporters of the thick ascending limb and the distal convoluted tubule,

respectively, are activated by phosphorylation processes through a kinase cascade involving SPAK/OSR1

and WNK kinases.

1.3.4 Endocrine Control of NKCC2 and NCC

Both NKCC2 and NCC are regulated by hormonal signaling pathways which affect sodium

reabsorption and arterial blood pressure. One of the most studied and most potent hor-

mones stimulating NKCC2 activity is arginine vasopressin (AVP) ([100], [101], [102]).

AVP can induce an increase of phosphorylation levels at the important regulatory sites

T96 and T101 and thereby activate NKCC2 transport ([77]). Hormones like AVP exert

their effects via an increase of intracellular cyclic adenosine monophosphate (cAMP) levels

which in turn activates the protein kinase A (PKA). Experiments with cAMP and PKA

stimulation have shown to control NKCC2 function but only phosphorylate the S126 site

within the N-terminal domain of NKCC2, leaving T96 and T101 sites unaffected ([103],

[104]). This suggested that phosphorylation at site S126 is PKA-dependent whereas AVP-

induced T96 and T101 phosphorylation must be regulated by other kinases activities, e.g.

SPAK or OSR1. It has further been shown that the vasopressin receptor 2(V2R)-specific

agonist desmopressin increases the abundance of phosphorylated SPAK/OSR1 suggesting

that SPAK mediates the effects of AVP on distal sodium reabsorption ([105]).

In the DCT, a stimulatory effect of AVP on NCC function and salt reabsorption has early

been demonstrated by Elalouf et al. ([55]). The crucial phosphorylation sites of NCC,

T53, T58 and S71, are regulated by AVP and there is compelling evidence that this effect

Introduction 14

is mediated via SPAK/OSR1 ([20], [106], [105], [23]).

As mentioned in chapter 1.1, an activated renin-angiotensin-aldosterone-system (RAAS)

has stimulatory effects on sodium reabsorption in the nephron. In the DCT, both al-

dosterone and angiotensin II are able to stimulate NCC expression and modulate NCC

phosphorylation, either independently or as a result of combined effects. The activation of

NCC by aldosterone and angiotensin II is likely mediated by SPAK ([81], [54], [107], [108]).

The discovery of the WNK-SPAK/OSR1 signaling cascade has helped understanding the

mechanisms of blood pressure regulation and the development of hypertension. The ki-

nase cascade provided a new link between hormones and salt transport in the distal

nephron ([105], [109], [110], [108]). However, the impact of endocrine pathways in the

pathomechanism of hypertension and especially CNI-induced hypertension still remain to

be clarified.

1.3.5 Pathophysiology of NKCC2 and NCC

Dysfunction of the distal salt transporters perturbs sodium reabsorption and may dete-

riorate blood pressure. Loss of function mutations within the genes encoding NKCC2

and NCC cause Bartter’s and Gitelman’s syndrome, respectively. While symptoms differ

among patients, both diseases are characterized by pronounced salt wasting, hypokalemia,

and potentially low blood pressure despite high plasma renin activity and high aldosterone

concentrations ([111], [64]).

Mutations in the genes encoding members of the WNK-SPAK/OSR1 cascade are asso-

ciated with the development of hypertension. Mutations in WNK1 and WNK4 have

been identified to be responsible for the development of the rare monogenic disease FHHt

(familial hyperkalemic hypertension) or otherwise named Gordon syndrome or pseudo-

hypoaldosteronism type 2, characterized by hyperkalemia and hypertension ([112], [24]).

Mutations in the WNK1 gene are large deletions which increase WNK1 expression, and

mutations in WNK4 lead to accumulation of WNK4. The overabundance of WNK protein

stimulates the downstream signaling cascade by phosphorylating SPAK/OSR1 which in

turn phosphorylate NCC. Activation of NCC, the major pathogenic factor of FHHt, leads

Introduction 15

to an excess in sodium reabsorption in the DCT which subsequently causes hypertension

([83], [109], [10]).

Since adverse effects associated with calcineurin inhibitors used in immunosuppressive

therapy resemble FHHt symptoms, it is tempting to assume that CNIs affect the same

signaling pathway. In fact, several studies have shown that CNIs stimulate NKCC2 and

NCC function ([16], [113]). The fact that FHHt-related hypertension is a sole result of

NCC activation while for this syndrome increased levels of phosphorylated NKCC2 have

not been reported indicates that NCC is sufficient to induce hypertension in certain condi-

tions. It remains unclear why in FHHt the accumulation of WNK protein is not associated

with a stimulation of NKCC2 function, although multiple studies have demonstrated that

WNK kinases are signaling via SPAK/OSR1 to NKCC2/NCC, providing compelling evi-

dence that WNKs can activate NKCC2.

Only recently, two upstream regulators of WNK kinases, cullin-3 (CUL3, 89 kDa) and

kelch-like protein 3 (KLHL3, 65 kDa), have been discovered and FHHt-causing mutations

were identified within their encoding genes. KLHL3 (encoded by KLHL3) is an adaptor

protein that brings WNK kinases in close proximity with the E3 ubiquitin-protein ligase

CUL3 (encoded by CUL3) which is able to ubiquitylate WNK and thereby tag it for

degradation ([114], [115]). Mutations found in the KLHL3-CUL3 complex all lead to

impaired degradation of WNK, either due to impairing the interaction of KLHL3 with

CUL3 or WNK, or due to ubiquitylation of KLHL3 itself ([116], [117], [118]).

Different expression levels of CUL3 and KLHL3 may play a role in the regulation of

NKCC2 and NCC in FHHt or CNI-induced renal side effects. However, it is still unclear

whether CUL3 and KLHL3 are differentially expressed along the distal nephron and, thus,

might have differential effects on NKCC2 and NCC function.

1.4 Effects of Cyclosporine and Tacrolimus on NKCC2 and

NCC Function

Besides their systemic action described in chapter 1.1 the two calcineurin inhibitors cy-

closporine and tacrolimus exert local effects on the kidney. Especially in the renal distal

Introduction 16

nephron, these effects play a crucial role in the development of hypertension ([16], [18],

[17]). Interestingly, NKCC2 and NCC are differentially affected by the two drugs. While

CsA has been shown to increase both NKCC2 and NCC activity, tacrolimus is suggested

to merely stimulate NCC function. The fact that calcineurin inhibition locally affects

the distal nephron suggests that the calcineurin phosphatase is involved in the regulation

of NKCC2 and NCC. The two transporters are under control of phosphorylation and

dephosphorylation processes and studies have indeed provided compelling evidence that

they are a target of calcineurin ([68], [19]).

1.4.1 Calcineurin

Calcineurin or protein phosphatase 3 (PP3, formerly called PP2B) is a Ca2+-calmodulin-

dependent serine/threonine phosphatase that is activated by increased intracellular Ca2+

concentrations. The full protein consists of a catalytic subunit of approximately 60 kDa

(PP3-A/CnA) encoded by the PPP3C gene and of a regulatory subunit of 19 kDa (PP3-

B/CnB) encoded by PPP3R. Three isoforms with a sequence identity of roughly 80 %

exist of the catalytic subunit: Isoform α,βand γ. The latter is only expressed in testis

and brain and is absent at least in the healthy kidney ([119], [120]). Two isoforms ex-

ist of the regulatory calcineurin B subunit: CnB1 which is ubiquitously expressed and

CnB2 which is only found in testis ([121]). The catalytic calcineurin A subunit possesses

a phosphatase domain that is necessary for interaction with phosphorylated substrates.

The regulatory CnB subunit binds calcium and calmodulin and, upon increased cytosolic

Ca2+ levels, facilitates the conformational change that is needed for activation of the CnA

phosphatase domain ([122], [123]).

Although ubiquitously expressed, the effects of activated calcineurin are best described in

T cells, where it activates the transcription factor NFAT through dephosphorylation and

thus induces gene expression of immunomodulatory genes (see chapter 1.1). In the kid-

ney, CnAα-CnB and CnAβ-CnB compose two functional enzymes; however, their renal

expression pattern is still controversial. Furthermore, different studies suggest distinct

cellular functions for the two isoforms ([124], [13]). While in vitro experiments show that

CnAαand CnAβhave similar catalytic activity on various substrates, they seem to differ

Introduction 17

in their physiological function as CnAαand CnAβknockout mice exhibit different phe-

notypes. Mice lacking the βisoform have an immature immune system, whereas mice

lacking the αisoform can still be immunosuppressed but manifest developmental defects

and kidney dysfunction ([125], [126], [127]). Thus, particularly in the kidney, the two

catalytic isoforms appear to have distinct functions.

Since calcineurin reportedly plays a role in distal nephron function ([128], [68]) and the re-

nal effects of calcineurin inhibition differ depending on the administered CNI ([19], [129]),

it has been suggested that the expression pattern of the calcineurin isoforms is the key to

this phenomenon.

Another tempting assumption is that calcineurin isoforms differ in substrate specificity.

Calcineurin binds its substrates via certain recognition sites in the respective proteins or

peptides. The PxIxT motif and the LxVP motif were detected in NFAT and other cal-

cineurin substrates ([130], [131], [132]). While roughly 50 calcineurin substrates have been

experimentally confirmed so far, almost 600 potential substrates were identified based on

bioinformatics analysis. It has been postulated that the two binding motifs establish the

affinity of calcineurin substrates or regulators in cooperation ([133]).

CnAαand CnAβfurther have significant sequence variance within their N-terminal region

which may cause substrate specific binding of the isoforms. Whether substrate specificity

of the catalytic subunits or their differential expression along the nephron is the key to

the distinct effects of calcineurin inhibition remains to be clarified.

1.4.2 Calcineurin Inhibition

Inhibition of calcineurin through CNIs occurs via binding of CsA and tacrolimus to their

respective intracellular receptors, the cyclophilins and Fkbp12, which collectively are

known as immunophilins. The hydrophobic groove at the interface of the CnA and CnB

subunits that interacts with the LxVP motif of calcineurin substrates has been identified

to be the docking site of the CNI-immunophilin-complex which competes with calcineurin

substrates. By binding to calcineurin, the CNI-immunophilin-complex completely inhibits

Introduction 18

enzymatic activity of the phosphatase, thus suppressing the phosphatase-controlled acti-

vation of NFAT and preventing lymphokine gene expression ([134], [135], [132]).

CsA and tacrolimus have been effectively used for immunosuppression for more than 20

years. The cyclic peptide CsA, extracted from a soil fungi, and the macrolide tacrolimus,

isolated from a soil bacterium, are both highly lipophilic and have a similar molecular

weight (approximately 1.200 g/mol and 800 g/mol, respectively) ([136], [137]). With re-

gard to immunosuppression, tacrolimus is more potent exhibiting a lower risk of rejection

than CsA. Tacrolimus is also less nephrotoxic than CsA but is associated with more neu-

rotoxic and gastrointestinal problems ([138], [139]). However, both drugs can ultimately

cause volume expansion and hypertension, but the local effects on the renal distal nephron

are dissimilar. CsA is able to stimulate NKCC2 and NCC function but tacrolimus only in-

creases NCC activity ([68], [18], [17], [140], [15]). A potential reason for this phenomenon

could be a differential expression of the CNI-binding proteins, the immunophilins, in the

distal nephron.

Besides being the endogenous receptors of CNIs, immunophilins have peptidyl-prolyl-

cis/trans-isomerase activity facilitating protein folding, and as chaperones they are in-

volved in protein trafficking and molecular assembly. Cyclophilin A (CypA) and cy-

clophilin B (CypB), encoded by PPIA and PPIB, are the isomerases most sensitive to

CsA-inhibition. The 18 kDa protein CypA is considered to be the principal target of

CsA, it binds the drug with high affinity. CypB, a 24 kDa protein, contains an ER signal

sequence that is absent in CypA and is believed to play a role in the secretory pathway and

maturation of proteins destined to the plasma membrane ([141], [142]). The tacrolimus-

binding protein Fkbp12 (encoded by FKBP1A) is the smallest protein of the Fkbp family

(12 kDa) containing the minimal sequence for a peptidyl-prolyl isomerase. Besides the iso-

merase function and the immunosuppressive activity of the tacrolimus-Fkbp12 complex,

Fkbp12 associates with membrane receptors and seems to play a role in their regulation

([143], [144]).

While cyclophilins and members of the Fkbp family equally inhibit calcineurin activity,

they have dissimilar protein structure, and their renal expression remains to be precisely

clarified.

Introduction 19

That CNI-induced hypertension is mediated not only by local and intraepithelial but also

by paracrine and endocrine mechanisms has been suggested in several in vitro and in vivo

studies, and these effects seem to be more pronounced with CsA than with tacrolimus

administration ([145]). CsA-induced vascular effects such as AVP-mediated contraction

of smooth muscle cells have been described ([146]). Also, renin as well as cyclooxygenase-

2 (COX-2) expression has been shown to be affected by CsA and tacrolimus treatment

([145], [21]). COX-2 is constitutively expressed in renal macula densa cells and its se-

cretion stimulates the local prostaglandin pathway ([147]). Different prostaglandins are

implicated in the regulation of blood pressure via modulation of the glomerular filtration

rate, renin synthesis and NKCC2 function ([148], [149]). Since COX-2 transcription is

regulated by the NFAT transcription factor and thus depends on calcineurin signaling

([150], [151]) it has been suggested that suppression of COX-2 by calcineurin inhibition is

one major factor that complicates the use of CNIs. Altogether, paracrine and endocrine

pathways such as AVP signaling, COX-2 expression and the RAAS may aggravate the

well-known CsA-induced adverse effects.

In view of their broad clinical use, unraveling the local and systemic mechanisms of

calcineurin inhibition that affect renal homeostasis would be highly beneficial.

Chapter 2

Hypotheses, Aims and Study

Design

The calcineurin inhibitors (CNIs) cyclosporine A (CsA) and tacrolimus are effective im-

munosuppressive agents for the treatment of several autoimmune diseases and the preven-

tion of organ rejection, since they block the calcineurin-NFAT signaling pathway and stop

the expression of inflammatory response genes. When chronically administered, however,

CNI treatment may be accompanied by severe side effects particularly affecting the kidney

which manifests in hypertension and electrolyte disorders. Previous studies have indicated

that calcineurin inhibition is associated with activation of the crucial cation coupled co-

transporters NKCC2 and NCC in the distal nephron, suggesting the involvement of the

calcineurin phosphatase in the regulation of the transporters. Interestingly, tacrolimus

has been shown to only stimulate NCC function, whereas CsA affects transport activity

in the entire distal nephron. Additionally, various studies suggest that tacrolimus-induced

activation of NCC and concomitant increase of arterial blood pressure primarily rely on lo-

cal calcineurin inhibition. In contrast, CsA-induced hypertension may partially be caused

by systemic effects of calcineurin inhibition, e.g. stimulated vasopressin signaling or acti-

vation of the renin angiotensin aldosterone system. In order to improve the benefit/risk

ratio of CNI treatment, we aimed to uncover the mechanism behind the distinct effects

of CsA and tacrolimus in the distal nephron.

Hypothesis, Aim and Study Design 21

Aim 1: The first part of this study focused on the characterization of the key factors

regulating the inhibition of calcineurin and the activation of the distal salt transporters.

With this, we sought to gain further insight into the molecular mechanisms underlying the

local signaling of calcineurin inhibition. We hypothesized that the distinct effects of CsA

and tacrolimus are due to differential expression of crucial proteins involved in calcineurin

inhibition.

Aim 2: The main objective was to differentiate between local and systemic effects of CsA

in the regulation of the distal salt transporters NKCC2 and NCC. Thus, we sought to

describe acute and chronic effects of CsA administration in vivo and in vitro, evaluate

physiological changes and study the role of endocrine factors in this context. We specu-

lated that additional systemic effects of calcineurin inhibition might be key to the different

CNI-sensitivities of the distal nephron.

Aim 3: In a third approach we sought to investigate the in vivo effects of CsA treat-

ment in the absence of NCC. In a previous study, tacrolimus treatment did not induce

hypertension in NCC knockout mice. We speculated that the impact of CsA treatment

on blood pressure is more diverse and that potential systemic mechanisms affect renal

homeostasis.

The present study aimed to contribute to a better understanding of the mechanisms of

CNI-induced hypertension and may help improve the clinical use of CNIs and alleviate

their negative side effects.

Chapter 3

Material and Methods

3.1 Animals, Tissues, Treatments

All animals, except for the NCC knockout mouse model, were bred and held at the re-

search facilities for experimental medicine (Forschungseinrichtungen f¨ur experimentelle

Medizin, FEM) and later in the facilities of the Charit´eCrossOver (CCO) at the Charit´e

University Medicine of Berlin. All animals used in this study were adult male rats and

mice, respectively, which were kept at a standard diet and tap water ad libitum.

Wistar rats, outbred albino rats developed in the Wistar Institute ([152]), were used as

wildtype (WT) model. For evaluation of short-term CsA effects, WT rats were divided

into groups (n=5 for biochemical evaluation and at least n=4 for morphology) and in-

jected intraperitoneally (i.p.) with CsA (30 mg/kg body weight, Sandimmune, Novartis,

N¨urnberg, Germany) or vehicle (Chremophor, Sigma-Aldrich, M¨unchen, Germany) for

1 h and 4 h, respectively. At end point, rats were sacrificed and the kidneys removed

and decapsulated for biochemical analysis. For morphologic evaluation, anesthetized rats

(ketamine/xylazine, 0.016 mg/g BW and 0.12 mg/g BW, Sigma-Aldrich) were perfusion-

fixed retrogradely via the aorta abdominalis ([153]).

For long-term CsA studies, WT rats received a daily dose of CsA (30 mg/kg body weight)

or vehicle (Chremophor) for 14 days by subcutaneous injection and animals were divided

into groups (n=10 for physiological analysis and n=4 for morphology). For physiolog-

ical analysis, animals were individually placed in metabolic cages and received normal

food and tap water ad libitum. 24 h urine was collected to determine urine sodium and

Material and Methods 23

creatinine concentrations as well as plasma renin activity. Mean arterial blood pressure

measurement was performed applying the non-invasive tail-cuff method on anesthetized

rats. For evaluation of NKCC2 activity, furosemide tests were performed by injecting one

dose of furosemide into the peritoneum (40 mg/kg body weight, Sigma-Aldrich) and col-

lecting urine 4 h following injection. Blood was collected via the tail vein during sacrifice

prior to organ collection. Perfusion fixation was performed for morphologic evaluation.

To evaluate the effects of calcineurin inhibition in the absence of AVP, Brattleboro rats

(n=6 per group) were injected with CsA (30 mg/kg body weight, i.p.) or vehicle (Chre-

mophor), sacrificed 4 h after injection, and kidneys were removed. Brattleboro rats lack

endogenous AVP due to a mutation in the AVP gene precursor which disturbs the ability

to concentrate urine and causes central diabetes insipidus ([154]).

For characterization of protein and mRNA expression patterns along the nephron, wild-

type (WT) mice with a Balb/c background were utilized. For morphologic analysis, mice

were perfusion fixed and in situ hybridization and immunofluorescent labeling were per-

formed on paraffin embedded kidney sections. For biochemical analysis, microdissected

tubule segments from mouse kidneys were supplied by the group of Markus Bleich (In-

stitute of Physiology, Kiel, Germany) and qPCR and immunoblotting was performed on

lysates from tubule sections.

These experiments were approved by the Regional Office for Health and Social Affairs

Berlin (LAGESO permission G0220/12).

For studies of the effects of CsA treatment in mice lacking the distal salt transporter NCC,

the NCC knockout (KO) mouse model was applied. The NCC KO animals were generated

by the group of David Ellison at the Oregon Health and Science University, Oregon, US.

The knockout of the NCC gene is achieved by the disruption of exon 12 through insertion

of a neo gene. NCC KO and WT mice can be distinguished by genotyping using a

standard RT-PCR protocol ([155]). Experiments on this mouse model were performed

at the facilities of David Ellison’s group and were approved by the OHSU Institutional

Animal Care and Usage Committee (Protocol IS03286).

For evaluation of long-term CsA effects in NCC KO mice, animals were divided into groups

and injected daily with CsA (30 mg/kg body weight, i.p.) or vehicle (Chremophor) for

12 days. Blood pressure measurement (n=6 mice per group) was performed using the

radio telemetry method. For morphologic evaluation, anesthetized mice were perfusion

fixed (at least n=2) ([153]) and immunofluorescent labeling was performed on paraffin

Material and Methods 24

embedded kidney sections. For biochemical analysis using immunoblotting, mice (n=5)

were sacrificed at end point and the kidneys removed and decapsulated.

3.2 Perfusion Fixation and Tissue Embedding

The animals were killed by in vivo perfusion fixation under ketamine/xylazine anesthesia.

The kidneys were perfused retrograde via the aorta abdominalis using PBS (phosphate

buffered saline) with sucrose (330 mosmol, pH 7.4) for 30 s followed by infusion of 3 %

paraformaldehyde in PBS (pH 7.4) for 5 min ([153]). Kidneys were removed and cut into

halfs. Kidney pieces were processed for embedding in paraffin, Tissue Freezing Medium

(Leica Biosystems, Nussloch, Germany)) or LR White medium (Electron Microscopy Sci-

ences, M¨unchen, Germany). For paraffin embedding, additional fixation was achieved by

incubating kidney pieces in 4 % formalin (Thermotex, Berin, Germany) at 4◦C overnight.

The tissues were then stored in 330 mosmol sucrose + 0.02 % sodium azide and paraffin

embedded at the Institute for Pathology (Charit´e University Medicine Berlin, CCM) or at

the Histopathology Shared Resource of the Oregon Health and Science University, respec-

tively. For cryo embedding, tissues were subsequently incubated in 800 mosmol sucrose

and then shock-frozen in liquid nitrogen-cooled isopentane and Tissue Freezing Medium

for subsequent cryostat sectioning. For electron microscopy, tissues were stored in 3◦C

paraformaldehyde plus 0.05 % glutaraldehyde at 4◦C followed by embedding in LR white

resin for subsequent preparation of ultrathin sections.

3.3 Cell Culture

All cells were cultivated in 75 cm2cell culture flasks at 95 % humidity, and 5 % CO2.

Rat medullary thick ascending limb cells (raTAL) ([156]) were cultured in renal epithelial

growth medium (Promo Cell, Heidelberg, Germany) with 1 % penicillin/streptomycin.

Mouse distal convoluted tubule cells (mDCT) ([157]) were cultured in RPMI (Roswell Park

Memorial Institute) medium (Biochrom, Berlin, Germany) with 10 % fetal calf serum and

1 % penicillin/streptavidin at 37◦C. Cells were grown to confluent monolayers, stimulated

with 1 µM CsA (Santa Cruz Biotechnology, Heidelberg, Germany), 10 µM desmopressin

(DDAVP, Sigma-Aldrich), or both agents simultaneously in culture medium for 4 h and

Material and Methods 25

then harvested in Igepal lysis buffer (Sigma-Aldrich). Whole cell lysates were prepared

for immunoblotting.

3.4 Antibodies

All antibodies used in this work were validated in previous studies by either preabsorption

tests or by using respective knockout control samples. Specific primary and secondary

antibodies used for immunoblotting, immunofluorescence and in situ hybridization are

listed in table A.1 and A.2.

3.5 Immunofluorescence

For immunofluorescence analysis, paraffin sections (4 µm) were dewaxed with xylene and

rehydrated through an ethanol series (100 % - 70 %) followed by boiling for 6 min in

citrate buffer (0.02 M citric acid, 0.09 M sodium citrate; pH 6) for antigen retrieval.

Cryo- (7 µm) sections were incubated in 0.5 % (v/v) Triton X-100 (Sigma-Aldrich) in

PBS for antigen retrieval. All tissues sections were washed in PBS prior to incubation

with primary antibodies diluted in blocking medium (1 h, overnight). Multiple stainings

were separated by washing steps and fluorescent Cy2-, Cy3- or Cy5-conjugated antibodies

were applied for detection. Evaluation of sections was performed using a Zeiss confocal

microscope (LSM 5 Exciter) and signals were analysed with respect to localization and

intensity. Kidney sections were double-labeled with antibodies against NKCC2 or NCC,

respectively, in order to identify TAL and DCT. For evaluation of signal intensities of

pNKCC2, pNCC and pSPAK/OSR1, double-labeling was performed applying antibodies

against the total protein. Micrographs were obtained using ZEN2008 software (Zeiss,

Jena, Germany) and ImageJ software ([158]) was applied to analyze signal intensities in

individual tubular profiles. Mean signal values of phospho-signals within 2 µm distance

to the apical membrane were normalized to respective non-phospho signals. In cases

where the respective non-phospho antibody was not available, signals were normalized

to background signals. For evaluation of pWNK1, signal intensities were normalized by

colocalized NCC signal in the DCT. Analysis of two to six animals per group with at least

20 similar tubular profiles per individual were performed in a blind fashion. For COX-2

Material and Methods 26

signal quantification at the macula densa and adjacent TAL portions, cells histochemically

positive for COX-2 were counted.

3.6 Immunoblotting

Whole kidneys and microdissected nephron segments were homogenized in buffer contain-

ing 250 mM sucrose, 10 mM triethanolamine and protease inhibitors (Complete, Roche

Diagnostics, Berlin, Germany), sonicated and centrifuged (1000 xg for 10 min) to remove

nuclei. The same protocol was used for protein extraction of cell lysates. Proteins were

electrophoretically separated using 10 % polyacrylamide minigels and then transferred

onto nitrocellulose membranes (Macherey-Nagel, D¨uren, Germany). Membranes were

blocked with respective blocking buffer for 30 min and subsequently incubated with pri-

mary antibody (see A.1) for 1 h at room temprature or overnight at 4◦C. HRP-conjugated

secondary antibodies (see A.2) were applied for detection following three washing steps be-

tween antibodies. Membranes were then incubated with chemiluminescent reagent (ECL

Western Blotting Detection Reagents, Amersham, UK) and signals were detected using

the ChemoCam Imager ECL (Intas, G¨ottingen, Germany). Densitometric evaluation was

performed using ImageJ software ([158]).

3.7 Quantitative PCR

RNA from cell and tissue samples was extracted using the RNA extraction kit (Stratec

biomedical, Birkenfeld, Germany). Reverse transcription into cDNA was performed in a

two-step reaction. For denaturation of RNA and hybridisation of Oligo(dT)18 primers

(Bioline GmbH, Luckenwalde, Germany), samples were first incubated 10 min at 70◦C.

Next, Tetro Reverse Transcriptase (Promega, Mannheim, Germany), 10 mM dNTPs (Bi-

oline GmbH) and 10 U Ribolock RNase Inhibitor were added and incubated 2 h at 37◦C

for cDNA synthesis. Quantitative PCR was performed using HOT FIREPol EvaGreen

qPCR Mix (Solis BioDyne, Tallinn, Estonia) using primers specific for target genes (see

table A.3). Experiments were run for 40 cycles and subsequent melting curve genera-

tion in a 7500 Fast Real-Time PCR System (Applied Biosystems, Darmstadt, Germany).

Material and Methods 27

The ∆∆Ct method was used for gene expression analysis and expression was normalized

against β-actin or against whole kidney control samples for tubule segments.

3.8 In Situ Hybridization

For evaluation of renin mRNA as well as for localization of CypA and CypB, in situ

hybridzation was performed on perfusion-fixed paraffin embedded mouse and rat kidney

sections. Digoxygenin (DIG)-labeled (Sigma-Aldrich) antisense RNA probes were gener-

ated via transcription from full length cDNA clones. Sections were dewaxed in xylene

and ethanol, digested with proteinase K (Roche) for membrane permeabilization and hy-

bridized with the respective probes at room temperature overnight. Anti-DIG-alkaline

phosphatase-conjugated antibody (Dako) was applied for detection of hybridized probes

and visualization was performed by incubating the sections with nitro blue tetrazolium

and 5-bromo-4-chloro-3-indolyl phosphate (Roche). Sections were analyzed with a Leica

DMRB microscope (Leitz).

3.9 Ultrastructural Analysis

For analysis of NKCC2 and NCC distribution, transmission electron microscopy was per-

formed on perfusion-fixed ultrathin LR white medium kidney sections. Primary antibodies

against NKCC2 and NCC and 10 nm nanogold-labeled secondary antibodies (Amersham)

were applied for detection. Visualization was performed using transmission electron mi-

croscopy (TECNAI G-2, Thermo Fisher Scientific, Berlin, Germany). Immunogold signals

of NKCC2 in TAL and NCC in DCT, respectively, were quantified on at least 10 profiles

and four cells per profile per individual animal. Signals within a 20 nm distance to the

plasma membrane were defined as membrane-bound, whereas signals detected above 20

nm until the nuclear envelope were defined as cytoplasmic.

3.10 Genotyping

Identification of the genotype of wildtype and NCC knockout mice was achieved applying

a standard genotyping protocol using tissue from tail biopsies. DNA was amplified in

Material and Methods 28

a PCR reaction with one forward (5-AGGGTCAAGGGCACGGTTGGC-3) and two reverse

primers (1: 5-GGTAAAGGGAGCGGGTCCGAGG-3; 2: 5-GCATGCTCCAGACTGCCTTG-3)

creating PCR products of different size depending on the genotype. The forward primer

and the reverse primer 1 correspond to the intron sequences flanking exon 12 which is

disrupted in NCC KO mice through insertion of the neo gene. In wildtype mice, these

primers will amplify a 265 base pair product. The reverse primer 2 is complementary to

sequences within the neo gene. Thus, in NCC KO, the forward primer and the reverse

primer 2 amplify a 188 base pair product.

3.11 Blood Pressure Measurements

Evaluation of mean arterial blood pressure in rats receiving long-term CsA treatment was

performed at the Max-Delbr¨uck-Center for Molecular Medicine (MDC Berlin, Germany)

applying the non-invasive tail-cuff method on anesthetized rats.

Evaluation of blood pressure in NCC KO mice was performed using the radio telemetry

method, the gold standard for the monitoring of arterial pressure in conscious mice. The

technique involves a surgical procedure under anesthesia where a thin, flexible catheter

is inserted from the left carotid artery into the aortic arch and the telemetry probe is

implanted under the skin ([159]). Data were recorded by the implanted telemetry device

at 10 min intervals and 24 h mean values were plotted.

3.12 Statistics

Samples from microdissected tubule sections were considered as independent and unre-

lated groups. Quantitative PCR data from these samples were determined applying one-

way ANOVA with Tukey’s HSD post hoc test in order to identify the sample-to-sample

differences. For statistical evaluation of the effects of different treatments in mice, rats

and cell culture, parametric Student’s t-test or non-parametric Mann-Whitney-test was

applied to test for significant differences between groups. Two-way ANOVA with Bonfer-

roni correction was used for physiological data received from metabolic cage experiments

and for evaluation of different treatments in wildtype and NCC KO mice. P-values of

P < 0.05 were considered significant. All data are expressed as means ±SEM.

Chapter 4

Results – Part I: Key Factors

Involved in Local Calcineurin

Inhibition

CNI therapy is accompanied by hypertension and salt retention mediated, among other

factors, by activation of the renal sodium chloride cotransporters. CsA and tacrolimus

herein display disparate effects with respect to the activation of NKCC2 and NCC. We

hypothesized that this is due to the differential expression of key factors involved in

calcineurin inhibition. Thus, in this part of the study, localization of the two pivotal cal-

cineurin isoforms CnAαand CnAβ, and the immunophilins CypA, CypB and Fkbp12 was

analyzed along the nephron applying qPCR, immunoblotting and in situ hybridization.

Since the recently identified adapter and ubiquitylation proteins KLHL3 and CUL3 might

play a role in hypertension, we further hypothesized that their expression pattern may

have differential effects on NKCC2 and NCC function in calcineurin inhibition conditions.

Therefore, analysis of KLHL3 and CUL3 localization was performed applying the same

protocols. Characterization of the expression pattern of these factors provides valuable

insights into the molecular mechanisms underlying calcineurin inhibition on a local level.

Results – Part I: Key Factors of Local Calcineurin Inhibition 30

4.1 Localization of Calcineurin Isoforms

Limited information is available about the renal expression of calcineurin. There is ev-

idence that CnAα, the predominant isoform in the kidney, is the major isoform in the

cortex while CnAβis primarily expressed in the medullary TAL ([160], [120], [15]). Our

group was recently able to confirm CnAβexpression in the TAL ([68]). To provide a clear

picture of calcineurin expression in the nephron, quantitative PCR and immunoblotting

were performed on microdissected mouse nephron segments. PCR analysis revealed ubiq-

uitous mRNA expression of both CnAαand CnAβisoforms with highest expression levels

in the two distal tubule segments TAL and DCT, but no significant differences between

the segments (Figure 4.1 A). Immunoblotting confirmed the presence of both isoforms in

all nephron segments including the glomerulus (Figure 4.1 B).

CnAα

CnAβ

Tubulin

Actin

60 kDa

40 kDa

60 kDa

50 kDa

Glom PT TAL DCT CD

Figure 4.1. Calcineurin Isoforms Are Expressed Ubiquitously along the Nephron. (A) Quan-

titative PCR analysis of calcineurin isoforms CnAαand CnAβon lysates of microdissected tubule sections.

Expression was normalized to whole kidney. Data are the means ±SEM. (B) Representative immunoblots

of lysates from microdissected tubule sections showing immunoreactive signals for CnAαand CnAβ, both

at approximately 60 kDa; β-actin (approximately 40 kDa) and tubulin (approximately 50 kDa) served as

loading control. PT: proximal tubule, TAL: thick ascending limb, DCT: distal convoluted tubule, CD:

collecting duct, Glom: glomerulus.

These results disprove the hypothesis that the distinct effects of local calcineurin inhibi-

tion are based on differential expression of the calcineurin isoforms in the distal nephron.

Nevertheless, differences in post-translational regulation cannot be excluded and substrate

specificity of the two isoforms may play a pivotal role. However, differences in the reg-

ulation of the sodium chloride cotransporters in the TAL and DCT by the calcineurin

phosphatase do not seem to be based on the expression pattern of CnAαand CnAβ.

Results – Part I: Key Factors of Local Calcineurin Inhibition 31

4.2 Localization of Immunophilins

Next, we hypothesized that the immunophilins, the cytosolic receptors of CsA and tacrolimus,

are differentially expressed along the nephron since the TAL and the DCT differ in their

sensitivity to CsA and tacrolimus. To shed light on this question, our group has re-

cently investigated the renal expression of CypA, CypB and Fkbp12 by performing im-

munoblots on lysates from microdissected nephron segments, and detected each of the

three immunophilins in all segments ([68]). To confirm this observation on the mRNA

level, additional quantitative PCR on lysates from the isolated nephron segments was

performed and confirmed the ubiquitous expression along the entire nephron (Figure 4.2).

The CsA binding proteins CypA and CypB appeared with highest expression in the dis-

tal tubule segments TAL and DCT; however, differences were not statistically significant.

Overall expression of the tacrolimus binding protein Fkbp12 was quite low showing similar

levels within the different segments and no significant changes in TAL and DCT (Figure

4.2).

Figure 4.2. Immunophilins Are Ubiquitously Expressed along the Nephron. Quantitative

PCR analysis of immunophilins CypA, CypB and Fkbp12 on lysates of microdissected tubule sections.

Expression was normalized to whole kidney. Data are the means ±SEM. PT: proximal tubule, TAL:

thick ascending limb, DCT: distal convoluted tubule, CD: collecting duct, Glom: glomerulus.

In situ hybridization on perfusion-fixed mouse and rat kidney sections was further per-

formed to verify mRNA expression of the immunophilins in the distal nephron. Strong

Results – Part I: Key Factors of Local Calcineurin Inhibition 32

signal intensities were detected for both CypA and CypB in the TAL and DCT, confirm-

ing the high expression of cyclophilins in the distal tubule segments (Figure 4.3 A and B).

Fkbp12 expression could not be evaluated by in situ hybridization for technical reasons.

Cyclophilin A

Cyclophilin B

Figure 4.3. CsA Binding Proteins Cyclophilin A and Cyclophilin B Are Highly Expressed

in the Distal Nephron. (B) Representative images of rat kidney sections showing mRNA signal of

(A) cyclophilin A and (B) cyclophilin B detected by in situ hybridization. Strong signal intensities were

found in the thick ascending limb (TAL, *) and the distal convoluted tubule (DCT, +), weaker expression

was detected in the proximal tubule (PT) and other segments confirming qPCR analysis. Bars indicate

20 µm.

The detection of the three immunophilins CypA, CypB and Fkbp12 in all nephron seg-

ments does not explain the different sensitivity to calcineurin inhibitors in the distal

tubule. Thus, the distinct effects of CsA and tacrolimus have to be mediated by other

mechanisms.

Results – Part I: Key Factors of Local Calcineurin Inhibition 33

4.3 Localization of KLHL3 and CUL3

The next step was to clarify the renal expression of the two recently identified WNK reg-

ulators, CUL3 and KLHL3, which are key factors in the ubiquitylation process of WNKs.

Mutations in the genes encoding these proteins cause familial hyperkalemic hypertension

(FHHt) and lead to accumulation of WNK protein which in turn upregulates NCC ac-

tivity. Since only in the DCT signaling pathways seem to be affected by this, we asked

whether CUL3 and KLHL3 were differentially expressed along the nephron. Quantita-

tive PCR and immunofluorescent staining was performed on isolated nephron segments

to characterize the expression pattern on the mRNA and the protein level, respectively.

qPCR revealed KLHL3 mRNA expression in all nephron segments with highest expression

levels in the DCT. CUL3 mRNA was detected in all nephron segments with no statistical

significance between segments (Figure 4.4).

TAL

DCT

Glom

TAL

DCT

Glom

expression vs. whole kidney

KLHL3

CUL3

Figure 4.4. WNK Regulators KLHL3 and CUL3 Are Ubiquitously Expressed along the

Nephron. Quantitative PCR analysis of WNK regulators KLHL3 and CUL3 on lysates of microdissected

tubule sections. Expression was normalized to whole kidney. Strongest KLHL3 expression was detected in

the distal nephron and collecting duct with highest levels in the DCT. CUL3 mRNA was relatively low in

all segments. Data are the means ±SEM. PT: proximal tubule, TAL: thick ascending limb, DCT: distal

convoluted tubule, CD: collecting duct, Glom: glomerulus.

Immunofluorescence staining revealed that medullary TAL and DCT are enriched with

KLHL3 protein. Low to intermediate expression was found in the other tubule sections

with the thin descending limb lacking any expression. Immunofluorescence further con-

firmed ubiquitous expression of CUL3, however, with lower signal intensities in the distal

nephron. Additionally, intercalated cells of the collecting duct displayed high expression

Results – Part I: Key Factors of Local Calcineurin Inhibition 34

levels of KLHL3 (Figure 4.5). These results illustrate the vital role of the WNK regulators

KLHL3 and CUL3 in the distal nephron; however, they do not explain the differential

actions of CNI in this area.

Figure 4.5. Protein Expression of KLHL3 and CUL3 along the Nephron. Immunofluorescence

of KLHL3 and CUL3 on paraffin embedded kidney sections as evaluated by confocal microscopy. (A, B)

KLHL3 in DCT segments (identified by NCC) and TAL (identified by NKCC2). (C, D) Colocalization

with WNK4 in DCT and TAL (butterfly sections of Aand B). (E) KLHL3 in the medulla with robust

expression along the thin limb (TL) and TAL. (F) KLHL3 in the cortex at the junction of a DCT, CNT,

and cortical collecting duct (cCD) with enriched expression in intercalated cells (IC). CNT and cCD

were identified by aquaporin 2 (AQP2). (G) KLHL3 in the medullary collecting duct (mCD) where it is

expressed in intercalated, but not principal cells. (H) CUL3 was detected most highly in proximal tubules

(PT) and expression in DCT was present but low. (I) Schematic expression sites of KLHL3 and CUL3.

OM: outer medulla; OS: outer stripe, IS: inner stripe; IM: inner medulla; C: cortex; MD: macula densa;

intercalated cells indicated by boxes. Bars indicate 20 µm ([161]).

Chapter 5

Results – Part II: Local and

Systemic Effects of Calcineurin

Inhibition

Since we showed that some major regulators of NKCC2/NCC activation display a similar

expression pattern in the TAL and the DCT, we supposed that other mechanisms must be

the key to the different CNI-sensitivities of the distal nephron segments. Various studies

suggest that, besides the effects of local calcineurin inhibition, CsA-induced hypertension

relies on additional systemic effects. To differentiate between local and systemic effects of

calcineurin inhibition in the regulation of the distal salt transporters and blood pressure,

in this part of the study acute and chronic effects of CsA treatment were investigated

in vivo and in vitro. Additionally, the role of vasopressin was examined in this context.

The characterization of the different effects of CsA treatment significantly contributes to

a better understanding of the mechanisms of action of CNI.

5.1 CsA Treatment Activates NKCC2 and NCC on the

Post-Translational Level

To assess acute and chronic effects of CsA treatment, Wistar rats were treated for three

time periods. 1 h and 4 h were selected as short-term periods whereas 14 d was defined

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 36

as long-term treatment. 1 h treatment served to evaluate fast changes on the post-

translational level such as phosphorylation and trafficking. 4 h treatment was sought

to assess potential early changes in mRNA expression and protein abundance. The 14

d treatment period served for the evaluation of changes in electrolyte handling, blood

pressure and endocrine mechanisms.

5.1.1 Acute CsA Treatment Induces NKCC2 and NCC Activation in

vivo

First, Wistar rats were treated with CsA or vehicle (Veh) for 1 h and 4 h, respectively.

Protein abundance and phosphorylation levels of NKCC2 and NCC were evaluated by

immunoblotting. In order to analyze activation of the transporters via the SPAK cas-

cade, the T96/T101 phosphorylation site of NKCC2 and the S71 phosphorylation site

of NCC were detected, respectively, using specific phospho-antibodies (see section 1.3.3).

Phosphorylation levels of NKCC2 and NCC were increased after both 1 h (+131 % for

pNKCC2 and +190 % for pNCC, P < 0.05, Figure 5.1 A, C) and 4 h (+197 % for pNKCC2

and +410 % for pNCC, P < 0.05, Figure 5.1 B, D) post CsA administration compared to

the vehicle control groups. Total protein abundance was unchanged after both time points.

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 37

pNCC

pNKCC2

Veh CsA

NKCC2

NCC

Actin

1 h

Actin

160 kDa

40 kDa

160 kDa

40 kDa

160 kDa

40 kDa

160 kDa

40 kDa

160 kDa

pNCC

pNKCC2

NKCC2

NCC

Actin

160 kDa

40 kDa

160 kDa

40 kDa

160 kDa

40 kDa

Veh CsA

4 h

Figure 5.1. Short-Term CsA Treatment Stimulates Phosphorylation of NKCC2 and NCC.

Representative immunoblots of kidney lysates from rats treated with cyclosporine A (CsA) for (A) 1 h

and (B) 4 h, showing immunoreactive signals for total NKCC2, phosphorylated NKCC2 (pNKCC2), total

NCC, and phosphorylated NCC (pNCC), all at approximately 160 kDa; β-actin served as loading control

(approximately 40 kDa). (C, D) Graphs showing respective densitometric evaluation of immunoreactive

signals normalized to loading controls. Levels of pNKCC2 and pNCC were increased after short-term

cyclosporine A (CsA) treatment compared to the vehicle (Veh) control group, but no change in their total

protein abundance was detected. Data are the means ±SEM; *P < 0.05 ([162]).

Trafficking of NKCC2 and NCC was assessed by applying immunogold electron microscopy

on ultrathin kidney sections. After 1 h of CsA treatment surface expression of the trans-

porters was unchanged in both TAL and DCT. After 4 h of CsA treatment a moderate

increase of NKCC2 and NCC signal in the plasma membrane was detected within the

respective nephron segment compared to the control group (+20 % for NKCC2 and +13

% for NCC, P < 0.05, Figure 5.2). These data indicate that acutely administered CsA

activates NKCC2 and NCC function by increasing their phosphorylation levels and stim-

ulating their trafficking.

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 38

NKCC2NCC

Veh CsA

4 h

Figure 5.2. Short-term CsA Treatment Stimulates Trafficking of NKCC2 and NCC. (A)

Representative immunoelectron microscopic images showing cellular distribution of NKCC2 and NCC in

the plasma membrane (arrows) and the cytoplasm (arrowheads) in kidneys from vehicle-treated (Veh)

and cyclosporine A (CsA)-treated rats; 5 nm gold grain labeling. (B) Numerical quantification of NKCC2

and NCC signals in plasma membrane (PM) per respective total cellular signals. 4 h CsA increased the

surface expression of NKCC2 and NCC compared to the Veh control group. Plots showing the means ±

SEM; *P < 0.05. Bars indicate 1 µm ([162]).

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 39

5.1.2 Acute CsA Treatment Activates the WNK-SPAK/OSR1 Cascade

in vivo

To assess the phosphorylation status of SPAK and OSR1, the kinases responsible for

NKCC2 and NCC function, we next used confocal microscopy on paraffin embedded

kidney sections from rats treated with CsA or vehicle. Activation of SPAK/OSR1 after

short-term (1 h and 4 h) CsA treatment was analyzed by evaluating apical abundance

of their phosphorylated species in the TAL and the DCT. Due to the high homology of

the two kinases, phospho-antibodies detect the phospho-sites of both SPAK and OSR1.

Signals of their phosphorylated regulatory domain (pS-SPAK/OSR1) were normalized to

colocalized SPAK signals.

Interestingly, in the TAL short-term CsA treatment did not induce significant changes

of pSPAK/OSR1 levels, albeit a trend could be detected (Figure 5.3 A, C). In contrast,

in the DCT substantially increased apical signals were detected upon short-term CsA

administration (Figure 5.3 B, C). This was true for both time points (+49 % after 1 h

CsA and +94 % after 4 h CsA, P < 0.05; (Figure 5.3 C, D)).

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 40

SPAKpS-SPAK/OSR1 merge

DCT

SPAKpS-SPAK/OSR1 merge

TAL

1 h

C4 h

Veh

4 h CsA

Veh4 h CsA

SPAKpS-SPAK/OSR1 merge

TAL

DCT

Figure 5.3. Increased Phosphorylation Levels of pS-SPAK/OSR1 after Short-Term CsA

Treatment in the Distal Convoluted Tubule (A, B) Representative images of kidney sections from

vehicle- (Veh) or cyclosporine A (CsA)-treated rats (4 h) showing immunofluorescent labeling of phos-

phorylated regulatory SPAK/OSR1 domain (pS-SPAK/OSR1) and double-labeling for SPAK in (A) the

thick ascending limb (TAL) and in (B) the distal convoluted tubule (DCT). Identification of TAL and

DCT was accomplished according to morphological criteria and specific SPAK signal patterns (predomi-

nant apical signal in TAL vs. apical and punctate cytoplasmic signal in DCT). (C, D) Graphs showing

pS-SPAK/OSR1:SPAK signal ratio in TAL and DCT of rats treated with Veh or CsA for 4 h (C) and 1

h (D) as evaluated using ZEN and ImageJ software; respective immunofluorescent images for 1 h are not

shown. Short-term CsA treatment increased the abundance of pS-SPAK/OSR1 in the DCT, but not in

the TAL. Data are the means ±SEM; *P < 0.05. Bars indicate 10 µm ([162]).

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 41

Immunoblots revealed that total protein abundance of SPAK and OSR1 were not affected

by acute CsA administration as compared to the vehicle treated control group (Figure

5.4).

SPAK

OSR1

Actin

1 h 4 h

Veh CsA Veh CsA

60 kDa

40 kDa

60 kDa

40 kDa

60 kDa

40 kDa

60 kDa

40 kDa

SPAK

OSR1

Actin

Figure 5.4. Short-Term CsA Treatment Has no Effect on Total SPAK and OSR1 Protein

Abundance. Representative immunoblots of kidney lysates from rats treated with cyclosporine A (CsA)

for (A) 1 h and (B) 4 h, showing immunoreactive signals for SPAK and OSR1, both at approximately

60 kDa; β-actin served as loading control (approximately 40 kDa). (C, D) Graphs showing respective

densitometric evaluation of immunoreactive signals normalized to loading controls. Levels of SPAK and

OSR1 were unchanged after short-term CsA treatment compared to the Veh control group. Data are the

means ±SEM; *P < 0.05 ([162]).

Since WNK kinases are the known upstream regulators of SPAK/OSR1, we next as-

sessed the phosphorylation level of WNK1, the predominant isoform expressed in the

DCT ([163]). The S382 phosphorylation site of WNK1 is reported to be required for

activation of the SPAK/OSR1 kinases (see section 1.3.3). The antibody used in this

study detects the S382 phospho-site of both pWNK1 and pWNK4. Therefore, individual

changes of WNK1 and WNK4 phosphorylation may not be detectable due to overlapping

signals. pWNK1 signals were normalized to overlapping NCC signals. We could not de-

tect any change in the phosphorylation level of WNK1 upon acute CsA administration as

evaluated by confocal microscopy (Figure 5.5).

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 42

NCCpS

WNK1

NCC

WNK1

Veh

CsA

4 h

Figure 5.5. Phosphorylation Levels of pS-WNK1 after Short-Term CsA Treatment in the

Distal Convoluted Tubule (A) Representative images of kidney sections from vehicle- (Veh) or cy-

closporine A (CsA)-treated rats (4 h) showing immunofluorescent labeling of phosphorylated WNK1

domain (pS-WNK1) and double-labeling for NCC in the distal convoluted tubule. Lacking change of

phosphorylation levels upon acute CsA treatment might be due to individual changes of pWNK1 and

pWNK4 which cannot be detected with the antibody used here. (B) Graphs showing relative signal in-

tensities of phosphorylated WNK signals normalized to colocalized NCC signals as evaluated using ZEN

and ImageJ software. Data are the means Data are the means ±SEM; *P < 0.05. Bars indicate 10 µm.

5.1.3 Acute CsA Treatment Does not Affect mRNA of the WNK-SPAK/OSR1

Cascade

Next, mRNA expression profiles were investigated applying qPCR. After 4 h of CsA

treatment no changes were detected for WNKs, SPAK, OSR1, NKCC2, and NCC (Figure

5.6). These data provide further evidence that short-term calcineurin inhibition using CsA

activates the WNK-SPAK/OSR1 cascade and their renal substrates, NKCC2 and NCC,

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 43

on the post-translational level by increasing their apical abundance and phosphorylation

but not mRNA levels.

4 h

Figure 5.6. Short-Term CsA Treatment Has No Effect on mRNA Expression of the WNK-

SPAK/OSR1-NKCC2/NCC Cascade Quantitative PCR analysis of WNK1, WNK4, SPAK, OSR1,

NKCC2, NCC mRNA in kidney lysates from rats treated with vehicle (Veh) or cyclosporine A (CsA) for

4 h. All results were normalized to GAPDH expression. Data are the means ±SEM; *P < 0.05 ([162]).

5.1.4 Chronic CsA Treatment Induces NKCC2 and NCC Activation in

vivo

To evaluate long-term effects, Wistar rats were treated with CsA or vehicle for 14 days.

Phosphorylation levels of NKCC2 and NCC were analyzed using confocal microscopy.

NKCC2 and NCC labeling was used to identify TAL and DCT, respectively, and for

normalization of phosphorylation signals to colocalized signals. This revealed a CsA-

induced increase in the abundance of their phosphorylated species without concomitant

changes in the total protein levels (pNKCC2: +46 %, pNCC: +19 %, P < 0.05, Figure

5.7).

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 44

NKCC2pNKCC2 merge

NCCpNCC merge

14 d

Veh

CsA

VehCsA

NKCC2pNKCC2 merge

NCCpNCC merge

DCT

TAL

DCT

Figure 5.7. Increased Phosphorylation Levels of pNKCC2 and pNCC after Long-Term

CsA Treatment (A, B) Representative images of kidney sections from vehicle- (Veh) or cyclosporine

A (CsA)-treated rats (14 d) showing immunofluorescent labeling of (A) pNKCC2 and double-labeling

for NKCC2 in the thick ascending limb TAL or of (B) pNCC and double-labeling for NCC in the distal

convoluted tubule (DCT). (C) Graphs showing relative signal intensities of phosphorylated NKCC2 and

phosphorylated NCC signals normalized to colocalized total NKCC2 or NCC signals as evaluated using

ZEN and ImageJ software. Data are the means ±SEM; *P < 0.05. Bars indicate 10 µm ([162]).

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 45

5.1.5 Chronic CsA Treatment Activates SPAK/OSR1 in vivo

Activation of SPAK/OSR1 kinases was analyzed by evaluating apical abundance of their

phosphorylated species using confocal microscopy. Signals were analyzed for the regu-

latory (pS-SPAK/OSR1) and the catalytic form (pT-SPAK/OSR1) and normalized to

colocalized SPAK/OSR1 signals. Relative signal intensities displayed upregulation of the

phosphorylated catalytic as well as regulatory SPAK phospho species; however, only in

the DCT (pT-SPAK/OSR1: +123 %, pS-SPAK/OSR1: +36 %; P < 0.05, Figure 5.8).

Total protein abundance remained unchanged.

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 46

DCT

B C

Veh

CsA

NCCpT

SPAK/OSR1 merge

NCC

SPAK/OSR1 merge

14 d 14 d

Figure 5.8. Increased Phosphorylation Levels of pSPAK/OSR1 after Long-Term CsA Treat-

ment in the Distal Nephron (A) Representative images of kidney sections from vehicle- (Veh) or

cyclosporine A (CsA)-treated rats (14 d) showing immunofluorescent labeling of phosphorylated catalytic

SPAK/OSR1 domain (pT-SPAK/OSR1) and double-labeling for NCC in the distal convoluted tubule

(DCT). (B, C) Graphs showing (B) pT-SPAK/OSR1:SPAK signal ratio in DCT of rats treated with

Veh or CsA for 14 d and (C) pS-SPAK/OSR1:SPAK signal ratio in TAL and DCT as evaluated us-

ing ZEN and ImageJ software; respective immunofluorescent images for pS-SPAK/OSR1:SPAK are not

shown. Long-term CsA treatment increased the abundance of catalytic pT-SPAK/OSR1 and regulatory

pS-SPAK/OSR1 in the DCT, but not in the TAL. Data are the means ±SEM; *P < 0.05. Bars indicate

10 µm ([162]).

5.1.6 Chronic CsA Treatment Does not Affect mRNA of the WNK-

SPAK/OSR1 Cascade

Subsequently, qPCR was performed to assess whether CsA-induced activation of the trans-

porters and their regulating kinases is a complete result of post-translational modification.

Chronic CsA administration did not alter mRNA expression of WNKs, SPAK, NKCC2

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 47

and NCC; however, OSR1 expression was moderately increased (+24 %, P < 0.05; Fig-

ure 5.9). These data suggest that long-term CsA treatment, just like the short-term

treatment, activates the distal salt transporters and their activating kinases chiefly by

post-translational mechanisms.

14 d

Figure 5.9. Long-Term CsA Treatment Has no Effect on mRNA Expression of WNK,

SPAK, NKCC2 and NCC Quantitative PCR analysis of WNK1, WNK4, SPAK, OSR1, NKCC2, NCC

mRNA in kidney lysates from rats treated with vehicle (Veh) or cyclosporine A (CsA) for 4 h. All results

were normalized to GAPDH expression. Only OSR1 shows moderately increased expression upon CsA

treatment. Data are the means ±SEM; *P < 0.05 ([162]).

5.2 Effects of CsA on Endocrine and Paracrine Regulation

of NKCC2 and NCC

Since calcineurin inhibition, particularly using CsA, reportedly modulates endocrine and

paracrine pathways with effects on distal salt handling, we next analyzed the function of

the juxtaglomerular apparatus upon CsA treatment. Local COX-2 and renin synthesis

was evaluated by qPCR, immunofluorescence and in situ hybridization. On the mRNA

level, COX-2, which is regulated by the calcineurin-NFAT pathway, ([145]) was expectedly

suppressed by acute and chronic CsA treatment (4 h: −41 %, 14 d: −54 %, P < 0.05;

Figure 5.10).

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 48

Figure 5.10. Short- and Long-Term CsA Treatment Downregulates COX-2 mRNA Expres-

sion Quantitative PCR analysis of COX-2 mRNA in kidney lysates from rats treated with vehicle (Veh)

or cyclosporine A (CsA) for 4 h and 14 d. All results were normalized to GAPDH expression. Data are

the means ±SEM; *P < 0.05.

Immunofluorescence confirmed the suppressed juxtaglomerular COX-2 expression after

chronic CsA administration (−93 %, P < 0.05; Figure 5.11 A, C). Furthermore, renin

mRNA expression was upregulated upon long-term CsA treatment as shown by in situ

hybridization on kidney sections (Figure 5.11 B, D).

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 49

Cox-2

Renin

AVeh CsA

Renin

Cox-2

14 d 14 d

Figure 5.11. Regulation of Endocrine and Paracrine Factors by Long-Term CsA Treatment

(A) Representative images of macula densa regions in kidney sections from vehicle- (Veh) or cyclosporine

A (CsA)-treated rats (14 d) showing immunofluorescent labeling of juxtaglomerular COX-2 expression.

(B) Representative images of afferent arterioles in kidney sections from Veh- and CsA-treated rats showing

renin mRNA signal detected by in situ hybridization. (C, D) Graphs showing numerical quantification of

are the means ±SEM; *P < 0.05. Bars indicate 10 µm ([162]).

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 50

Additionally, plasma renin activity was determined in chronically treated rats and was

strongly increased after 14 d CsA (+966 %, P < 0.05; Figure 5.12).

14 d

Figure 5.12. Long-Term CsA Stimualtes Plasma Renin Activity Plasma renin activity in vehicle-

(Veh) and cyclosporine A- (CsA; 14 d) treated rats. Data are the means ±SEM; *P < 0.05 ([162]).

These results suggest that CsA stimulates renal salt reabsorption via additional systemic

mechanisms.

5.3 CsA-Induced Salt Retention and Hypertension

For evaluation of physiological changes induced by CsA treatment, rats were kept in

metabolic cages and blood pressure was measured. Urinary salt excretion was markedly

decreased on day 1 (−31 %, P < 0.05; Figure 5.13 A) and day 10 (−51 %, P < 0.05;

Figure 5.13 A) in CsA treated animals compared with the vehicle-treated control group.

Mean arterial blood pressure displayed a moderate but significant increase on day 7 of

treatment compared to the control group (113 (Veh) vs. 121 (CsA) mmHg; P < 0.05;

Figure 5.13 B).

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 51

A B

MAP

Figure 5.13. Regulation of Sodium Excretion and Mean Arterial Blood Pressure by Long-

Term CsA Treatment (A) 24 h urine sodium excretion in vehicle- (Veh) and cyclosporine A- (CsA)

treated rats on treatment days 1 and 10. Long-term CsA treatment induced marked decreases in urinary

salt excretion. (B) Mean arterial blood pressure (MAP) in anesthetized Veh- and CsA-treated rats

measured by non-invasive tail cuff measurement. Long-term CsA treatment induced a moderate but

significant increase in MAP on day 7 of treatment compared to Veh group. Data are the means ±SEM;

*P < 0.05 for Veh vs. CsA, #P < 0.05 for day 1 vs. day 10 ([162]).

Previous studies on CNI have reported that tacrolimus-induced salt retention is the chief

result of NCC activation ([51], [15]). To demonstrate the respective contribution of

NKCC2 in a CsA-induced salt retention and hypertension, we performed a furosemide

test. CsA-receiving animals displayed a stronger furosemide-induced salt loss than the

vehicle treated group, indicating that NKCC2 function is indeed increased upon CsA

treatment (+104 %, P < 0.05; Figure 5.14).

14 d

Figure 5.14. Furosemide Test Revealed CsA-Induced Activation of NKCC2 Furosemide test in

vehicle- (Veh) and cyclosporine A- (CsA) treated rats, as evaluated by the ratio of urinary sodium excre-

tion after (UNa-Furo) and before furosemide application (UNa-Veh); data were normalized to creatinine

excretion. Data are the means ±SEM; *P < 0.05 ([162]).

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 52

5.4 CsA-Induced Activation of NKCC2 Depends on AVP

5.4.1 CsA Stimulates NCC but not NKCC2 in AVP-Deficient Brattle-

boro Rats

To study whether AVP plays a role in CNI-induced activation of NKCC2 and NCC, AVP-

deficient Brattleboro rats were treated with CsA or vehicle for 4 h and immunoblot profiles

of the distal transporters were compared between the groups. CsA treatment stimulated

NCC but not NKCC2 function, as demonstrated by significantly increased levels of pNCC

but no change in pNKCC2 or the unphosphorylated species of the two transporters (+68

% for pNCC, P < 0.05; Figure 5.15). These data suggest that CsA-induced activation of

NKCC2 requires AVP signaling whereas activation of NCC occurs independently of AVP

levels.

pNKCC2

pNCC

GAPDH

4 h

A B

Veh CsA

160 kDa

40 kDa

160 kDa

40 kDa

Figure 5.15. Short-term Effects of CsA on NKCC2 and NCC Phosphorylation in Vasopressin

Deficient Brattleboro Rats (A) Representative immunoblots of kidney lysates from vehicle- (Veh)

and cyclosporine A- (CsA, 4 h) treated Brattleboro rats showing immunoreactive signals for pNKCC2

and pNCC (both approximately 160 kDa); GAPDH served as loading control (approximately 40 kDa).

(B) Densitometric evaluation of immunoreactive signals normalized to loading controls. Short-term CsA

treatment increased levels of pNCC but not pNKCC2. Data are the means ±SEM; *P < 0.05 ([162]).

5.4.2 AVP is Required for CsA-Induced Activation of NKCC2 in vitro

To prove the hypothesis that CsA-induced activation of NKCC2 and NCC differs in their

dependence on AVP, we next studied the local effects of calcineurin inhibition in cultured

epithelial cells of the distal tubule. For this, mDCT and raTAL cells were each stimulated

with CsA or vehicle for 4 h and immunoblots were performed to analyze NKCC2, NCC

and SPAK/OSR1. In mDCT cells, 4 h of CsA strongly increased levels of phosphorylated

NCC (+375 % for pNCC, P < 0.05) and SPAK/OSR1 (+68 % for pS-SPAK/OSR1,

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 53

P < 0.05; Figure 5.16 A, C). In contrast, CsA did not induce any significant changes in

NKCC2 or SPAK/OSR1 phosphorylation in raTAL cells (Figure 5.16 B, C).

pNCC

GAPDH

pNKCC2

Actin

pS-SPAK/OSR1

raTAL cells

mDCT cells

Veh CsA

pS-SPAK/OSR1

mDCT raTAL

160 kDa

40 kDa

60 kDa

40 kDa

160 kDa

60 kDa

40 kDa

4 h 4 h

Figure 5.16. CsA Activates NCC but not NKCC2 in vitro (A) Representative immunoblots of cell

lysates from vehicle- (Veh) and cyclosporine A- (CsA, 4 h) treated mDCT cells showing immunoreactive

signals for pNCC (approximately 160 kDa) and pS-SPAK/OSR1 (approximately 60 kDa); GAPDH served

as loading control (approximately 40 kDa). (B) Representative immunoblots of cell lysates from Veh-

and CsA-treated raTAL cells showing immunoreactive signals for pNKCC2 (approximately 160 kDa) and

pS-SPAK/OSR1 (approximately 60 kDa); β-actin served as loading control (approximately 40 kDa). (C)

Densitometric evaluation of immunoreactive signals normalized to loading controls. mDCT: mouse distal

convoluted tubule cells, raTAL: rat thick ascending limb cells. Data are the means ±SEM; *P < 0.05

([162]).

To provide further support for the idea that CsA-induced activation of NKCC2 requires

AVP signaling, we added the AVP receptor 2 agonist DDAVP to the treatment proto-

col. DDAVP-stimulated raTAL cells indeed displayed markedly increased phosphoryla-

tion levels of NKCC2 (+46 % for DDAVP and +97 % for DDAVP+CsA, P < 0.05) and

SPAK/OSR1 (+65 % for DDAVP, not significant; +92 % for DDAVP+CsA, P < 0.05;

Results – Part II: Local and Systemic Effects of Calcineurin Inhibition 54

Figure 5.17 A, B). In sum, these results clearly illustrate that local calcineurin inhibi-

tion plays a dominant role in the DCT, regulating NCC function. However, compelling

evidence is provided for a permissive role of AVP in the TAL, regulating NKCC2 function.

raTAL cells

Veh dDAVP dDAVP+CsA

pNKCC2

GAPDH

pS-SPAK/OSR1

160 kDa

60 kDa

40 kDa

GAPDH

40 kDa

Figure 5.17. DDAVP Stimulates NKCC2 and SPAK/OSR1 Phosphorylation in Cultured

TAL Cells (A) Representative immunoblots of cell lysates from vehicle- (Veh), desmopressin- (DDAVP),

and DDAVP+CsA-treated TAL cells showing immunoreactive signals for pNKCC2 (approximately 160

kDa) and pS-SPAK/OSR1 (approximately 60 kDa); GAPDH served as loading control (approximately 40

kDa). (B) Densitometric evaluation of immunoreactive signals normalized to loading controls. Data are

the means ±SEM; *P < 0.05 for DDAVP vs. Veh or DDAVP+CsA vs. Veh, #P < 0.05 for DDAVP vs.

DDAVP+CsA ([162]).

Chapter 6

Results – Part III: Key Role of

NKCC2 in Blood Pressure

Regulation

In a collaborative work our group has previously shown that NCC knockout (NCC KO)

mice are protected from developing hypertension when treated with the calcineurin in-

hibitor tacrolimus. That work demonstrated that tacrolimus only stimulates NCC activity

and that it has no effect on NKCC2 function. In the present study, we provide further

evidence that CsA treatment, in contrast, activates both, NKCC2 and NCC, and we

demonstrate that these effects are partially mediated via systemic signaling. Based on

these findings we speculated that NKCC2 plays a major role in CsA-induced hypertension.

Thus, we hypothesized that CsA treatment, unlike tacrolimus, will induce hypertension

regardless of NCC activity. Here, we used the NCC KO mouse model in order to further

define the role of NKCC2 and evaluate the effects of CsA in the absence of NCC activity.

Mice were treated with CsA and mean arterial blood pressure was monitored over a period

of 12 days. Protein expression was evaluated at end point. With this we aimed to gain

further insight into the differential effects of the two CNIs tacrolimus and CsA on salt

homeostasis and blood pressure.

Results – Part III: Key Role of NKCC2 in Blood Pressure Regulation 56

6.1 Effects of CsA Treatment in NCC Knockout Mice - A

Pilot Study

To assess the effects of CsA treatment on blood pressure in NCC KO mice, animals were

injected with CsA for 12 days. The well-established radio telemetry method was used to

measure blood pressure in unanesthetized mice (see section 3.11). Three days of baseline

levels were recorded while a vehicle solution was applied to the mice. Unfortunately, due

to technical problems the experiment was finished with only n= 2 mice per group and,

thus, the results may only give an idea of the actual effects.

By day 5 of CsA treatment, wildtype mice had developed a marked increase in blood

pressure (109 mmHG: mean baseline level; 133 mmHg: CsA day 5) and levels remained

high throughout the rest of the experiment. NCC knockout mice, despite high variations

within the group, also displayed a marked albeit slower increase of blood pressure. By day

6, NCC KO mice had developed high blood pressure (105 mmHG: mean baseline level;

133 mmHG: CsA day 6) and levels remained at a fairly high level throughout the rest of

the experiment (Figure 6.1).

These data, although without statistical significance, indicate that CsA might induce

high blood pressure regardless of NCC activity, albeit with a certain delay. These results

corroborate the hypothesis that NKCC2 plays a key role in CsA-induced hypertension.

MAP

Figure 6.1. CsA Induces Hypertension in NCC KO Mice Mean arterial blood pressure (MAP)

in CsA-treated wildtype (WT) and NCC knockout (NCC KO) mice measured by radio telemetry. CsA

treatment induced hypertension in WT mice shortly after the beginning of treatment. NCC KO mice also

developed hypertension, albeit with a certain delay. Data are the means ±SEM.

Results – Part III: Key Role of NKCC2 in Blood Pressure Regulation 57

In order to provide evidence for the NKCC2 activation, we sought to assess NKCC2 phos-

phorylation levels in CsA-treated NCC KO mice. To this aim, we analysed immunofluo-

rescent signals in 20 representative TAL motifs in kidneys of WT and NCC KO mice. In

both, WT and NCC KO mice, a marked CsA-induced increase of NKCC2 phosphoryla-

tion was detected. The increase was slightly stronger in NCC KO mice compared to their

WT littermates (Figure 6.2). Since this part of the project was performed during a three

months research stay in David Ellison’s laboratory at the Oregon Health and Science

University, US, the number of available animals was limited. Due to the initial problems

with the telemetry method described above, we could only proceed with a small number

of animals per group throughout the rest of the study. Proper statistical evaluation was

not possible for the immunofluorescent data. However, it is tempting to conclude that

the CsA-induced rise in blood pressure observed in NCC KO mice is a result of the strong

upregulation of NKCC2 activity, possibly via the induction of AVP signaling.

Results – Part III: Key Role of NKCC2 in Blood Pressure Regulation 58

pNKCC2

CsAVeh

WTNCC KO

pNKCC2

TAL TAL

TAL

pNKCC2

Figure 6.2. Effects of CsA Treatment on NKCC2 Phosphorylation in WT and NCC KO

Mice (A) Representative images of kidney sections from vehicle- (Veh) or cyclosporine A (CsA)-treated

wildtype (WT) and NCC knockout (NCC KO) mice showing immunofluorescent labeling of phosphorylated

NKCC2 in the thick ascending limb (TAL). (B) Graphs showing pNKCC2 signal in TAL of WT and NCC

KO mice treated with Veh or CsA as evaluated using ZEN and ImageJ software. pNKCC2 signal was

normalized to background signal. CsA treatment increased the abundance of pNKCC2 in both groups,

but to a stronger extent in NCC KO mice. Data are the means ±SEM. Bars indicate 10 µm.

6.2 Stimulated Salt Reabsorption Cascade in the TAL of

NCC Knockout Mice

Based on these findings we hypothesized that NCC KO mice possess a compensatory

mechanism that upregulates NKCC2 function. To gain further insight into the role of

Results – Part III: Key Role of NKCC2 in Blood Pressure Regulation 59

NKCC2, in a new experiment we assessed baseline protein expression and phosphory-

lation levels of NKCC2 in untreated wildtype and NCC knockout mice. Immunoblots

revealed similar levels of NKCC2 in both groups. Interestingly, phosphorylation levels of

NKCC2 were found to be significantly higher in NCC knockout mice compared to wild-

type animals (Figure 6.3). Since NCC function is minimized in NCC KO mice, the high

levels of phosphorylated NKCC2 might indeed reflect a compensatory mechanism in order

to upregulate NKCC2 activity for adjustment of salt homeostasis. Such mechanism would

explain that NCC knockout mice display a steady state blood pressure only slightly lower

than wildtype mice (Figure 6.1) and develop hypertension on the CsA protocol.

NKCC2

GAPDH

pNKCC2

160 kDa

40 kDa

160 kDa

WT NCC KO *

A B

Figure 6.3. Baseline Activity of NKCC2 Is Higher in NCC KO Mice Representative immunoblots

of kidney lysates from wildtype (WT) and NCC knockout (NCC KO) mice showing (A) immunoreactive

signals for total NKCC2 and phosphorylated NKCC2 (pNKCC2), both at approximately 160 kDa; GAPDH

served as loading control (approximately 40 kDa). (B) Graphs showing respective densitometric evaluation

of immunoreactive signals normalized to loading controls. NCC KO mice have higher baseline pNKCC2

levels than their WT littermates, but total protein abundance is not different in the two groups. Data are

the means ±SEM; *P < 0.05.

In order to investigate whether the SPAK kinase is responsible for NKCC2 activation,

we next analyzed SPAK expression and phosphorylation in WT and NCC KO mice.

Immunoblot analysis revealed that SPAK protein expression is similar in both groups.

Surprisingly, NCC KO mice displayed slightly lower pS-SPAK levels, albeit not significant

(Figure 6.4).

Results – Part III: Key Role of NKCC2 in Blood Pressure Regulation 60

SPAK

GAPDH

pS-SPAK/OSR1

60 kDa

40 kDa

60 kDa

WT NCC KO

A B

Figure 6.4. Similar Baseline Activity of SPAK in WT and NCC KO Mice Representative

immunoblots of kidney lysates from wildtype (WT) and NCC knockout (NCC KO) mice showing (A)

immunoreactive signals for total SPAK and phosphorylated regulatory pS-SPAK/OSR1, both at approxi-

mately 60 kDa; GAPDH served as loading control (approximately 40 kDa). (B) Graphs showing respective

densitometric evaluation of immunoreactive signals normalized to loading controls. WT and NCC KO mice

have similar baseline SPAK levels, pS-SPAK/OSR1 trended to be lower in NCC KO mice than in WT.

Data are the means ±SEM.

With respect to the increased pNKCC2 levels in NCC KO mice we had expected up-

regulation of SPAK phoshphorylation. To gain better understanding of the mechanism

responsible for NKCC2 activation, we analyzed TAL and DCT motifs in WT and NCC

KO kidneys by immunofluorescent labeling of the two SPAK/OSR1 phosphorylation sites.

Interestingly, we found that kidneys of the NCC KO mice lack the apical signal of the

regulatory pS-SPAK/OSR1 form in their DCTs. Instead, in the DCT only cytosolic pS-

SPAK/OSR1 was detected (Figure 6.5) while WT mice displayed normal pS-SPAK/OSR1

distribution (apical in TAL, apical and cytosolic in DCT). The missing apical expression

of regulatory SPAK/OSR1 might reflect a deficiency of SPAK function in NCC KO mice

due to NCC depletion.

Results – Part III: Key Role of NKCC2 in Blood Pressure Regulation 61

Figure 6.5. Expression Profile of Regulatory pS-SPAK/OSR1 in WT and NCC KO Mice

Representative images of kidney sections from wildtype (WT) and NCC knockout (NCC KO) mice showing

immunofluorescent labeling of phosphorylated regulatory SPAK/OSR1 domain (pS-SPAK/OSR1) in the

thick ascending limb (TAL) and the distal convoluted tubule (DCT). In WT mice, pS-SPAK/OSR1 shows

apical expression in the TAL, and apical and cytosolic expression in the DCT. NCC KO mice lack the

apical expression in the DCT. Bars indicate 10 µm.

For the catalytic pT-SPAK/OSR1 form, we saw a reversed picture in NCC KO mice

compared to WT mice. pT-SPAK/OSR1, which in wildtype mice shows strong apical

expression in the DCT and mild apical expression in the TAL, could not be detected in

the DCT of NCC KO mice at all. However, in NCC KO mice, pT-SPAK/OSR1 was found

to be expressed in the apical membrane of the TAL (Figure 6.6). These data indicate that

the SPAK kinase is not fully functional in NCC KO mice but compensates this lack of

function through enhanced activity in the TAL, thereby balancing NKCC2 function and

salt transport.

Results – Part III: Key Role of NKCC2 in Blood Pressure Regulation 62

Figure 6.6. Expression Profile of Catalytic pT-SPAK/OSR1 in WT and NCC KO Mice

Representative images of kidney sections from wildtype (WT) and NCC knockout (NCC KO) mice showing

immunofluorescent labeling of phosphorylated catalytic SPAK/OSR1 domain (pT-SPAK/OSR1) in the

thick ascending limb (TAL) and the distal convoluted tubule (DCT). In WT mice, pT-SPAK/OSR1 shows

strong apical expression in the DCT and lower apical expression in the TAL. NCC KO mice lack the apical

expression in the DCT, but display apical expression in the TAL. Bars indicate 10 µm.

Taken together, these data underline the significance of NKCC2 in the regulation of

blood pressure and its involvement in the development of hypertension. They further

corroborate the hypothesis of a potentiating effect of CsA-induced AVP signaling on

NKCC2 activation. Additionally, SPAK expression and function seem to adapt to distal

salt transport activity and to be able to compensate for deteriorated resorption in the

DCT.

Chapter 7

Discussion

Since their market launch over 20 years ago, the calcineurin inhibitors (CNIs) cyclosporine

A and tacrolimus have been increasingly used for immunosuppression after organ trans-

plantation and in the treatment of several autoimmune diseases. Due to the great treat-

ment success and significant prolongation of patient survival, long-term treatment with

calcineurin inhibitors (CNIs) became a standard treatment protocol ([1], [2]). However,

severe side effects associated with chronic CNI administration complicate their use. The

kidney, the essential organ for blood pressure regulation, is most affected by CNI treat-

ment. One of the early symptoms occurring in a majority of patients is elevated blood

pressure that often manifests as hypertension. Besides, the most prevalent adverse effect

of chronic CNI treatment is renal damage ([7], [8]).

The kidney controls blood pressure via regulation of water and electrolyte homeostasis.

While filtering the whole body blood volume, the kidney precisely coordinates tubular

reabsorption and secretion processes to permanently ensure a constant blood filtration

rate ([38],[36], [35]). Since Na+ions are the predominant ions in the interstitium, reab-

sorption of sodium is essential for blood pressure regulation. Increased sodium uptake

stimulates water uptake which increases extracellular fluid volume and finally enhances

blood pressure. The distal tubule herein plays a key role as a substantial amount (up to

35 %) of sodium is reabsorbed in the TAL and the DCT ([34], [37], [36], [51]).

Regarding CNI administration, the distal tubule becomes particularly interesting as CNIs

differentially affect the two major salt transporters, NKCC2 and NCC, expressed in this

Discussion 64

part of the nephron. Acute and chronic CNI treatment is associated with activation of

NKCC2 and NCC, and activation of the transporters can lead to hypertension at long-

term. Previous studies suggested that tacrolimus treatment only stimulates NCC, whereas

CsA treatment affects both NKCC2 and NCC function ([68],[18], [17], [140], [19], [129]).

CsA and tacrolimus also differ in their systemic effects and, thus, have distinct side effects.

Therefore, both drugs are indispensable for individual patient care and various disease

patterns ([138], [139]).

Unraveling the mechanism of how NKCC2 and NCC are regulated will thus be of high

clinical value for the improvement of blood pressure control upon CNI treatment. Since

not only CNIs but certain other drugs as well as hereditary diseases and life-style factors

can induce hypertension, anti-hypertensive treatment has become increasingly important

([22], [23], [24]). Therefore, elucidating the regulation of the distal salt transporters can

further contribute to a better understanding of the development of high blood pressure

in general.

In this study we provide new insights into the effects of CsA on salt handling in the distal

nephron. While activation of NKCC2 and NCC is regulated by a kinase cascade compris-

ing members of the WNK family and the two homologous SPAK and OSR1 kinases ([68],

[23]), it has been suggested that the transporters and the kinases are also regulated by

the calcineurin phosphatase since calcineurin inhibition affects NKCC2 and NCC func-

tion. Thus, we hypothesized that differential expression of the key factors regulating

calcineurin inhibition might be the key for the distinct effects of CsA and tacrolimus in

the distal nephron. To investigate this, we first localized the renal isoforms of the catalytic

subunit of calcineurin, CnAαand CnAβ, and showed that both isoforms are ubiquitously

expressed along the nephron. Other groups suggested that CnAαis primarily expressed

in the renal cortex while CnAβis the major isoform of the medullary TAL ([160], [123],

[15]). In this study, we could not detect any significant differences in the expression of the

two isoforms along the nephron on either the mRNA or the protein level. These results

indicate that the distinct effects of CsA and tacrolimus are not mediated by differential

expression of calcineurin isoforms. Instead, differences in post-translational regulation or

substrate specificity might be responsible for different effects of calcineurin inhibition. In

fact, it has been postulated by Sheftic et al. ([133]) that two recognition sites detected

in calcineurin substrates, the PxIxT motif and the LxVP motif, regulate the binding

Discussion 65

affinity of calcineurin substrates ([130], [131], [132]). But neither the two transporters

NKCC2 and NCC nor their upstream regulating kinases possess these binding motifs.

However, when co-immunoprecipitating, they show multiple interactions suggesting scaf-

folding mechanisms between calcineurin and its substrates ([68], [105]). For example, a

recent study of our group revealed that SORLA (sorting-related receptor with A-type

repeats), an intracellular receptor involved in sorting and trafficking of various proteins,

interacts with CnAβand modulates dephosphorylation and activation of NKCC2 ([132],

[68]). Thus, substrate specificity of calcineurin isoforms as a reason for the differential

effects of CNIs cannot be ruled out.

We next hypothesized that the CNI-binding proteins, the immunophilins, which are es-

sential for calcineurin inhibition, are differentially expressed along the nephron. Since

tacrolimus treatment only affects NCC function, it was very tempting to suggest that the

tacrolimus-binding protein Fkbp12 is expressed primarily in the DCT. Analogically, we

presumed to detect the CsA-binding proteins CypA and CypB in both, TAL and DCT,

since CsA treatment is associated with activation of NKCC2 and NCC. Moreover, a previ-

ous study revealed that renal deletion of Fkbp12 in mice abolishes the effects of tacrolimus

on NCC phosphorylation ([164]). However, we found no significant differences in CypA,

CypB and Fkbp12 expression on the mRNA or protein level. Thus, we concluded that

other mechanisms must be the key to the different CNI sensitivities of the TAL and the

DCT. Finally, we analyzed expression patterns of two proteins involved in WNK ubiquity-

lation, KLHL3 and CUL3. Since mutations in KLHL3 and CUL3 cause hypertension via

activation of NCC, a phenotype that resembles the renal effects of tacrolimus treatment,

we suggested that these proteins are majorly expressed in the DCT. However, we found

strong expression of KLHL3 in both the TAL and the DCT and lower but ubiquitous

expression of CUL3 along the entire nephron. In sum, these data provide compelling

evidence that the distinct effects of CsA and tacrolimus are not mediated by differential

expression of key factors within the signaling pathways regulating NKCC2 and NCC.

Thus, we pursued a different approach to unravel the mechanisms behind the distinct

effects of CNIs. Various studies suggest that tacrolimus-induced hypertension and NCC

activation primarily rely on local effects of calcineurin inhibition while CsA-induced ele-

vation of blood pressure may partially be caused by systemic effects. To assess local and

Discussion 66

systemic effects of CsA treatment we analyzed the effects of acute and chronic treatment

protocols in rat TAL and DCT. Acute treatment served for evaluation of fast changes on

the post-translational level such as phosphorylation and trafficking as well as potential

early changes in mRNA expression and protein abundance. Chronic treatment served to

evaluate the impact of endocrine/paracrine mechanims on electrolyte handling and blood

pressure. With regard to the local effects of calcineurin inhibition in distal nephron cells,

our findings in the short- and long-term treated rats corroborate and extend previous

reports on the role of post-translational modification of NKCC2 and NCC by phosphory-

lation ([68], [77]). Conflicting data exist regarding the effects of CNI treatment on protein

content of the kinases and the salt transporters in animals and cultured cells ([17], [18],

[140], [165], [166], [19]). However, with the treatment protocols used in this study, no

substantial changes in total protein abundance or mRNA levels of the transporters or

WNK and SPAK/OSR1 were detected. Additional evidence for the activated state of

the two transporters is provided by the parallel increase in surface expression upon CsA

treatment. Previous studies hypothesize that phosphorylation stabilizes the transporters

within the plasma membrane ([77], [84], [85]). Our data support this idea of a correlation

of phosphorylation and trafficking processes. Since we observed that the initial effects of

CsA treatment were increased phosphorylation levels of the transporters followed by in-

creased surface expression after 4 h of CsA, we conclude that phosphorylation of NKCC2

and NCC might stimulate trafficking in order to augment transport function.

Taken together, these results highlight the role of post-translational regulation of NKCC2

and NCC by phosphorylation/dephosphorylation reactions. Our data indicate that in-

creased levels of phosphorylated NKCC2 and NCC are, at least in part, a result of the

activation of the WNK-SPAK/OSR1 kinase cascade. Additionally, local inhibition of the

calcineurin phosphatase might play a role in the regulation of NKCC2 and NCC activation

([167], [168]). However, it remains to be clarified how calcineurin binds to the distal salt

transporters. It is known that substrates of calcineurin possess recognition sites for cal-

cineurin binding, the PxIxT and the LxVP motif, but these sites have not been detected

in NKCC2 and NCC nor in the WNK-SPAK/OSR1 kinases. Thus, they might interact

with the calcineurin phosphatase via scaffolding proteins ([68], [105]). Additionally, other

phosphatases like the protein phosphatase 1 may be involved in the regulation of NKCC2

and NCC and mediate the effects of calcineurin ([168]).

Apart from the local effects in the renal distal tubule cells, CNIs reportedly stimulate

Discussion 67

endocrine and paracrine mechanisms as well as renal sympathetic innervation and may

thereby facilitate renal salt reabsorption ([145], [21], [169]). In our rat model, CsA treat-

ment clearly increased renin expression and activity, likely reflecting enhanced sympa-

thetic activity and stimulated renin synthesis in juxtaglomerular cells ([21]). Activation

of the renin-angiotensin-aldosterone-system can be induced by COX-2 secretion which

stimulates the local prostaglandins pathway and subsequently activates renin expression.

This leads to upregulation of the glomerular filtration rate and NKCC2 function ([148],

[149]). However, COX-2 expression was suppressed in CsA-treated animals, indicating

that stimulation of renin activity was unrelated to paracrine mechanisms within the jux-

taglomerular apparatus. Since transcription of COX-2 is controlled by the nuclear factor of

activated T cells (NFAT), which remains inactive under calcineurin inhibition conditions,

downregulation of COX-2 upon CsA treatment was an expected observation ([145], [150]).

While the effects of renal COX-2 suppression on salt homesostasis are complex ([5]), global

inhibition of COX-2 can result in salt-sensitive hypertension ([170]). Thus, synergistic ef-

fects of a stimulated RAAS and inhibition of COX-2 may enhance salt reabsorption along

the distal nephron and thereby contribute to the development of CNI-induced hyperten-

sion ([16], [170]).

Another principal component of the endocrine control of NKCC2 and NCC is arginine va-

sopressin (AVP) ([100], [101], [55]). While the vasopressin V2 receptor (V2R) is reportedly

expressed along the entire distal nephron ([171], [172], [84]), the impact of AVP signal-

ing on salt reabsorption processes might be different in the TAL and the DCT ([173]).

In fact, kidneys of AVP-deficient Brattleboro rats display almost complete absence of

phosphorylated NKCC2, whereas levels of phosphorylated NCC are less affected in this

model ([84], [20]). In our study, short-term CsA treatment induced an increase of NCC

phosphorylation in kidneys of Brattleboro rats but did not stimulate NKCC2, suggesting

that activation of NKCC2 depends on AVP signaling. Our in vitro studies on cultured

distal nephron cells, where systemic effects can be excluded, support these results. Herein,

CsA induced activation of NCC and SPAK in cultured mDCT cells but did not change

NKCC2 and SPAK function in raTAL cells. These findings support previous suggestions

that particularly the TAL strongly depends on the presence of AVP ([102]). This is fur-

ther substantiated by one of our recent studies showing that rats with a segment-specific

overexpression of a dominant-negative V2R mutant exhibit substantial deficits in NKCC2

Discussion 68

phosphorylation and function ([172]), underlining the major role of the TAL in the uri-

nary concentration process. Overall, these data provide compelling evidence that AVP

is permissive for CsA-induced activation of NKCC2, while local calcineurin inhibition is

sufficient for the regulation of NCC.

The molecular mechanism underlying the stimulatory effects of AVP on NKCC2 and

NCC function are still unclear, but several studies suggest facilitating effects of AVP on

relevant kinases such as WNK-SPAK/OSR1 kinases, protein kinase A (PKA), or AMP-

activated protein kinase ([77], [55], [105]). Stimulation of basal SPAK/OSR1 activity

may be achieved via activation of PKA through induction of intracellular cAMP release

([103], [104], [105]). Furthermore, it has been increasingly recognized that sensing of in-

tracellular chloride concentration is a major function of WNK kinases. For instance, it

has been demonstrated that low intracellular chloride concentration stimulates WNK to

activate NCC ([174]). Besides, in a previous study we have shown that uromodulin, a pro-

tein expressed in the TAL and potentially involved in urinary concentrating mechanism,

contributes to AVP-induced NKCC2 phosphorylation via effects on intracellular chloride

concentration ([175], [176]). It is therefore tempting to speculate that AVP either affects

chloride concentration or modulates chloride sensitivity of WNK-SPAK/OSR1 kinases in

TAL and DCT cells. Therefore, decreased function of NKCC2-activating kinases in the

absence of AVP may explain that CsA treatment fails to enhance NKCC2 phosphoryla-

tion in Brattleboro rats or cultured raTAL cells lacking AVP.

To confirm the essential role of AVP in NKCC2 activation, we added DDAVP to the CsA-

treated raTAL and mDCT cells and found clear increases of phosphorylation levels of

SPAK/OSR1 and NKCC2 in raTAL cells. Thus, in the presence of AVP, CsA treatment

induces activation of NKCC2 and might thereby contribute to salt retention occuring at

the long-term.

Taken together, these data confirm that in DCT cells local calcineurin inhibition is suf-

ficient to induce NCC activation, whereas in the TAL, stimulation by AVP appears to

be indispensable for activation of NKCC2. While stimulatory effects of AVP signaling

have been described for both, TAL and DCT, the different sensitivities of the two seg-

ments are still unclear, but the distal nephron seems to adapt to AVP depending on the

Discussion 69

treatment protocol ([55], [105], [20]). Direct and indirect effects of AVP on the distal salt

transporters cannot be ruled out but remain to be elucidated.

Our analysis highlights that NKCC2 function depends on systemic signaling. Since the

systemic effects associated with calcineurin inhibitors have been reported to be more pro-

nounced with CsA treatment than with tacrolimus and tacrolimus being less nephrotoxic

than CsA ([145], [138], [139]), we suggested that activation of NKCC2 is a key mechanism

in CsA-induced hypertension. CsA-induced activation of systemic factors may explain

the differential effects of CsA and tacrolimus in the distal nephron. Thus, we assumed

that CsA treatment, unlike tacrolimus, enhances blood pressure in NCC KO mice. In a

collaborative study we have previously applied the NCC KO model and therein we were

able to show that tacrolimus-induced hypertension is chiefly mediated by NCC activation

([19]). In that study, NCC KO mice were protected from developing high blood pressure

upon tacrolimus treatment. In contrast, in a pilot study we detected an increase in blood

pressure in CsA-treated NCC KO mice by day 6 of treatment. These results suggest

that CsA induces hypertension regardless of NCC activity, supporting the pivotal role of

NKCC2 in CsA-induced hypertension. In line with this, NCC KO mice displayed higher

baseline levels of phosphorylated NKCC2 and the CsA-induced increase of pNKCC2 was

stronger than in the wildtype group. The upregulation of NKCC2 activity in NCC KO

mice may likely reflect a compensatory mechanism counterbalancing the lack of NCC

function. However, due to limited animal numbers these data were not statistically sig-

nificant.

Compensatory reactions of the SPAK kinase have been reported previously ([177], [105]).

The findings of the present study further point to the ability of SPAK to compensate

for imbalanced salt transport in the distal nephron. While total SPAK protein levels

were similar in WT and NCC KO mice, apical expression of regulatory pS-SPAK/OSR1

and catalytic pT-SPAK/OSR1 in the DCT of NCC KO mice was below detection lev-

els, likely reflecting a diminished demand for NCC activating kinases. Instead, enhanced

apical expression of phosphorylated SPAK kinase was detected in the TAL of NCC KO

mice suggesting that SPAK might balance NKCC2 function as well as salt transport in

the TAL of NCC KO mice. Since in NCC KO mice baseline blood pressure was slightly

lower and CsA-induced increase of blood pressure was delayed compared to their wildtype

littermates, SPAK in the TAL seemed not to be able to fully compensate for the lack of

NCC.

Discussion 70

Notably, NCC KO mice resemble many features of SPAK KO mice where a tendency to-

ward salt wasting is attenuated by activation of NKCC2 ([177]). In line with our findings

on AVP signaling and the fact that compensatory effects in the kidney are commonly

stimulated by endocrine factors ([147], [148], [149], [106]), the upregulation of NKCC2

might be triggered by AVP. Alternative kinase pathways such as PKA might mediate the

effects of AVP ([69], [57])

In sum, the data of the present study underline the key role of NKCC2 in CsA-induced

hypertension and expand the understanding of the involvement of AVP signaling in this

context.

Chapter 8

Conclusion

In this study, we present a novel mechanism that might be key to the more diverse effects of

CsA treatment compared to tacrolimus. We succeeded in localizing all major components

within the signaling pathways regulating the function of the renal distal salt transporters.

Our analysis indicates that the distinct effects of the two CNIs CsA and tacrolimus are

not mediated via differential expression of those components. We further show that CsA-

induced activation of NKCC2 and NCC chiefly occurs at the post-translational level via

enhanced phosphorylation of the transporters and their activating kinases. With the use

of an AVP-deficient rat model and in vitro experiments on cultured distal tubule cells

we were able to show that CsA-induced activation of NKCC2, unlike NCC, depends on

AVP signaling. These results provide compelling evidence that NKCC2 plays a pivotal

role in the development of hypertension upon CsA treatment. In fact, though in a small

group of animals, we demonstrate that CsA induces high blood pressure regardless of

NCC activity, thus underlining the strong impact of NKCC2 function in blood pressure

regulation. Based on these findings, we conclude that differential effects of CsA and

tacrolimus may be mediated via AVP signaling in the TAL.

In sum, this study provides a better understanding of the mechanisms of CNI-induced

hypertension and may help clinical therapeutic strategy.

Chapter 9

Perspectives

The present study provides new insights into the mechanism of CsA-induced renal side

effects. Based on previous findings on the effects of CNIs, our data suggest to favour the

use of tacrolimus in certain situations. Even though CsA is more frequently used in the

post-transplantational therapy, the use of tacrolimus gradually increases and alternative

immunosuppressant drugs are also increasingly prescribed ([4]). Regarding safety and

efficacy, tacrolimus is superior to CsA exhibiting less renal complications with less need

for antihypertensive treatment ([139]). However, CsA treatment is associated with less

post-transplant diabetes and reduced severity of liver fibrosis ([178]). Thus, a conversion

to CsA or other immunosuppressive drugs can be appropriate in some cases and might be

advantageous especially for lower risk patients ([138]). Our data further support the use of

furosemide diuretics and suggest AVP blocker, V2R antagonists or other substances that

specifically inhibit NKCC2 function to control blood pressure in CsA-treated patients.

A previous study which focused on the effects of tacrolimus suggested that the toxic side

effects in the kidney arise from local calcineurin inhibition ([164]). Our findings also point

to the involvement of the calcineurin phosphatase in the regulation of salt transport and

thus blood pressure control, at least in the DCT. To reduce calcineurin inhibitor toxicity,

we envision novel agents that directly target immune cells or specific calcineurin isoforms.

Future experiments should grant a better insight as to whether different effects of CNIs

additionally arise from substrate specific binding of calcineurin.

Altogether, this study can help improve blood pressure control and may contribute to a

Perspectives 73

better quality of life for patients who depend on successive immunosuppressive therapy.

Supplementary Tables 75

Appendix A

Supplementary Tables

A.1 Primary Antibodies

Antigen Species Source of Supply

β-actin mouse Sigma Aldrich

CnAαrabbit Millipore

CnAβrabbit Millipore

CUL3 rabbit provided by D. H. Ellison, US

CypA rabbit Abcam

CypB rabbit Abcam

COX-2 goat Santa Cruz Biotechnology

Fkbp12 rabbit Abcam

GAPDH rabbit Santa Cruz Biotechnology

KLHL3 rabbit provided by D. H. Ellison, US

NCC rabbit provided by D. H. Ellison, US

NKCC2 guinea pig [153]

OSR1 sheep University of Dundee, UK

phospho-S71-NCC rabbit Pineda Antik¨orper-Service

phospho-T96/T101-NKCC2 rabbit [77]

phospho-S383-SPAK/pS325-OSR1 sheep University of Dundee, UK

phospho-T243-SPAK/pT185-OSR1 sheep University of Dundee, UK

phospho-S382-WNK1 sheep University of Dundee, UK

SPAK rabbit [177], US

WNK4 rabbit provided by J. A. McCormick, US

Table A.1. Validated Primary Antibodies Used in this Study Listed are all antibodies used for

immunoblotting, immunofluorescence and in situ hybridization

Supplementary Tables 76

A.2 Secondary Antibodies

Species Source of Supply

DyLight488 donkey anti rabbit IgG DIANOVA, Germany

Cy2 donkey anti guinea pig IgG DIANOVA, Germany

Cy3 donkey anti rabbit IgG DIANOVA, Germany

Cy3 donkey anti goat IgG DIANOVA, Germany

Cy3 donkey anti sheep IgG DIANOVA, Germany

Cy3 donkey anti guinea pig IgG DIANOVA, Germany

Cy5 donkey anti Polyclonal swine anti rabbit IgG/HRP Dako, Denmark

Goat anti mouse IgG/HRP Dako, Denmark

Polyclonal swine anti guinea pig IgG/HRP Dako, Denmark

Rabbit anti sheep IgG/HRP Dako, Denmark

Polyclonal rabbit anti goat IgG/HRP Dako, Denmark

Anti-DIG-alkaline phosphatase-conjugated antibody Dako, Denmark

Table A.2. Validated Secondary Antibodies Used in this Study Listed are all antibodies used for

immunoblotting, immunofluorescence and in situ hybridization

Supplementary Tables 77

A.3 qPCR Primers

Gene Forward Primer Reverse Primer

GAPDH TGCACCACCAACTGCTTAGC GGCATGGACTGTGGTCATGAG

β-actin ACGGCCAGGTCATCACTATT GCCACAGGATTCCATACCCA

CnAαCCAACACTCGCTACCTCTTC GTGCCTACATTCATGGTTTCC

CnAβGCAACCATGAATGCAGACACC CAAGGGGCAAGCTGTCAAAAG

CypA CTGGACCAAACACAAACGGT TGCCTTCTTTCACCTTCCCA

CypB GCACAGGAGGAAAGAGCATC TGAGCCATTGGTGTCTTTGC

KLHL3 GCCATGAAGTACCACCTCCT ACCAACCACAATCATGACCTTG

CUL3 GCACCATGTCGAATCTGAGC TTCATCCATGGTCATCGGAAA

WNK1 AAGTATGCCTCAGTCCGTGG ACTTTCGGTGGACAGGTAGG

WNK3 GAGCTACAGGACCGCAAATTA TCGAACTATATTGGGATGCTGGA

SPAK TGCCAGACGAGTATGGATGA CCACAGCTCATCTTTGACCA

OSR1 AAAGACGTTTGTTGGCACCC GCCCCTGTGGCTAGTTCAAT

NKCC2 TGTGAAGTTTGGATGGGT CCGCTTCTCCTACAATCC

NCC TGATCATCCTTACCTTGCCCA ACGTTCTCCTGGTTACCTCG

COX-2 TGACAGCCCACCAACTTACA TCCTTATTTCCTTTCACACCCA

Table A.3. qPCR Primers Used in this Study Listed are all primers used for quantitative PCR

A.4 Primers for In Situ Hybridization

Gene Primer

CypA Forward AGTTCCAAAGACAGCAGAAAACT

CypA Reverse TAATACGACTCACTATAGGATGCCAGGACCTGTATGCTT

CypB Forward ACAGGAGAGAAAGGATTTGGC

CypB Reverse TAATACGACTCACTATAGGCCAGGCTCTCTACTCCTTGG

Renin Forward TCTCTGGGCACTCTTGTTGCTCTGGACCTCTTGTAGCTT

Renin Reverse AGTCTCCCGACAGACACAGCCAGCTTTGGACGAATCTTGCTCAAGAAA

Table A.4. In Situ Hybridization Primers Used in this Study Listed are all primers used for in

situ hybridization

References

[1] A. A. Fernandez-Ramos, V. Poindessous, C. Marchetti-Laurent, N. Pallet, and M.-

A. Loriot, “The effect of immunosuppressive molecules on t-cell metabolic repro-

gramming,” Biochimie, vol. 127, pp. 23–36, 2016.

[2] I. J. Saldanha, O. Akinyede, and K. A. Robinson, “Immunosuppressive drug ther-

apy for preventing rejection following lung transplantation in cystic fibrosis,” The

Cochrane database of systematic reviews, no. 11, p. CD009421, 2015.

[3] D. C. Brennan, “Long-term trends in allograft survival,” Advances in chronic kidney

disease, vol. 13, no. 1, pp. 11–17, 2006.

[4] C.-L. Chou, C.-Y. Chou, Y.-Y. Huang, M.-S. Wu, C.-C. Hsu, and Y.-C. Chou,

“Prescription trends of immunosuppressive drugs in post-heart transplant recipients

in taiwan, 2000-2009,” Pharmacoepidemiology and drug safety, 2014.

[5] M. Naesens, D. R. J. Kuypers, and M. Sarwal, “Calcineurin inhibitor nephrotoxic-

ity,” Clinical journal of the American Society of Nephrology : CJASN, vol. 4, no. 2,

pp. 481–508, 2009.

[6] A. J. Olyaei, A. M. de Mattos, and W. M. Bennett, “Immunosuppressant-induced

nephropathy: pathophysiology, incidence and management,” Drug safety, vol. 21,

no. 6, pp. 471–488, 1999.

[7] C. R. Williams and J. L. Gooch, “Calcineurin inhibitors and immunosuppression -

a tale of two isoforms,” Expert reviews in molecular medicine, vol. 14, p. e14, 2012.

[8] A. Busauschina, P. Schnuelle, and F. J. van der Woude, “Cyclosporine nephrotoxi-

city,” Transplantation proceedings, vol. 36, no. 2 Suppl, pp. 229S–233S, 2004.

References 79

[9] L. E. Kassel and L. E. Odum, “Our own worst enemy: pharmacologic mechanisms

of hypertension,” Advances in chronic kidney disease, vol. 22, no. 3, pp. 245–252,

2015.

[10] P. W. Flatman, “Cotransporters, wnks and hypertension: Important leads from

the study of monogenetic disorders of blood pressure regulation,” Clinical science

(London, England : 1979), vol. 112, no. 4, pp. 203–216, 2007.

[11] J. Roy and M. S. Cyert, “Cracking the phosphatase code: docking interactions

determine substrate specificity,” Science signaling, vol. 2, no. 100, p. re9, 2009.

[12] F. Rusnak and P. Mertz, “Calcineurin: Form and function,” Physiological reviews,

vol. 80, no. 4, pp. 1483–1521, 2000.

[13] H. Jiang, F. Xiong, S. Kong, T. Ogawa, M. Kobayashi, and J. O. Liu, “Distinct

tissue and cellular distribution of two major isoforms of calcineurin,” Molecular

immunology, vol. 34, no. 8-9, pp. 663–669, 1997.

[14] E. S. Masuda, R. Imamura, Y. Amasaki, K. Arai, and N. Arai, “Signalling into

the t-cell nucleus: Nfat regulation,” Cellular signalling, vol. 10, no. 9, pp. 599–611,

1998.

[15] E. J. Hoorn, S. B. Walsh, J. A. McCormick, A. F¨urstenberg, C.-L. Yang,

T. Roeschel, A. Paliege, A. J. Howie, J. Conley, S. Bachmann, R. J. Unwin, and

D. H. Ellison, “The calcineurin inhibitor tacrolimus activates the renal sodium

chloride cotransporter to cause hypertension,” Nature medicine, vol. 17, no. 10,

pp. 1304–1309, 2011.

[16] E. J. Hoorn, S. B. Walsh, J. A. McCormick, R. Zietse, R. J. Unwin, and D. H.

Ellison, “Pathogenesis of calcineurin inhibitor-induced hypertension,” Journal of

nephrology, vol. 25, no. 3, pp. 269–275, 2012.

[17] C. Esteva-Font, E. Ars, E. Guillen-Gomez, J. M. Campistol, L. Sanz, W. Jimenez,

M. A. Knepper, F. Torres, R. Torra, J. A. Ballarin, and P. Fernandez-Llama,

“Ciclosporin-induced hypertension is associated with increased sodium transporter

of the loop of henle (nkcc2),” Nephrology, dialysis, transplantation : official publi-

cation of the European Dialysis and Transplant Association - European Renal As-

sociation, vol. 22, no. 10, pp. 2810–2816, 2007.

References 80

[18] S. Damiano, R. Scanni, R. Ciarcia, S. Florio, and G. Capasso, “Regulation of sodium

transporters in the kidney during cyclosporine treatment,” Journal of nephrology,

vol. 23 Suppl 16, pp. S191–8, 2010.

[19] E. J. Hoorn, J. H. Nelson, J. A. McCormick, and D. H. Ellison, “The wnk kinase net-

work regulating sodium, potassium, and blood pressure,” Journal of the American

Society of Nephrology : JASN, vol. 22, no. 4, pp. 605–614, 2011.

[20] K. Mutig, T. Saritas, S. Uchida, T. Kahl, T. Borowski, A. Paliege, A. B¨ohlick,

M. Bleich, Q. Shan, and S. Bachmann, “Short-term stimulation of the thiazide-

sensitive na+-cl- cotransporter by vasopressin involves phosphorylation and mem-

brane translocation,” American journal of physiology. Renal physiology, vol. 298,

no. 3, pp. F502–9, 2010.

[21] A. Kurtz, R. Della Bruna, and K. Kuhn, “Cyclosporine a enhances renin secretion

and production in isolated juxtaglomerular cells,” Kidney international, vol. 33,

no. 5, pp. 947–953, 1988.

[22] P. A. Heidenreich, J. G. Trogdon, O. A. Khavjou, J. Butler, K. Dracup, M. D.

Ezekowitz, E. A. Finkelstein, Y. Hong, S. C. Johnston, A. Khera, D. M. Lloyd-Jones,

S. A. Nelson, G. Nichol, D. Orenstein, P. W. F. Wilson, and Y. J. Woo, “Forecasting

the future of cardiovascular disease in the united states: A policy statement from

the american heart association,” Circulation, vol. 123, no. 8, pp. 933–944, 2011.

[23] F. H. Rafiqi, A. M. Zuber, M. Glover, C. Richardson, S. Fleming, S. Jovanovic,

A. Jovanovic, K. M. O’Shaughnessy, and D. R. Alessi, “Role of the wnk-activated

spak kinase in regulating blood pressure,” EMBO molecular medicine, vol. 2, no. 2,

pp. 63–75, 2010.

[24] F. H. Wilson, S. Disse-Nicod`eme, K. A. Choate, K. Ishikawa, C. Nelson-Williams,

I. Desitter, M. Gunel, D. V. Milford, G. W. Lipkin, J. M. Achard, M. P. Feely,

B. Dussol, Y. Berland, R. J. Unwin, H. Mayan, D. B. Simon, Z. Farfel, X. Je-

unemaitre, and R. P. Lifton, “Human hypertension caused by mutations in wnk

kinases,” Science (New York, N.Y.), vol. 293, no. 5532, pp. 1107–1112, 2001.

[25] K. Besseghir and F. Roch-Ramel, “Renal excretion of drugs and other xenobiotics,”

Renal physiology, vol. 10, no. 5, pp. 221–241, 1987.

References 81

[26] S. GLABMAN, H. S. AYNEDJIAN, and N. BANK, “Micropuncture study of the

effect of acute reductions in glomerular filtration rate on sodium and water reab-

sorption by the proximal tubules of the rat,” The Journal of clinical investigation,

vol. 44, pp. 1410–1416, 1965.

[27] L. WERKO, J. EK, E. VARNAUSKAS, H. BUCHT, B. THOMASSON, and

H. ELIASCH, “The relationship between renal blood flow, glomerular filtration rate

and sodium excretion, cardiac output and pulmonary and systemic blood pressures

in various heart disorders,” American heart journal, vol. 49, no. 6, pp. 823–837,

1955.

[28] B. J. Potter, “The effect of an intravenous infusion of hypertonic saline on re-

nal mechanisms and on electrolyte changes in sheep,” The Journal of physiology,

vol. 184, no. 3, pp. 605–617, 1966.

[29] W. Kriz and L. Bankir, “A standard nomenclature for structures of the kidney. the

renal commission of the international union of physiological sciences (iups),” Kidney

international, vol. 33, no. 1, pp. 1–7, 1988.

[30] A. B. Donnersberger and D. R. Donnersberger, A laboratory textbook of anatomy

and physiology: Cat version. Sudbury, Mass.: Jones and Bartlett Publishers, 9th

ed. ed., 2010.

[31] A. Benninghoff and D. Drenckhahn, Anatomie: Makroskopische Anatomie, Embry-

ologie und Histologie des Menschen. M¨unchen [u.a.]: Urban & Schwarzenberg, 17.,

durchges. aufl. ed., 2008.

[32] M. H. Little, Kidney development, disease, repair and regeneration. Amsterdam:

Elsevier, 2016.

[33] R. J. Alpern, S. C. Herbert, D. W. Seldin, and G. H. Giebisch, Seldin and Giebisch’s

the kidney: Physiology & pathophysiology. Amsterdam and Boston: Elsevier Aca-

demic Press, 4th ed. ed., 2008.

[34] K. Hierholzer and M. Wiederholt, “Some aspects of distal tubular solute and water

transport,” Kidney international, vol. 9, no. 2, pp. 198–213, 1976.

[35] R. Greger, “Ion transport mechanisms in thick ascending limb of henle’s loop of

mammalian nephron,” Physiological reviews, vol. 65, no. 3, pp. 760–797, 1985.

References 82

[36] M. B. Burg, “Tubular chloride transport and the mode of action of some diuretics,”

Kidney international, vol. 9, no. 2, pp. 189–197, 1976.

[37] C. M. Bennett, B. M. Brenner, and R. W. Berliner, “Micropuncture study of

nephron function in the rhesus monkey,” The Journal of clinical investigation,

vol. 47, no. 1, pp. 203–216, 1968.

[38] M. Burg and D. Good, “Sodium chloride coupled transport in mammalian

nephrons,” Annual review of physiology, vol. 45, pp. 533–547, 1983.

[39] M. B. Burg, S. PAPPER, and J. D. ROSENBAUM, “Factors influencing the diuretic

response to ingested water,” The Journal of laboratory and clinical medicine, vol. 57,

pp. 533–545, 1961.

[40] S. Nielsen, A. B. Maunsbach, C. A. Ecelbarger, and M. A. Knepper, “Ultrastructural

localization of na-k-2cl cotransporter in thick ascending limb and macula densa of

rat kidney,” The American journal of physiology, vol. 275, no. 6 Pt 2, pp. F885–93,

1998.

[41] J. Schnermann, H. Osswald, and M. Hermle, “Inhibitory effect of methylxanthines

on feedback control of glomerular filtration rate in the rat kidney,” Pflugers Archiv

: European journal of physiology, vol. 369, no. 1, pp. 39–48, 1977.

[42] S. Thomson, D. Bao, A. Deng, and V. Vallon, “Adenosine formed by 5’-nucleotidase

mediates tubuloglomerular feedback,” The Journal of clinical investigation, vol. 106,

no. 2, pp. 289–298, 2000.

[43] T. A. Kotchen, W. J. Welch, J. N. Lorenz, and C. E. Ott, “Renal tubular chloride

and renin release,” The Journal of laboratory and clinical medicine, vol. 110, no. 5,

pp. 533–540, 1987.

[44] T. K. Keeton and W. B. Campbell, “The pharmacologic alteration of renin release,”

Pharmacological reviews, vol. 32, no. 2, pp. 81–227, 1980.

[45] R. M. Carey, “The intrarenal renin-angiotensin system in hypertension,” Advances

in chronic kidney disease, vol. 22, no. 3, pp. 204–210, 2015.

[46] P. H. Baylis, “Regulation of vasopressin secretion,” Bailliere’s clinical endocrinology

and metabolism, vol. 3, no. 2, pp. 313–330, 1989.

References 83

[47] E. E. Windhager, Handbook of physiology. Bethesda, Md.: American Physiological

Soc, 1992.

[48] M. Carlstrom, C. S. Wilcox, and W. J. Arendshorst, “Renal autoregulation in health

and disease,” Physiological reviews, vol. 95, no. 2, pp. 405–511, 2015.

[49] D. H. Ellison, H. Vel´azquez, and F. S. Wright, “Thiazide-sensitive sodium chloride

cotransport in early distal tubule,” The American journal of physiology, vol. 253,

no. 3 Pt 2, pp. F546–54, 1987.

[50] G. Gamba, S. N. Saltzberg, M. Lombardi, A. Miyanoshita, J. Lytton, M. A. Hediger,

B. M. Brenner, and S. C. Hebert, “Primary structure and functional expression of a

cdna encoding the thiazide-sensitive, electroneutral sodium-chloride cotransporter,”

Proceedings of the National Academy of Sciences of the United States of America,

vol. 90, no. 7, pp. 2749–2753, 1993.

[51] G. Gamba, “Molecular physiology and pathophysiology of electroneutral cation-

chloride cotransporters,” Physiological reviews, vol. 85, no. 2, pp. 423–493, 2005.

[52] R. Schmitt, D. H. Ellison, N. Farman, B. C. Rossier, R. F. Reilly, W. B. Reeves,

I. Oberbaumer, R. Tapp, and S. Bachmann, “Developmental expression of sodium

entry pathways in rat nephron,” The American journal of physiology, vol. 276, no. 3

Pt 2, pp. F367–81, 1999.

[53] T. Wang and G. Giebisch, “Effects of angiotensin ii on electrolyte transport in the

early and late distal tubule in rat kidney,” The American journal of physiology,

vol. 271, no. 1 Pt 2, pp. F143–9, 1996.

[54] G. H. Kim, S. Masilamani, R. Turner, C. Mitchell, J. B. Wade, and M. A. Knep-

per, “The thiazide-sensitive na-cl cotransporter is an aldosterone-induced protein,”

Proceedings of the National Academy of Sciences of the United States of America,

vol. 95, no. 24, pp. 14552–14557, 1998.

[55] J. M. Elalouf, N. Roinel, and C. de Rouffignac, “Effects of antidiuretic hormone

on electrolyte reabsorption and secretion in distal tubules of rat kidney,” Pflugers

Archiv : European journal of physiology, vol. 401, no. 2, pp. 167–173, 1984.

References 84

[56] P. Igarashi, G. B. Vanden Heuvel, J. A. Payne, and B. r. Forbush, “Cloning, em-

bryonic expression, and alternative splicing of a murine kidney-specific na-k-cl co-

transporter,” The American journal of physiology, vol. 269, no. 3 Pt 2, pp. F405–18,

1995.

[57] G. R. Ares, P. S. Caceres, and P. A. Ortiz, “Molecular regulation of nkcc2 in the

thick ascending limb,” American journal of physiology. Renal physiology, vol. 301,

no. 6, pp. F1143–59, 2011.

[58] D. B. Mount, A. Baekgaard, A. E. Hall, C. Plata, J. Xu, D. R. Beier, G. Gamba,

and S. C. Hebert, “Isoforms of the na-k-2cl cotransporter in murine tal i. molecular

characterization and intrarenal localization,” The American journal of physiology,

vol. 276, no. 3 Pt 2, pp. F347–58, 1999.

[59] J. A. Payne and B. r. Forbush, “Alternatively spliced isoforms of the putative re-

nal na-k-cl cotransporter are differentially distributed within the rabbit kidney,”

Proceedings of the National Academy of Sciences of the United States of America,

vol. 91, no. 10, pp. 4544–4548, 1994.

[60] I. Carota, F. Theilig, M. Oppermann, P. Kongsuphol, A. Rosenauer, R. Schreiber,

B. L. Jensen, S. Walter, K. Kunzelmann, and H. Castrop, “Localization and func-

tional characterization of the human nkcc2 isoforms,” Acta physiologica (Oxford,

England), vol. 199, no. 3, pp. 327–338, 2010.

[61] M. Oppermann, D. Mizel, G. Huang, C. Li, C. Deng, F. Theilig, S. Bachmann,

J. Briggs, J. Schnermann, and H. Castrop, “Macula densa control of renin secretion

and preglomerular resistance in mice with selective deletion of the b isoform of the

na,k,2cl co-transporter,” Journal of the American Society of Nephrology : JASN,

vol. 17, no. 8, pp. 2143–2152, 2006.

[62] K. Fiedler and K. Simons, “The role of n-glycans in the secretory pathway,” Cell,

vol. 81, no. 3, pp. 309–312, 1995.

[63] N. Mastroianni, A. Bettinelli, M. Bianchetti, G. Colussi, M. de Fusco, F. Sereni,

A. Ballabio, and G. Casari, “Novel molecular variants of the na-cl cotransporter

gene are responsible for gitelman syndrome,” American journal of human genetics,

vol. 59, no. 5, pp. 1019–1026, 1996.

References 85

[64] D. B. Simon, C. Nelson-Williams, M. J. Bia, D. Ellison, F. E. Karet, A. M. Molina,

I. Vaara, F. Iwata, H. M. Cushner, M. Koolen, F. J. Gainza, H. J. Gitleman, and

R. P. Lifton, “Gitelman’s variant of bartter’s syndrome, inherited hypokalaemic al-

kalosis, is caused by mutations in the thiazide-sensitive na-cl cotransporter,” Nature

genetics, vol. 12, no. 1, pp. 24–30, 1996.

[65] B. Ko, A. C. Mistry, L. Hanson, R. Mallick, L. L. Cooke, B. K. Hack, P. Cunning-

ham, and R. S. Hoover, “A new model of the distal convoluted tubule,” American

journal of physiology. Renal physiology, vol. 303, no. 5, pp. F700–10, 2012.

[66] R. S. Hoover, E. Poch, A. Monroy, N. Vazquez, T. Nishio, G. Gamba, and S. C.

Hebert, “N-glycosylation at two sites critically alters thiazide binding and activity

of the rat thiazide-sensitive na(+):cl(-) cotransporter,” Journal of the American

Society of Nephrology : JASN, vol. 14, no. 2, pp. 271–282, 2003.

[67] S. Kunchaparty, M. Palcso, J. Berkman, H. Velazquez, G. V. Desir, P. Bernstein,

R. F. Reilly, and D. H. Ellison, “Defective processing and expression of thiazide-

sensitive na-cl cotransporter as a cause of gitelman’s syndrome,” The American

journal of physiology, vol. 277, no. 4 Pt 2, pp. F643–9, 1999.

[68] A. Borschewski, N. Himmerkus, C. Boldt, K. I. Blankenstein, J. A. McCormick,

R. Lazelle, T. E. Willnow, V. Jankowski, A. Plain, M. Bleich, D. H. Ellison, S. Bach-

mann, and K. Mutig, “Calcineurin and sorting-related receptor with a-type repeats

interact to regulate the renal na(+)-k(+)-2cl(-) cotransporter,” Journal of the Amer-

ican Society of Nephrology : JASN, vol. 27, no. 1, pp. 107–119, 2016.

[69] C. Richardson, K. Sakamoto, P. de los Heros, M. Deak, D. G. Campbell, A. R.

Prescott, and D. R. Alessi, “Regulation of the nkcc2 ion cotransporter by spak-osr1-

dependent and -independent pathways,” Journal of cell science, vol. 124, no. Pt 5,

pp. 789–800, 2011.

[70] I. Gimenez and B. Forbush, “Regulatory phosphorylation sites in the nh2 terminus

of the renal na-k-cl cotransporter (nkcc2),” American journal of physiology. Renal

physiology, vol. 289, no. 6, pp. F1341–5, 2005.

[71] P. A. Ortiz, “camp increases surface expression of nkcc2 in rat thick ascending

limbs: role of vamp,” American journal of physiology. Renal physiology, vol. 290,

no. 3, pp. F608–16, 2006.

References 86

[72] M. B. Sandberg, A. B. Maunsbach, and A. A. McDonough, “Redistribution of distal

tubule na+-cl- cotransporter (ncc) in response to a high-salt diet,” American journal

of physiology. Renal physiology, vol. 291, no. 2, pp. F503–8, 2006.

[73] P. Meade, R. S. Hoover, C. Plata, N. V´azquez, N. A. Bobadilla, G. Gamba, and

S. C. Hebert, “camp-dependent activation of the renal-specific na+-k+-2cl- cotrans-

porter is mediated by regulation of cotransporter trafficking,” American journal of

physiology. Renal physiology, vol. 284, no. 6, pp. F1145–54, 2003.

[74] C. Dathe, A.-L. Daigeler, W. Seifert, V. Jankowski, R. Mrowka, R. Kalis, E. Wanker,

K. Mutig, S. Bachmann, and A. Paliege, “Annexin a2 mediates apical trafficking

of renal na+-k+-2cl−cotransporter,” The Journal of biological chemistry, vol. 289,

no. 14, pp. 9983–9997, 2014.

[75] G. R. Ares and P. A. Ortiz, “Dynamin2, clathrin, and lipid rafts mediate endocytosis

of the apical na/k/2cl cotransporter nkcc2 in thick ascending limbs,” The Journal

of biological chemistry, vol. 287, no. 45, pp. 37824–37834, 2012.

[76] L. L. Rosenbaek, M. L. A. Kortenoeven, T. S. Aroankins, and R. A. Fenton, “Phos-

phorylation decreases ubiquitylation of the thiazide-sensitive cotransporter ncc and

subsequent clathrin-mediated endocytosis,” The Journal of biological chemistry,

vol. 289, no. 19, pp. 13347–13361, 2014.

[77] I. Gim´enez and B. Forbush, “Short-term stimulation of the renal na-k-cl cotrans-

porter (nkcc2) by vasopressin involves phosphorylation and membrane translocation

of the protein,” The Journal of biological chemistry, vol. 278, no. 29, pp. 26946–

26951, 2003.

[78] S. A. Fraser, I. Gimenez, N. Cook, I. Jennings, M. Katerelos, F. Katsis, V. Levid-

iotis, B. E. Kemp, and D. A. Power, “Regulation of the renal-specific na+-k+-2cl-

co-transporter nkcc2 by amp-activated protein kinase (ampk),” The Biochemical

journal, vol. 405, no. 1, pp. 85–93, 2007.

[79] D. Pacheco-Alvarez, P. S. Cristobal, P. Meade, E. Moreno, N. Vazquez, E. Munoz,

A. Diaz, M. E. Juarez, I. Gimenez, and G. Gamba, “The na+:cl- cotransporter

is activated and phosphorylated at the amino-terminal domain upon intracellular

chloride depletion,” The Journal of biological chemistry, vol. 281, no. 39, pp. 28755–

28763, 2006.

References 87

[80] G. Gamba, “Regulation of the renal na+-cl- cotransporter by phosphorylation and

ubiquitylation,” American journal of physiology. Renal physiology, vol. 303, no. 12,

pp. F1573–83, 2012.

[81] M. Chiga, T. Rai, S.-S. Yang, A. Ohta, T. Takizawa, S. Sasaki, and S. Uchida,

“Dietary salt regulates the phosphorylation of osr1/spak kinases and the sodium

chloride cotransporter through aldosterone,” Kidney international, vol. 74, no. 11,

pp. 1403–1409, 2008.

[82] C. Richardson, F. H. Rafiqi, H. K. R. Karlsson, N. Moleleki, A. Vandewalle, D. G.

Campbell, N. A. Morrice, and D. R. Alessi, “Activation of the thiazide-sensitive

na+-cl- cotransporter by the wnk-regulated kinases spak and osr1,” Journal of cell

science, vol. 121, no. Pt 5, pp. 675–684, 2008.

[83] S.-S. Yang, T. Morimoto, T. Rai, M. Chiga, E. Sohara, M. Ohno, K. Uchida, S.-H.

Lin, T. Moriguchi, H. Shibuya, Y. Kondo, S. Sasaki, and S. Uchida, “Molecu-

lar pathogenesis of pseudohypoaldosteronism type ii: Generation and analysis of a

wnk4(d561a/+) knockin mouse model,” Cell metabolism, vol. 5, no. 5, pp. 331–344,

2007.

[84] K. Mutig, A. Paliege, T. Kahl, T. J¨ons, W. M¨uller-Esterl, and S. Bachmann, “Va-

sopressin v2 receptor expression along rat, mouse, and human renal epithelia with

focus on tal,” American journal of physiology. Renal physiology, vol. 293, no. 4,

pp. F1166–77, 2007.

[85] L. L. Rosenbaek, M. Assentoft, N. B. Pedersen, N. MacAulay, and R. A. Fenton,

“Characterization of a novel phosphorylation site in the sodium-chloride cotrans-

porter, ncc,” The Journal of physiology, vol. 590, no. 23, pp. 6121–6139, 2012.

[86] K. Piechotta, J. Lu, and E. Delpire, “Cation chloride cotransporters interact with

the stress-related kinases ste20-related proline-alanine-rich kinase (spak) and oxida-

tive stress response 1 (osr1),” The Journal of biological chemistry, vol. 277, no. 52,

pp. 50812–50819, 2002.

[87] S.-S. Yang, Y.-F. Lo, C.-C. Wu, S.-W. Lin, C.-J. Yeh, P. Chu, H.-K. Sytwu,

S. Uchida, S. Sasaki, and S.-H. Lin, “Spak-knockout mice manifest gitelman syn-

drome and impaired vasoconstriction,” Journal of the American Society of Nephrol-

ogy : JASN, vol. 21, no. 11, pp. 1868–1877, 2010.

References 88

[88] A. N. Anselmo, S. Earnest, W. Chen, Y.-C. Juang, S. C. Kim, Y. Zhao, and M. H.

Cobb, “Wnk1 and osr1 regulate the na+, k+, 2cl- cotransporter in hela cells,”

Proceedings of the National Academy of Sciences of the United States of America,

vol. 103, no. 29, pp. 10883–10888, 2006.

[89] T. Moriguchi, S. Urushiyama, N. Hisamoto, S.-i. Iemura, S. Uchida, T. Nat-

sume, K. Matsumoto, and H. Shibuya, “Wnk1 regulates phosphorylation of cation-

chloride-coupled cotransporters via the ste20-related kinases, spak and osr1,” The

Journal of biological chemistry, vol. 280, no. 52, pp. 42685–42693, 2005.

[90] B. Xu, J. M. English, J. L. Wilsbacher, S. Stippec, E. J. Goldsmith, and M. H.

Cobb, “Wnk1, a novel mammalian serine/threonine protein kinase lacking the cat-

alytic lysine in subdomain ii,” The Journal of biological chemistry, vol. 275, no. 22,

pp. 16795–16801, 2000.

[91] X. Min, B.-H. Lee, M. H. Cobb, and E. J. Goldsmith, “Crystal structure of the

kinase domain of wnk1, a kinase that causes a hereditary form of hypertension,”

Structure (London, England : 1993), vol. 12, no. 7, pp. 1303–1311, 2004.

[92] L. Y. Lenertz, B.-H. Lee, X. Min, B.-e. Xu, K. Wedin, S. Earnest, E. J. Goldsmith,

and M. H. Cobb, “Properties of wnk1 and implications for other family members,”

The Journal of biological chemistry, vol. 280, no. 29, pp. 26653–26658, 2005.

[93] A. T. Piala, T. M. Moon, R. Akella, H. He, M. H. Cobb, and E. J. Goldsmith, “Chlo-

ride sensing by wnk1 involves inhibition of autophosphorylation,” Science signaling,

vol. 7, no. 324, p. ra41, 2014.

[94] B.-e. Xu, X. Min, S. Stippec, B.-H. Lee, E. J. Goldsmith, and M. H. Cobb, “Regula-

tion of wnk1 by an autoinhibitory domain and autophosphorylation,” The Journal

of biological chemistry, vol. 277, no. 50, pp. 48456–48462, 2002.

[95] J. O. Thastrup, F. H. Rafiqi, A. C. Vitari, E. Pozo-Guisado, M. Deak, Y. Mehel-

lou, and D. R. Alessi, “Spak/osr1 regulate nkcc1 and wnk activity: Analysis of

wnk isoform interactions and activation by t-loop trans-autophosphorylation,” The

Biochemical journal, vol. 441, no. 1, pp. 325–337, 2012.

References 89

[96] A. C. Vitari, J. Thastrup, F. H. Rafiqi, M. Deak, N. A. Morrice, H. K. R. Karlsson,

and D. R. Alessi, “Functional interactions of the spak/osr1 kinases with their up-

stream activator wnk1 and downstream substrate nkcc1,” The Biochemical journal,

vol. 397, no. 1, pp. 223–231, 2006.

[97] K. B. E. Gagnon, R. England, and E. Delpire, “A single binding motif is required

for spak activation of the na-k-2cl cotransporter,” Cellular physiology and biochem-

istry : international journal of experimental cellular physiology, biochemistry, and

pharmacology, vol. 20, no. 1-4, pp. 131–142, 2007.

[98] K. B. Gagnon and E. Delpire, “On the substrate recognition and negative regulation

of spak, a kinase modulating na+-k+-2cl- cotransport activity,” American journal

of physiology. Cell physiology, vol. 299, no. 3, pp. C614–20, 2010.

[99] K. B. E. Gagnon, R. England, and E. Delpire, “Characterization of spak and osr1,

regulatory kinases of the na-k-2cl cotransporter,” Molecular and cellular biology,

vol. 26, no. 2, pp. 689–698, 2006.

[100] D. A. Molony, W. B. Reeves, S. C. Hebert, and T. E. Andreoli, “Adh increases

apical na+, k+, 2cl- entry in mouse medullary thick ascending limbs of henle,” The

American journal of physiology, vol. 252, no. 1 Pt 2, pp. F177–87, 1987.

[101] S. C. Hebert, P. A. Friedman, and T. E. Andreoli, “Effects of antidiuretic hormone

on cellular conductive pathways in mouse medullary thick ascending limbs of henle:

I. adh increases transcellular conductance pathways,” The Journal of membrane

biology, vol. 80, no. 3, pp. 201–219, 1984.

[102] C. A. Ecelbarger, G. H. Kim, J. B. Wade, and M. A. Knepper, “Regulation of the

abundance of renal sodium transporters and channels by vasopressin,” Experimental

neurology, vol. 171, no. 2, pp. 227–234, 2001.

[103] R. Gunaratne, D. W. W. Braucht, M. M. Rinschen, C.-L. Chou, J. D. Hoffert,

T. Pisitkun, and M. A. Knepper, “Quantitative phosphoproteomic analysis reveals

camp/vasopressin-dependent signaling pathways in native renal thick ascending limb

cells,” Proceedings of the National Academy of Sciences of the United States of

America, vol. 107, no. 35, pp. 15653–15658, 2010.

References 90

[104] H. Amlal, C. Legoff, C. Vernimmen, M. Paillard, and M. Bichara, “Na(+)-

k+(nh4+)-2cl- cotransport in medullary thick ascending limb: Control by pka, pkc,

and 20-hete,” The American journal of physiology, vol. 271, no. 2 Pt 1, pp. C455–63,

1996.

[105] T. Saritas, A. Borschewski, J. A. McCormick, A. Paliege, C. Dathe, S. Uchida,

A. Terker, N. Himmerkus, M. Bleich, S. Demaretz, K. Laghmani, E. Delpire, D. H.

Ellison, S. Bachmann, and K. Mutig, “Spak differentially mediates vasopressin ef-

fects on sodium cotransporters,” Journal of the American Society of Nephrology :

JASN, vol. 24, no. 3, pp. 407–418, 2013.

[106] N. B. Pedersen, M. V. Hofmeister, L. L. Rosenbaek, J. Nielsen, and R. A. Fenton,

“Vasopressin induces phosphorylation of the thiazide-sensitive sodium chloride co-

transporter in the distal convoluted tubule,” Kidney international, vol. 78, no. 2,

pp. 160–169, 2010.

[107] V. Vallon, J. Schroth, F. Lang, D. Kuhl, and S. Uchida, “Expression and phos-

phorylation of the na+-cl- cotransporter ncc in vivo is regulated by dietary salt,

potassium, and sgk1,” American journal of physiology. Renal physiology, vol. 297,

no. 3, pp. F704–12, 2009.

[108] N. van der Lubbe, C. H. Lim, R. A. Fenton, M. E. Meima, A. H. Jan Danser,

R. Zietse, and E. J. Hoorn, “Angiotensin ii induces phosphorylation of the thiazide-

sensitive sodium chloride cotransporter independent of aldosterone,” Kidney inter-

national, vol. 79, no. 1, pp. 66–76, 2011.

[109] K. T. Kahle, F. H. Wilson, Q. Leng, M. D. Lalioti, A. D. O’Connell, K. Dong,

A. K. Rapson, G. G. MacGregor, G. Giebisch, S. C. Hebert, and R. P. Lifton,

“Wnk4 regulates the balance between renal nacl reabsorption and k+ secretion,”

Nature genetics, vol. 35, no. 4, pp. 372–376, 2003.

[110] P. San-Cristobal, D. Pacheco-Alvarez, C. Richardson, A. M. Ring, N. Vazquez, F. H.

Rafiqi, D. Chari, K. T. Kahle, Q. Leng, N. A. Bobadilla, S. C. Hebert, D. R. Alessi,

R. P. Lifton, and G. Gamba, “Angiotensin ii signaling increases activity of the renal

na-cl cotransporter through a wnk4-spak-dependent pathway,” Proceedings of the

National Academy of Sciences of the United States of America, vol. 106, no. 11,

pp. 4384–4389, 2009.

References 91

[111] D. B. Simon, F. E. Karet, J. M. Hamdan, A. DiPietro, S. A. Sanjad, and R. P.

Lifton, “Bartter’s syndrome, hypokalaemic alkalosis with hypercalciuria, is caused

by mutations in the na-k-2cl cotransporter nkcc2,” Nature genetics, vol. 13, no. 2,

pp. 183–188, 1996.

[112] R. D. Gordon, “Syndrome of hypertension and hyperkalemia with normal glomerular

filtration rate,” Hypertension (Dallas, Tex. : 1979), vol. 8, no. 2, pp. 93–102, 1986.

[113] C. H. Lee and G.-H. Kim, “Electrolyte and acid-base disturbances induced by

clacineurin inhibitors,” Electrolyte & blood pressure : E & BP, vol. 5, no. 2, pp. 126–

130, 2007.

[114] L. M. Boyden, M. Choi, K. A. Choate, C. J. Nelson-Williams, A. Farhi, H. R. Toka,

I. R. Tikhonova, R. Bjornson, S. M. Mane, G. Colussi, M. Lebel, R. D. Gordon, B. A.

Semmekrot, A. Poujol, M. J. Valimaki, M. E. de Ferrari, S. A. Sanjad, M. Gutkin,

F. E. Karet, J. R. Tucci, J. R. Stockigt, K. M. Keppler-Noreuil, C. C. Porter, S. K.

Anand, M. L. Whiteford, I. D. Davis, S. B. Dewar, A. Bettinelli, J. J. Fadrowski,

C. W. Belsha, T. E. Hunley, R. D. Nelson, H. Trachtman, T. R. P. Cole, M. Pinsk,

D. Bockenhauer, M. Shenoy, P. Vaidyanathan, J. W. Foreman, M. Rasoulpour,

F. Thameem, H. Z. Al-Shahrouri, J. Radhakrishnan, A. G. Gharavi, B. Goilav,

and R. P. Lifton, “Mutations in kelch-like 3 and cullin 3 cause hypertension and

electrolyte abnormalities,” Nature, vol. 482, no. 7383, pp. 98–102, 2012.

[115] H. Louis-Dit-Picard, J. Barc, D. Trujillano, S. Miserey-Lenkei, N. Bouatia-Naji,

O. Pylypenko, G. Beaurain, A. Bonnefond, O. Sand, C. Simian, E. Vidal-Petiot,

C. Soukaseum, C. Mandet, F. Broux, O. Chabre, M. Delahousse, V. Esnault, B. Fi-

quet, P. Houillier, C. I. Bagnis, J. Koenig, M. Konrad, P. Landais, C. Mourani,

P. Niaudet, V. Probst, C. Thauvin, R. J. Unwin, S. D. Soroka, G. Ehret, S. Os-

sowski, M. Caulfield, P. Bruneval, X. Estivill, P. Froguel, J. Hadchouel, J.-J. Schott,

and X. Jeunemaitre, “Klhl3 mutations cause familial hyperkalemic hypertension by

impairing ion transport in the distal nephron,” Nature genetics, vol. 44, no. 4,

pp. 456–60, S1–3, 2012.

[116] K. Susa, E. Sohara, T. Rai, M. Zeniya, Y. Mori, T. Mori, M. Chiga, N. Nomura,

H. Nishida, D. Takahashi, K. Isobe, Y. Inoue, K. Takeishi, N. Takeda, S. Sasaki, and

S. Uchida, “Impaired degradation of wnk1 and wnk4 kinases causes phaii in mutant

References 92

klhl3 knock-in mice,” Human molecular genetics, vol. 23, no. 19, pp. 5052–5060,

2014.

[117] Y. Mori, M. Wakabayashi, T. Mori, Y. Araki, E. Sohara, T. Rai, S. Sasaki, and

S. Uchida, “Decrease of wnk4 ubiquitination by disease-causing mutations of klhl3

through different molecular mechanisms,” Biochemical and biophysical research

communications, vol. 439, no. 1, pp. 30–34, 2013.

[118] M. Wakabayashi, T. Mori, K. Isobe, E. Sohara, K. Susa, Y. Araki, M. Chiga,

E. Kikuchi, N. Nomura, Y. Mori, H. Matsuo, T. Murata, S. Nomura, T. Asano,

H. Kawaguchi, S. Nonoyama, T. Rai, S. Sasaki, and S. Uchida, “Impaired klhl3-

mediated ubiquitination of wnk4 causes human hypertension,” Cell reports, vol. 3,

no. 3, pp. 858–868, 2013.

[119] C. B. Klee, H. Ren, and X. Wang, “Regulation of the calmodulin-stimulated protein

phosphatase, calcineurin,” The Journal of biological chemistry, vol. 273, no. 22,

pp. 13367–13370, 1998.

[120] J. L. Gooch, J. J. Toro, R. L. Guler, and J. L. Barnes, “Calcineurin a-alpha but

not a-beta is required for normal kidney development and function,” The American

journal of pathology, vol. 165, no. 5, pp. 1755–1765, 2004.

[121] K. Ueki, T. Muramatsu, and R. L. Kincaid, “Structure and expression of two iso-

forms of the murine calmodulin-dependent protein phosphatase regulatory subunit

(calcineurin b),” Biochemical and biophysical research communications, vol. 187,

no. 1, pp. 537–543, 1992.

[122] T. Khatlani, S. Pradhan, Q. Da, F. C. Gushiken, A. L. Bergeron, K. W. Langlois,

J. D. Molkentin, R. E. Rumbaut, and K. V. Vijayan, “The b isoform of the catalytic

subunit of protein phosphatase 2b restrains platelet function by suppressing outside-

in aiibb3 integrin signaling,” Journal of thrombosis and haemostasis : JTH, 2014.

[123] J. L. Gooch, “An emerging role for calcineurin aalpha in the development and

function of the kidney,” American journal of physiology. Renal physiology, vol. 290,

no. 4, pp. F769–76, 2006.

References 93

[124] M. Buttini, S. Limonta, M. Luyten, and H. Boddeke, “Distribution of calcineurin

a isoenzyme mrnas in rat thymus and kidney,” The Histochemical journal, vol. 27,

no. 4, pp. 291–299, 1995.

[125] B. W. Zhang, G. Zimmer, J. Chen, D. Ladd, E. Li, F. W. Alt, G. Wiederrecht,

J. Cryan, E. A. O’Neill, C. E. Seidman, A. K. Abbas, and J. G. Seidman, “T

cell responses in calcineurin a alpha-deficient mice,” The Journal of experimental

medicine, vol. 183, no. 2, pp. 413–420, 1996.

[126] O. F. Bueno, E. B. Brandt, M. E. Rothenberg, and J. D. Molkentin, “Defective t

cell development and function in calcineurin a beta -deficient mice,” Proceedings of

the National Academy of Sciences of the United States of America, vol. 99, no. 14,

pp. 9398–9403, 2002.

[127] V. S. F. Chan, C. Wong, and P. S. Ohashi, “Calcineurin aalpha plays an exclusive

role in tcr signaling in mature but not in immature t cells,” European journal of

immunology, vol. 32, no. 5, pp. 1223–1229, 2002.

[128] K. Hocherl, F. Kees, B. K. Kramer, and A. Kurtz, “Cyclosporine a attenuates the

natriuretic action of loop diuretics by inhibition of renal cox-2 expression,” Kidney

international, vol. 65, no. 6, pp. 2071–2080, 2004.

[129] D. C. Hu, C. Burtner, A. Hong, P. I. Lobo, and M. D. Okusa, “Correction of renal

hypertension after kidney transplantation from a donor with gitelman syndrome,”

The American journal of the medical sciences, vol. 331, no. 2, pp. 105–109, 2006.

[130] J. Aramburu, F. Garcia-Cozar, A. Raghavan, H. Okamura, A. Rao, and P. G.

Hogan, “Selective inhibition of nfat activation by a peptide spanning the calcineurin

targeting site of nfat,” Molecular cell, vol. 1, no. 5, pp. 627–637, 1998.

[131] J. Roy, H. Li, P. G. Hogan, and M. S. Cyert, “A conserved docking site modulates

substrate affinity for calcineurin, signaling output, and in vivo function,” Molecular

cell, vol. 25, no. 6, pp. 889–901, 2007.

[132] A. Rodriguez, J. Roy, S. Martinez-Martinez, M. D. Lopez-Maderuelo, P. Nino-

Moreno, L. Orti, D. Pantoja-Uceda, A. Pineda-Lucena, M. S. Cyert, and J. M.

Redondo, “A conserved docking surface on calcineurin mediates interaction with

References 94

substrates and immunosuppressants,” Molecular cell, vol. 33, no. 5, pp. 616–626,

2009.

[133] S. R. Sheftic, R. Page, and W. Peti, “Investigating the human calcineurin interaction

network using the pilxvp slim,” Scientific reports, vol. 6, p. 38920, 2016.

[134] J. Liu, M. W. Albers, T. J. Wandless, S. Luan, D. G. Alberg, P. J. Belshaw, P. Co-

hen, C. MacKintosh, C. B. Klee, and S. L. Schreiber, “Inhibition of t cell signaling

by immunophilin-ligand complexes correlates with loss of calcineurin phosphatase

activity,” Biochemistry, vol. 31, no. 16, pp. 3896–3901, 1992.

[135] N. A. Clipstone, D. F. Fiorentino, and G. R. Crabtree, “Molecular analysis of the

interaction of calcineurin with drug-immunophilin complexes,” The Journal of bio-

logical chemistry, vol. 269, no. 42, pp. 26431–26437, 1994.

[136] A. Fahr, “Cyclosporin clinical pharmacokinetics,” Clinical pharmacokinetics,

vol. 24, no. 6, pp. 472–495, 1993.

[137] K. Iwasaki, “Metabolism of tacrolimus (fk506) and recent topics in clinical phar-

macokinetics,” Drug metabolism and pharmacokinetics, vol. 22, no. 5, pp. 328–335,

2007.

[138] B. F. Leas, S. Uhl, D. L. Sawinski, J. Trofe-Clark, S. Tuteja, J. L. Kaczmarek, and

C. A. Umscheid, “Calcineurin inhibitors for renal transplant.,” 2016.

[139] B. K. Kr¨amer, G. Montagnino, B. Kr¨uger, R. Margreiter, C. J. Olbricht, R. Marcen,

U. Sester, U. Kunzendorf, K.-H. Dietl, P. Rigotti, C. Ronco, S. H¨orsch, B. Banas,

F. M¨uhlbacher, and M. Arias, “Efficacy and safety of tacrolimus compared with

ciclosporin a in renal transplantation: seven-year observational results,” Transplant

international : official journal of the European Society for Organ Transplantation,

2015.

[140] S. Melnikov, H. Mayan, S. Uchida, E. J. Holtzman, and Z. Farfel, “Cyclosporine

metabolic side effects: association with the wnk4 system,” European journal of

clinical investigation, vol. 41, no. 10, pp. 1113–1120, 2011.

[141] G. Sune, E. Sarro, M. Puigmule, J. Lopez-Hellin, M. Zufferey, T. Pertel, J. Luban,

and A. Meseguer, “Cyclophilin b interacts with sodium-potassium atpase and is

References 95

required for pump activity in proximal tubule cells of the kidney,” PloS one, vol. 5,

no. 11, p. e13930, 2010.

[142] P. Wang and J. Heitman, “The cyclophilins,” Genome biology, vol. 6, no. 7, p. 226,

2005.

[143] S. Barik, “Immunophilins: For the love of proteins,” Cellular and molecular life

sciences : CMLS, vol. 63, no. 24, pp. 2889–2900, 2006.

[144] M. Tong and Y. Jiang, “Fk506-binding proteins and their diverse functions,” Cur-

rent molecular pharmacology, vol. 9, no. 1, pp. 48–65, 2015.

[145] K. H¨ocherl, F. Dreher, H. Vitzthum, J. K¨ohler, and A. Kurtz, “Cyclosporine a

suppresses cyclooxygenase-2 expression in the rat kidney,” Journal of the American

Society of Nephrology : JASN, vol. 13, no. 10, pp. 2427–2436, 2002.

[146] H. Meyer-Lehnert and R. W. Schrier, “Potential mechanism of cyclosporine a-

induced vascular smooth muscle contraction,” Hypertension (Dallas, Tex. : 1979),

vol. 13, no. 4, pp. 352–360, 1989.

[147] R. C. Harris, J. A. McKanna, Y. Akai, H. R. Jacobson, R. N. Dubois, and M. D.

Breyer, “Cyclooxygenase-2 is associated with the macula densa of rat kidney and

increases with salt restriction,” The Journal of clinical investigation, vol. 94, no. 6,

pp. 2504–2510, 1994.

[148] G.-H. Kim, “Renal effects of prostaglandins and cyclooxygenase-2 inhibitors,” Elec-

trolyte & blood pressure : E & BP, vol. 6, no. 1, pp. 35–41, 2008.

[149] A. Paliege, D. Mizel, C. Medina, A. Pasumarthy, Y. G. Huang, S. Bachmann, J. P.

Briggs, J. B. Schnermann, and T. Yang, “Inhibition of nnos expression in the macula

densa by cox-2-derived prostaglandin e(2),” American journal of physiology. Renal

physiology, vol. 287, no. 1, pp. F152–9, 2004.

[150] M. A. Iniguez, S. Martinez-Martinez, C. Punzon, J. M. Redondo, and M. Fresno,

“An essential role of the nuclear factor of activated t cells in the regulation of the

expression of the cyclooxygenase-2 gene in human t lymphocytes,” The Journal of

biological chemistry, vol. 275, no. 31, pp. 23627–23635, 2000.

References 96

[151] W. M. Flanagan, B. Corth´esy, R. J. Bram, and G. R. Crabtree, “Nuclear association

of a t-cell transcription factor blocked by fk-506 and cyclosporin a,” Nature, vol. 352,

no. 6338, pp. 803–807, 1991.

[152] G. Rovera, “The wistar institute,” Molecular medicine (Cambridge, Mass.), vol. 3,

no. 4, pp. 229–230, 1997.

[153] R. Schmitt, E. Klussmann, T. Kahl, D. H. Ellison, and S. Bachmann, “Renal ex-

pression of sodium transporters and aquaporin-2 in hypothyroid rats,” American

journal of physiology. Renal physiology, vol. 284, no. 5, pp. F1097–104, 2003.

[154] H. Schmale and D. Richter, “Single base deletion in the vasopressin gene is the cause

of diabetes insipidus in brattleboro rats,” Nature, vol. 308, no. 5961, pp. 705–709,

1984.

[155] P. J. Schultheis, J. N. Lorenz, P. Meneton, M. L. Nieman, T. M. Riddle, M. Flagella,

J. J. Duffy, T. Doetschman, M. L. Miller, and G. E. Shull, “Phenotype resembling

gitelman’s syndrome in mice lacking the apical na+-cl- cotransporter of the distal

convoluted tubule,” The Journal of biological chemistry, vol. 273, no. 44, pp. 29150–

29155, 1998.

[156] B. Eng, S. Mukhopadhyay, C. P. Vio, P. L. Pedraza, S. Hao, S. Battula, P. B. Sehgal,

J. C. McGiff, and N. R. Ferreri, “Characterization of a long-term rat mtal cell line,”

American journal of physiology. Renal physiology, vol. 293, no. 4, pp. F1413–22,

2007.

[157] F. A. Gesek and P. A. Friedman, “Mechanism of calcium transport stimulated

by chlorothiazide in mouse distal convoluted tubule cells,” The Journal of clinical

investigation, vol. 90, no. 2, pp. 429–438, 1992.

[158] C. A. Schneider, W. S. Rasband, and K. W. Eliceiri, “Nih image to imagej: 25 years

of image analysis,” Nature methods, vol. 9, no. 7, pp. 671–675, 2012.

[159] D. A. Huetteman and H. Bogie, “Direct blood pressure monitoring in laboratory

rodents via implantable radio telemetry,” Methods in molecular biology (Clifton,

N.J.), vol. 573, pp. 57–73, 2009.

References 97

[160] J. M. Foster, P. K. Carmines, and J. S. Pollock, “Pp2b-dependent no production in

the medullary thick ascending limb during diabetes,” American journal of physiol-

ogy. Renal physiology, vol. 297, no. 2, pp. F471–80, 2009.

[161] J. A. McCormick, C.-L. Yang, C. Zhang, B. Davidge, K. I. Blankenstein, A. S.

Terker, B. Yarbrough, N. P. Meermeier, H. J. Park, B. McCully, M. West,

A. Borschewski, N. Himmerkus, M. Bleich, S. Bachmann, K. Mutig, E. R. Argaiz,

G. Gamba, J. D. Singer, and D. H. Ellison, “Hyperkalemic hypertension-associated

cullin 3 promotes wnk signaling by degrading klhl3,” The Journal of clinical inves-

tigation, 2014.

[162] K. I. Blankenstein, A. Borschewski, R. Labes, A. Paliege, C. Boldt, J. A. Mc-

Cormick, D. H. Ellison, M. Bader, S. Bachmann, and K. Mutig, “Calcineurin in-

hibitor cyclosporine a activates renal na-k-cl cotransporters via local and systemic

mechanisms,” American journal of physiology. Renal physiology, vol. 312, no. 3,

pp. F489–F501, 2017.

[163] A. S. Terker, C. Zhang, K. J. Erspamer, G. Gamba, C.-L. Yang, and D. H. Ellison,

“Unique chloride-sensing properties of wnk4 permit the distal nephron to modulate

potassium homeostasis,” Kidney international, 2015.

[164] R. A. Lazelle, B. H. McCully, A. S. Terker, N. Himmerkus, K. Blankenstein,

K. Mutig, M. Bleich, S. Bachmann, C.-L. Yang, and D. H. Ellison, “Renal dele-

tion of 12 kda fk506-binding protein attenuates tacrolimus-induced hypertension,”

Journal of the American Society of Nephrology : JASN, 2015.

[165] C. E. Deppe, P. J. Heering, H. Tinel, E. Kinne-Saffran, B. Grabensee, and R. K.

Kinne, “Effect of cyclosporine a on na+/k(+)-atpase, na+/k+/2cl- cotransporter,

and h+/k(+)-atpase in mdck cells and two subtypes, c7 and c11,” Experimental

nephrology, vol. 5, no. 6, pp. 471–480, 1997.

[166] S. W. Lim, K. O. Ahn, M. R. Sheen, U. S. Jeon, J. Kim, C. W. Yang, and H. M.

Kwon, “Downregulation of renal sodium transporters and tonicity-responsive en-

hancer binding protein by long-term treatment with cyclosporin a,” Journal of the

American Society of Nephrology : JASN, vol. 18, no. 2, pp. 421–429, 2007.

[167] J. A. McCormick and D. H. Ellison, “The wnks: atypical protein kinases with

pleiotropic actions,” Physiological reviews, vol. 91, no. 1, pp. 177–219, 2011.

References 98

[168] N. Picard, K. Trompf, C.-L. Yang, R. L. Miller, M. Carrel, D. Loffing-Cueni, R. A.

Fenton, D. H. Ellison, and J. Loffing, “Protein phosphatase 1 inhibitor-1 deficiency

reduces phosphorylation of renal nacl cotransporter and causes arterial hypoten-

sion,” Journal of the American Society of Nephrology : JASN, vol. 25, no. 3, pp. 511–

522, 2014.

[169] N. G. Moss, S. L. Powell, and R. J. Falk, “Intravenous cyclosporine activates afferent

and efferent renal nerves and causes sodium retention in innervated kidneys in rats,”

Proceedings of the National Academy of Sciences of the United States of America,

vol. 82, no. 23, pp. 8222–8226, 1985.

[170] M.-Z. Zhang, B. Yao, Y. Wang, S. Yang, S. Wang, X. Fan, and R. C. Harris,

“Inhibition of cyclooxygenase-2 in hematopoietic cells results in salt-sensitive hy-

pertension,” The Journal of clinical investigation, vol. 125, no. 11, pp. 4281–4294,

2015.

[171] R. A. Fenton, L. Brønd, S. Nielsen, and J. Praetorius, “Cellular and subcellular

distribution of the type-2 vasopressin receptor in the kidney,” American journal of

physiology. Renal physiology, vol. 293, no. 3, pp. F748–60, 2007.

[172] K. Mutig, T. Borowski, C. Boldt, A. Borschewski, A. Paliege, E. Popova, M. Bader,

and S. Bachmann, “Demonstration of the functional impact of vasopressin signalling

in thick ascending limb by a targeted transgenic rat approach,” American journal

of physiology. Renal physiology, p. ajprenal.00126.2016, 2016.

[173] M. L. A. Kortenoeven, N. B. Pedersen, L. L. Rosenbaek, and R. A. Fenton, “Vaso-

pressin regulation of sodium transport in the distal nephron and collecting duct,”

American journal of physiology. Renal physiology, vol. 309, no. 4, pp. F280–99, 2015.

[174] A. S. Terker, C. Zhang, J. A. McCormick, R. A. Lazelle, C. Zhang, N. P. Meermeier,

D. A. Siler, H. J. Park, Y. Fu, D. M. Cohen, A. M. Weinstein, W.-H. Wang, C.-

L. Yang, and D. H. Ellison, “Potassium modulates electrolyte balance and blood

pressure through effects on distal cell voltage and chloride,” Cell metabolism, vol. 21,

no. 1, pp. 39–50, 2015.

[175] K. Mutig, T. Kahl, T. Saritas, M. Godes, P. Persson, J. Bates, H. Raffi, L. Rampoldi,

S. Uchida, C. Hille, C. Dosche, S. Kumar, M. Casta˜neda-Bueno, G. Gamba, and

References 99

S. Bachmann, “Activation of the bumetanide-sensitive na+,k+,2cl- cotransporter

(nkcc2) is facilitated by tamm-horsfall protein in a chloride-sensitive manner,” The

Journal of biological chemistry, vol. 286, no. 34, pp. 30200–30210, 2011.

[176] S. Bachmann, K. Mutig, J. Bates, P. Welker, B. Geist, V. Gross, F. C. Luft, N. Alen-

ina, M. Bader, B. J. Thiele, K. Prasadan, H. S. Raffi, and S. Kumar, “Renal effects

of tamm-horsfall protein (uromodulin) deficiency in mice,” American journal of

physiology. Renal physiology, vol. 288, no. 3, pp. F559–67, 2005.

[177] J. A. McCormick, K. Mutig, J. H. Nelson, T. Saritas, E. J. Hoorn, C.-L. Yang,

S. Rogers, J. Curry, E. Delpire, S. Bachmann, and D. H. Ellison, “A spak isoform

switch modulates renal salt transport and blood pressure,” Cell metabolism, vol. 14,

no. 3, pp. 352–364, 2011.

[178] G. Levy, F. G. Villamil, F. Nevens, H. J. Metselaar, P.-A. Clavien, G. Klintmalm,

R. Jones, M. Migliaccio, H. Prestele, and R. Orsenigo, “Refine: a randomized trial

comparing cyclosporine a and tacrolimus on fibrosis after liver transplantation for

hepatitis c,” American journal of transplantation : official journal of the American

Society of Transplantation and the American Society of Transplant Surgeons, vol. 14,

no. 3, pp. 635–646, 2014.