Dissecting the role of amyloid fibril deposition in the kidney in familial amyloidotic polyneuropathy

LUCIANA MARIA VAZ MOREIRA DISSECTING THE ROLE OF AMYLOID FIBRIL DEPOSITION IN THE KIDNEY IN FAMILIAL AMYLOIDOTIC POLYNEUROPATHY Tese de Candidatura ao grau de Doutor em Ciências Biomédicas submetida ao Instituto de Ciências Biomédicas Abel Salazar da Universidade do Porto. Orientador: Doutor Paulo Pinho e Costa Categorias: Professor Auxiliar Convidado e Investigador Afiliações: Instituto de Ciências Biomédicas Abel Salazar da Universidade do Porto e Instituto Nacional de Saúde Dr. Ricardo Jorge Coorientadora: Doutora Idalina Mª Melo Beirão Categorias: Professora Auxiliar Convidada Afiliações: Instituto de Ciências Biomédicas Abel Salazar da Universidade do Porto e Centro Hospitalar do Porto Coorientadora: Doutora Paola Romagnani Categoria: Professora Associada Afiliações: Universidade de Florença e Hospital Pediátrico Meyer de Florença De acordo com o disposto no nº 1, do artigo 34º do Decreto-Lei nº 74/2006, publicado em Diario da Républica 1ªserie nº 60 de 24 de Março de 2006 e republicado pelo Decreto Lei nº 115/2013 publicado no Diario da Républica 1ª serie de 7 de Agosto de 2013, utilizaram-se nesta tese resultados contidos nos trabalhos já publicados ou em vias de publicação, Em revista de circulação internacional com arbitragem científica: Moreira L, Beirão J, Beirão I, Costa P. Oligomeric TTR V30M aggregates compromise cell viability, erythropoietin gene expression and promoter activity in the human hepatoma cell line Hep3B. Amyloid. DOI:10.3109/13506129.2015.1007497 (in press). Em acta de encontro científico: Moreira L, Ballerini L, Peired A, Sagrinati C, Parente E, Angelotti ML, Ronconi E, Lazzeri E, Mazzinghi B, Lacerda P, Beirão I, Lasagni L, Costa PP, Romagnani P. TTRV30M oligomeric aggregates inhibit proliferation of renal progenitor cells but maintain their capacity to differentiate into podocytes in vitro. The Proceedings of the XIIIth International Symposium on Amyloidosis, May 6-10, 2012, Groningen, The Netherlands, GUARD (Groningen Unit for Amyloidosis Research & Development), UMC Groningen, c2013; 9093. ISBN 978-90-821593. No cumprimento do Decreto-Lei supra mencionado, a autora desta tese declara que interveio na concepção e execução do trabalho experimental, assim como na interpretação e discussão dos resultados e na sua redação. Outros artigos publicados pela autora durante o seu doutoramento que, não tendo sido usados nos resultados desta tese, estão no âmbito do tema aqui desenvolvido: Lacerda PC*, Moreira L*, Vitorino R, Costa PP. Use of MALDI-TOF Mass Spectrometry to Assay the Transthyretin V30M Mutation in Serum From a Liver Transplant Donor: A Case Report. Transplantation (in press). * The first two authors should be regarded as joint First Authors. Beirão JM, Moreira LM, Oliveira JC, Menéres MJ, Pessoa BB, Matos ME, Costa PP, Torres PA, Beirão IB. Aqueous humor erythropoietin levels in open-angle glaucoma patients with and without TTR V30M familial amyloid polyneuropathy. Mol Vis. 2014 Jul 2;20:970-6. Beirão JM, Moreira LV, Lacerda PC, Vitorino RP, Beirão IB, Torres PA, Costa PP. Inability of mutant transthyretin V30M to cross the blood-eye barrier. Transplantation. 2012 Oct 27;94(8):e54-6. Este trabalho foi financiado pela Fundação para a Ciência e Tecnologia através de uma bolsa de doutoramento (SFRH/BD/46441/2008) e co-financiado pelo POPH/FSE. Aos meus pais. Agradecimentos Começo por agradecer aos directores das instituições de acolhimento que tornaram possível a realização deste trabalho, quer pelo financiamento disponibilizado como pela utilização das instalações e equipamentos: Departamento de Genética Humana (DGH) do Instituto Nacional de Saúde Dr. Ricardo Jorge (INSA), Unidade Multidisciplinar de Investigação Biomédica (UMIB) do Instituto de Ciências Biomédicas Abel Salazar (ICBAS) da Universidade do Porto e Departamento de Fisiopatologia Clínica da Universidade de Florença. Agradeço à Fundação para a Ciência e a Tecnologia pelo apoio financeiro disponibilizado através da bolsa de doutoramento. Ao Dr. Paulo Pinho e Costa agradeço, antes de mais, o facto de me ter recebido novamente no seu laboratório após a minha vinda da Alemana sem qualquer oposição, e após a defesa do Mestrado ter aceitado ser meu orientador de doutoramento. O meu percurso no seu laboratório já vai longo, mas tenho sempre que lhe agradecer a confiança depositada em mim e no meu trabalho. Sempre me deu liberdade para planear e desenvolver o trabalho de forma autónoma e independente, mas sem as suas sugestões pertinentes e orientação a realização desta tese não teria sido possível. À Dra. Idalina Beirão agradeço a dedicação, as discussões científicas tão úteis para o trabalho e acima de tudo o estímulo constante para continuar face às adversidades. Foi a “fada-madrinha” das principais experiências e é o meu exemplo de perseverança e força de vontade. Alla professoressa Paola Romagnani ringrazio la colaborazione, orientamento e mi avere ricevuto benissimo nel suo laboratorio per tanti mesi. È stata una grande esperienza e un apprendimento costante e stimolante in una área scientifica che mi era poco conosciuta. Alla Laura Lasagni, Elena Lazzeri, Maria Lucia Angelotti, Lara Ballerini, Costanza Sagrinati, Eliana Parente, Anna Peired, Elisa Ronconi e Benedetta Mazzinghi, ringrazio prima di tutto, il modo accogliente con che mi hanno accolto e fatto sentire subito integrata sia in laboratorio che fuori. So che avrò sempre da voi una grande amicizia. Inoltre, voglio ringraziarvi tutto che mi avete insegnato in laboratorio perché ho sempre imparato qualcosa da ciascuno di voi. iv Abbreviations v Abbreviations most frequently used in the text AGEs – advanced glycation end products AL – immunoglobulin amyloidosis ATF3 – activating transcription factor 3 ATTR amyloidosis – amyloidosis caused by transthyretin amyloid deposition ATTRV30M – transthyretin with a methionine-for-valine substitution at position 30 ATTRV30M amyloidosis - amyloidosis caused by the mutant transthyretin with a methionine-for-valine substitution at position 30 CNS – central nervous system CSF – cerebrospinal fluid Ct – threshold cycle DN – Diabetic nephropathy DMOG – dimethyloxalylglycine DLS – Dynamic Light Scattering EPO – erythropoietin EPOR – EPO receptor ER – endoplasmic reticulum ERK – extracellular signal-regulated kinase FACS – fluorescence-activated cell sorting FAP – Familial Amyloidotic Polyneuropathy FBS – fetal bovine serum FIH1 – HIF inhibition factor GAG – sulfated glycosaminoglycans GATA-2 – GATA binding protein 2 GATA-4 – GATA binding protein 4 HEK293T – human embryonic kidney 293 cell line Hep3B – human hepatocellular carcinoma HIF-1 – hypoxia inducible factor 1 HIF-2 – hypoxia inducible factor 1 HNF-4 – hepatocyte nuclear factor 4 HRE – hypoxia response element HUVECs – primary human umbilical vein endothelial cells IHC – Immunohistochemistry IL-1β – interleukin-1β IMAC – immobilized metal ion affinity chromatography Abbreviations vi iNOS – inducible nitric oxide synthase IPTG – Isopropyl β-D-1-thiogalactopyranoside JAK2 – Janus kinase 2 LT – liver transplantation MAPK – mitogen-activated protein kinase MTS – 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2Htetrazolium MTT – 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide NF-κB – nuclear factor-κB NO – nitric oxide PDGF – platelet-derived growth factor PHDs – prolyl-hydroxylases qPCR – quantitative PCR RAGE – receptor of advanged glycation end products RBP – retinol binding protein RFLP – restriction fragment length polymorphism ROS – reactive oxygen species RPC – renal progenitor cells CD133+CD24+ RPE – human retinal pigment epithelial cells RT-PCR – real time PCR SAA – senile systemic amyloidosis SAP – Serum Amyloid P component SD – standard deviation SDS-PAGE – sodium dodecyl sulfate polyacrylamide gel electrophoresis SH-SY5Y – human neuroblastoma cell line siRNAs – small interfering RNAs T4 – thyroxine hormone TBP – TATA-binding protein TFA – trifluoroacetic acid ThT – thioflavin T TNF-α – tumour necrosis factor-α TTR – transthyretin TTRV30M – transthyretin with a methionine-for-valine substitution at position 30 wt-TTR – wild-type TTR Resumo vii Resumo A Polineuropatia Amiloidótica Familiar (PAF) ou amiloidose ATTRV30M é uma doença neurodegenerativa, autossómica dominante, causada pela deposição de amilóide extracelular de transtirretina mutante (TTRV30M), afectando principalmente o sistema nervoso periférico. É caracterizada por uma polineuropatia periférica sensitivo-motora progressiva e disfunção autonômica, com manifestações renais, cardíacas e oculares. A anemia afecta cerca de 25% dos doentes PAF sintomáticos e é caracterizada por uma produção ineficiente de eritropoietina (EPO), independentemente da presença de insuficiência renal. Precede por vezes a doença clínica, sugerindo um bloqueio das células renais produtoras de EPO. Excluímos anteriormente um efeito inibitório dos depósitos de amiloide fibrilar e da TTRV30M em circulação, mas o papel dos agregados de TTR não-fibrilares na produção de EPO renal ainda precisava ser explorado. Os agregados não-fibrilares ou oligómeros de TTR são citotóxicos, induzindo stress oxidativo e a expressão de moléculas relacionadas com a apoptose, e a secreção de citoquinas pró-inflamatórias. Alguns destes marcadores também são capazes de inibir a produção de EPO. A expressão do gene EPO é regulada ao nível da transcrição pelos factores de transcrição HIF, NF-kB e GATA-2. O HIF induz a expressão de EPO em condições de hipóxia, por ligação ao enhancer na região 3’, enquanto o GATA-2 e NF-kB inibem a sua expressão por ligação ao promotor na região 5’ do gene. Neste trabalho propusémo-nos explorar os mecanismos moleculares envolvidos no bloqueio da produção de EPO na PAF. Utilisando diferentes modelos de cultura de células, foi avaliada a influência dos agregados oligoméricos de TTR na viabilidade celular, capacidade de diferenciação, expressão do gene da EPO e atividade do seu promotor. Oligómeros de TTR foram preparados envelhecendo a proteína a pH fisiológico, seguido de 5 minutos de agitação. As preparações que continham predominantemente espécies de 300 nm foram usadas para os ensaios celulares. Uma redução modesta mas estatisticamente significativa da viabilidade celular foi induzida pela TTRV30M oligomérica após 24 horas de incubação, independentemente do tipo de célula. Concomitantemente com a redução da viabilidade, foi observado um aumento da actividade das caspases 3/7 em células Hep3B, SH-SY5Y e RPE expostas aos agregados oligoméricos de TTRV30M. Estes resultados estão de acordo com estudos anteriores realizados noutros outros tipos de células, e mostram que a apoptose está implicada na perda de viabilidade celular. Resumo viii A influência dos agregados oligoméricos sobre a diferenciação celular foi avaliada em células progenitoras renais (RPC). Estas têm potencial de diferenciação em podócitos e células tubulares renais. Embora os oligómeros de TTRV30M tenham inibido a proliferação das RPC, eles não influenciaram a sua capacidade de se diferenciar em podócitos funcionalmente maduros, e, portanto, não devem comprometer a regeneração dos tecidos. A expressão do gene da EPO foi avaliada por PCR em tempo real. Células Hep3B e RPE tratadas com TTRV30M oligomérica mostraram uma redução significativa (aproximadamente 50%) da expressão de EPO após 24 horas em condições de normóxia quando comparado com o controlo e com a exposição à forma tetramérica. Os oligómeros de TTR normal também reduziram a expressão de EPO em 22% em células Hep3B em normóxia, quando comparada com a com a exposição à forma tetramérica. Estes resultados apoiam a nossa hipótese de que espécies oligoméricas citotóxicas estarão envolvidas na génese da anemia em doentes com PAF. Além disso, mostrámos recentemente que as concentrações de EPO no humor aquoso de olhos com glaucoma de doentes não-PAF estão significativamente aumentadas relativamente a olhos sem glaucoma. No entanto, em olhos com glaucoma de doentes PAF os níveis de EPO não aumentam e mantêm níveis semelhantes aos dos olhos normais, o que mostra a incapacidade destes doentes regularem positivamente a produção ocular de EPO. Um ensaio repórter foi construído com um fragmento de PCR do promotor da EPO, que contém os locais de reconhecimento para GATA-2 e NF-kB, ligados ao gene da luciferase para avaliar o papel dos factores de transcrição do promotor. Células Hep3B transfectadas e expostas durante 24 horas a TTRV30M oligomérica mostraram uma redução significativa da actividade do promotor de EPO (53%) quando comparado com o controlo e exposição à forma tetramérica. Estes resultados sugerem que a expressão de EPO é inibida pelos agregados oligoméricos de TTRV30M, pelo menos em parte através da inibição da actividade do promotor. Imunofluorescência e imunohistoquímica foram realizadas para explorar o envolvimento do NF-kB e GATA-2 na redução da atividade do promotor da EPO, mas não foram observadas diferenças significativas entre as células tratadas com oligómeros ou com a forma tetramérica de TTR. Em conclusão, os agregados de TTR não-fibrilares podem inibir a produção de EPO e contribuir para o aparecimento precoce da anemia nos doentes PAF. Estudos adicionais são necessários para elucidar os mecanismos que levam à inibição da EPO, a fim de proporcionar marcadores úteis para a avaliação do doente e, eventualmente, novos alvos terapêuticos. As células RPE, sendo produtoras quer de TTR como de EPO, poderão constituir um bom modelo para estes estudos futuros. Abstract ix Abstract ATTRV30M amyloidosis or Familial Amyloidotic Polyneuropathy (FAP) is a neurodegenerative, autosomal dominant disease, caused by the extracelular amyloid deposition of a mutant transthyretin (TTRV30M), affecting particularly the peripheral nervous system. It is characterized by progressive sensorimotor peripheral polyneuropathy and autonomic dysfunction, and renal, cardiac, and ocular manifestations. Anemia affects about 25% of symptomatic FAP patients and courses with low erythropoietin (EPO) levels, independently of the presence of renal failure. It sometimes precedes clinical disease, suggesting a blockage of kidney’s EPO-producing cells. We had previously excluded an inhibitory effect of the amyloid fibril deposits and of circulating TTRV30M, but the role of early non-fibrillar TTR aggregates on renal EPO production still needed to be explored. Early non-fibrillar TTR aggregates are highly cytotoxic, inducing oxidative stress and the expression of apoptosis-related molecules, and secretion of proinflammatory cytokines. Some of these markers are also capable of inhibiting EPO production. EPO gene expression is regulated at the transcriptional level by HIF, NF-kB and GATA-2 transcription factors. HIF induces EPO expression under hypoxic conditions, by binding to the enhancer in the 3’ region, whereas GATA-2 and NF-kB inhibit its expression by binding to the promoter in the 5' region of the gene. In this work, our aim was to explore the molecular mechanisms involved in the blockage of EPO production in FAP. Using different cell culture models, we assessed the influence of oligomeric TTR aggregates on cell viability, differentiation capacity, EPO gene expression and promoter activity. TTR oligomers were prepared by aging the protein at physiological pH, followed by 5 minutes of stirring. Preparations containing species mainly of 300 nm were used for the cell culture-based assays. A modest but statistically significant reduction in cell viability was induced by oligomeric TTRV30M after 24 hours of incubation, independently of the cell type. Concomitantantly with the reduction of cell viability, an increase in caspase 3/7 activity was seen in Hep3B, SH-SY5Y and RPE cells exposed to the oligomeric TTRV30M aggregates. These results agree with those of previous studies performed with other cell types, and implicate apoptosis in the loss of viability. The influence of oligomeric aggregates on cell differentiation was evaluated on renal progenitor cells (RPC). RPC have self-renewal and multidifferentiation potential into podocytes and renal tubular cells. Although TTRV30M oligomers inhibited RPC Abstract x proliferation, they did not influence their capacity to differentiate into functionally mature podocytes, and thus should not compromise tissue regeneration. EPO mRNA expression was evaluated by real-time PCR. Hep3B and RPE cells treated with oligomeric TTRV30M showed a significant reduction (about 50%) of EPO mRNA expression after 24 hours in normoxia, when compared to the control and cells exposed to the tetrameric form. Oligomers from normal (wild type) TTR also reduced EPO expression by 22% in normoxic Hep3B cells when compared to exposure to the tetrameric form. These results support our hypothesis that cytotoxic oligomeric species are involved in the genesis of anemia in FAP patients. Besides this evidence, we recently showed that EPO concentrations in the aqueous humor of glaucomatous eyes of non-amyloidotic patients are significantly increased relatively to normal non-glaucomatous eyes. However, in glaucomatous eyes of FAP patients the EPO levels did not increase and maintained similar levels to those of control eyes, showing an inability of these patients to upregulate ocular EPO production. A reporter assay was constructed with a PCR fragment of the EPO promoter, containing the recognition sites for GATA-2 and NF-kB linked to the luciferase gene, to evaluate the role of transcription factors targeting the promoter. Transfected Hep3B cells exposed for 24 hours to oligomeric TTRV30M showed a significant reduction of the erythropoietin promoter activity (53%) when compared to the control and exposure to the tetrameric form. These results suggest that EPO expression is inhibited by oligomeric TTRV30M aggregates, at least in part through inhibition of promoter activity. Immunofluorescence and immunohistochemistry were performed to explore the involvement of NF-kB and GATA-2 in the reduction of EPO promoter activity, but no significant differences were observed between the oligomeric TTR-treated and tetrameric TTR-treated cells. In conclusion, early non-fibrillar TTR aggregates can inhibit EPO production and may contribute to the early onset of anemia in these patients. Further studies are needed to elucidate the mechanisms that lead to EPO inhibition, in order to provide useful markers for patient evaluation, and possibly new targets for therapeutic development. RPE cells, as producers of both TTR and EPO, could be a good model for these future studies. INTRODUCTION Introduction 3 1. Amyloid and amyloidosis: short story of its discovery Amyloid is an insoluble substance that deposits in tissues and organs, mainly in the extracellular spaces, leading to progressive organ dysfunction and disease [1-3]. Amyloidosis is the group of diseases associated with amyloid deposition [4]. Different types of amyloidosis exist, depending on the protein that originates the fibrillar deposits. Some proteins undergo conformational changes of their structure due to abnormal polymeric assemblies of its subunits, and form amyloid fibrils. Amyloid deposits are composed by non-branching fibrils with a β-sheet structure and approximately 10 nm in diameter [4]. The fibrils bind the dye Congo red and exhibit green birefringence when the Congo red-stained deposits are viewed with polarized light. 1.1. Historical review The historical review of the discovery of amyloid presented here had as major bibliographic sources the detailed accounts of Sipe JD et al. [5] and Kyle RA et al. [6]. The first description of what we now call amyloidosis may have occured in 1639, when Nicolaus Fontanus reported the autopsy of a young man who had an abscess in the liver and a large spleen filled with white stones, probably a “sago spleen” amyloidosis. In 1838, Matthias Schleiden, a German botanist, used the term amyloid to describe a normal amylaceous constituent of plants. Later, in 1854, Rudolph Virchow used the same term to describe the corpora amylacea of the nervous system and the substance implicated in lardaceous degeneration. He found that, using iodine, these stained blue, turning violet upon the subsequent addition of sulfuric acid. This peculiar reaction made Virchow consider that these lardaceous deposits were identical to starch. In 1859, Carl Friedreich and August Kekule saw that this amyloid “mass” had a proteic nature instead of carbohydrate, unlike amylon or cellulose. From this point, amyloid has been considered a protein material and, later, as a group of proteins with a propensity to undergo conformational changes that result in the formation of fibrils. The name amyloid prevailed nonetheless. Nowadays, it is known that amyloid deposits in tissues have also other non-fibrillar components besides the protein, such as proteoglycans (heparan sulfate or chondroitin sulfate type), basement membrane constituents (laminin, fibronectin and collagen IV), Introduction 10 northern regions of Portugal, mainly Póvoa do Varzim, where allele frequency is approximately 1:550 [36, 41]. 2.2. Clinical features of ATTRV30M amyloidosis ATTRV30M amyloidosis is a systemic disease characterized by progressive sensorimotor peripheral polyneuropathy and autonomic dysfunction, with renal, cardiac, and ocular manifestations, among others [42]. Usually symptoms begin in the third to fourth decade of life, and develop gradually for 10-20 years, leading to death. Progressive impairment of thermal sensitivity and pain are common initial symptoms. The disease progresses with lowering of the general state of health, alimentary and sexual dysfunction, malabsorption, urinary bladder dysfunction, abnormal glomerular function, cardiac insufficiency and vitreous opacities [43-46]. Besides the peripheral nerves, where the amyloid is preferentially deposited causing myelin sheet destruction, other organs, such as kidney, pancreas, heart, stomach, aorta, skin and eye are affected [47]. Although TTR is produced mainly by the liver, this organ is not significantly affected. Figure 2 – Schematic representation of clinical manifestations in FAP patients. From Ueda et al. [47]. Introduction 11 In this work, more attention will be given to renal and ocular complications, and in particular, to the expression of erythropoietin by these two organs in FAP patients. This subject will be discussed later in further detail. 2.3. Therapeutic strategies for ATTR amyloidosis Advances are being made in the treatment of ATTR amyloidosis and many therapeutic strategies have been proposed, which act on 3 important checkpoints: A) Blocking the synthesis of TTR protein B) Stabilization of the TTR tetramer to inhibit TTR disaggregation C) Disruption and promotion of TTR amyloid fibrils clearance Figure 3 – Schematic representation of the mechanism proposed for TTR amyloidogenesis and some therapeutic strategies for FAP. From Hanna 2014 [48]. Block TTR synthesis Liver transplantation Since 1990, liver transplantation (LT) has become a therapeutic option for FAP patients, as it eliminates more than 95% of the abnormal protein from circulation [49-50]. Approximately 120 LTs are performed worldwide each year [http://www.fapwtr.org]. When LT is performed in the early stage of the disease, it can prolong survival, halt the Introduction 12 progression of clinical manifestations and even improve autonomic and, to a lesser extent, peripheral nerve function [51-53]. A strategy to manage the scarcity of healthy livers for transplantation was adopted with sequential or domino LT, in which FAP patients receive a healthy liver from a deceased donor and their liver is transplanted into patients with malignant or end-stage liver diseases. As FAP only manifests after the 2nd or 3rd decade of life, it would be expected that a FAP liver recipient would not manifest the disease. However, some recipients of FAP livers started developing TTR amyloid deposits and disease symptoms less than 10 years after the surgery [54-56]. LT has important drawbacks: it is extremely expensive; transplanted patients must be treated with immunosuppressants lifelong; non-symptomatic carriers of TTR mutations as well as FAP patients in advanced stage do not undergo liver transplantation. Although circulating mutant TTR is virtually eliminated, there are reports of continuing formation and deposition of fibrils from wild-type TTR, causing cardiomyopathy and/or neuropathy after liver transplantation, findings that seem similar to those observed in senile systemic amyloidosis (SSA) [57-60]. Also, LT does not prevent the production of mutant TTR in the cerebrospinal fluid and in the eyes, where the mutant protein is still secreted by choroid plexus and retinal pigmented epithelium, respectively. Ocular and central nervous system (CNS) problems may occur, highlighting the importance of non-neural dysfunction in posttransplant patient management [61-62]. Gene therapy Antisense oligonucleotides (ASO) and small interfering RNAs (siRNAs) are effective genesilencing tools that could be promising for FAP gene therapy [48]. Blocking hepatocyte synthesis of TTR would prevent the production of both mutant and wild type TTR, which could be a potential treatment for both hereditary and acquired ATTR amyloidosis. Phase 3 studies are currently ongoing for some ASOs and siRNAs [48, 63]. Antisense oligonucleotides (ASOs) are synthetic single stranded oligomers designed to be complementary to a specific region in a target mRNA, promoting its degradation, thus preventing production of the associated protein [64]. The drug ISIS-TTR Rx from Isis Pharmaceuticals (Carlsbad, CA) targets the TTR mRNA preventing the production of both mutant and wild-type TTR protein. This ASO suppresses TTR mRNA levels in the liver and in the choroid plexus of the brain [65-67]. siRNAs are sequence-specific post-transcriptional gene silencing molecules. Patisiran, also known as ALN-TTR02, from Alnylam Pharmaceuticals (Cambridge, MA) is a lipid nanoparticle formulation of a synthetic siRNA that selectively silences both mutant and wild-type TTR gene expression, both in vitro and in vivo [65, 68-69]. Introduction 13 Stabilize the TTR tetramer TTR binding of its natural ligand thyroxine (T4) stabilizes the tetramer against dissociation. The same strategy has been tested using small molecules that could bind to the TTR T4 pocket to kinetically stabilize the native tetramer and avoid the conformational changes that lead to amyloid fibril formation [70]. Diflunisal, a non-steroidal anti-inflammatory drug (NSAID) with a molecular structure similar to T4, binds to the T4 binding sites on the TTR tetramer, stabilizing it and preventing acid mediated fibril formation in vitro [71-74], as well as in serum of amyloidosis patients, without adverse effects [73, 75]. A 2 years long clinical trial showed that Diflunisal treated patients had substantially less polyneuropathy progression compared to placebo [76]. Tafamidis (Vyndaqel) or Fx-1006A, like Diflunisal, is a small molecule that selectively binds to the T4 binding site and stabilizes the TTR tetramer [77], but without NSAID activity. FAP patients treated with Tafamadis have shown a significant benefit relatively to an untreated control group. Tafamidis stabilizes plasmatic TTRV30M protein, slowing disease progression [77-78]. It became the first pharmacological treatment for FAP patients to be approved in Europe [79]. Clearance of TTR amyloid fibrils The presence of other compounds in the amyloid fibrils besides the core protein led several groups to investigate the possibility to target them as a strategy to promote clearance of amyloid fibrils, regardless of the amyloid type. Doxycycline, an antibiotic, was shown to be an effective fibril disrupter, disaggregating TTR amyloid fibris in mice transgenic for human TTRV30M, promoting amyloid deposit reabsorption [80-81]. TUDCA (Tauroursodeoxycholic acid), a biliary acid with antioxidant and antiapoptotic activities, significantly decreased the amount of TTR aggregates, as well as oxidative and apoptotic biomarkers associated with disease, also in a transgenic mouse model [82-83]. A trial of a combination of TUDCA with Doxycyline was conducted in aged FAP mice [82] and was more effective than either doxydoxycycline or TUDCA alone. It significantly lowered TTR deposition and associated tissue markers and also disaggregated mature amyloid deposits in the gastrointestinal tract. EGCG (Epigallocatechin-3-gallate), the predominant polyphenol in green tea, has shown in vitro the capacity to both inhibit fibril formation and also disrupt amyloid fibrils by converting existing fibrils into non-fibril conformers [84-86]. Introduction 14 Anti-SAP antibodies could contribute to clear amyloid fibrils. Serum Amyloid P component (SAP) is a plasma glycoprotein universally associated with amyloid fibrils. Encouraging data was seen in animal models as well as in a heterogeneous group of amyloid patients treated with anti-SAP antibodies [87-88]. 2.4. Diagnosis As for other amyloidosis, the diagnosis is based on histological examination, genetic testing or mass spectrometry. ATTRV30M amyloidosis results from a single adenine for guanine nucleotide change in exon 2 [89], which creates a restriction site for the enzyme NsiI. This facilitates the molecular diagnosis, which can be performed by PCR followed by RFLP, or by real-time PCR (rtPCR) genotyping based on melting curve analysis. Also, amino acid substitution results in a known mass shift in the protein molecular weight [90], making it possible to use mass spectrometry analysis [91-93]. This is particularly useful in the setting of liver transplantation. Recently, our group described a case report of a FAP patient who underwent orthotopic liver transplantation from a cadaveric donor [94]. Continuing deterioration in this patient raised the suspicion, confirmed in 2012, that the liver donor was also a TTRV30M carrier, so retransplantation was proposed and carried out. Immunoprecipitation of TTR from the serum of the cadaveric donor, followed by mass spectrometry analysis [95] enabled to confirm the absence of TTRV30M, concluding that this second time the transplanted liver was FAP free. Introduction 15 3. Renal and ocular complications in ATTRV30M Amyloidosis: its association with low erythropoietin production TTRV30M amyloid deposits are commonly found both in the kidney as well as in the eye of FAP patients and give rise, respectively, to renal and ocular problems in these patients. 3.1. Renal complications: nephrotic syndrome and anemia In ATTRV30M amyloidosis, renal amyloid deposition occurs mainly in the glomerular mesangium and medulla, with a typical involvement of the distal convoluted tubule and Henle's loop, which may lead to tubular atrophy and loss of tubular function [27]. Clinical nephropathy manifests initially as a microalbuminuria and then progresses to proteinuria and renal failure. In FAP patients, nephrotic syndrome may be associated with anemia with decreased levels of EPO, sometimes below the lower limit of the normal range, without associated iron deficit [96]. These patients do not respond to iron therapy but treatment with recombinant EPO is effective [97]. According to World Health Organization criteria, anemia is defined as a concentration of hemoglobin (Hb) <13 g / dL in men and <12 g / dL in women [98]. 3.1.1. Defective EPO production by the kidney in ATTRV30M amyloidosis Anemia has been described in ATTRV30M amyloidosis patients:  Moderate normocytic normochromic anemia was observed in 39% of the Swedish FAP patients [99];  macrocytic and hypochromic anemia was reported in a group of 35 Japanese FAP patients [100].  normocytic and normochromic anemia was present in 25% from a total of 165 symptomatic FAP patients, even in the presence of normal renal function, and is associated with a defective renal production of EPO, revealed by serum levels lower than expected [96]. A deficit of EPO in Portuguese FAP patients is an early event that was observed independently of the presence of renal failure and sometimes preceding clinical disease Introduction 16 [96]. Serum EPO levels were, on average, 11.2+6.7 mU/mL, lower than the expected levels (35+13.9 mU/ml) for the degree of anemia, while iron stores, B12 vitamin, and serum folate levels were normal in these patients. Normalization of iron status was insufficient for the correction of anemia, but therapy with low doses of recombinant EPO was effective [97], excluding a defective response of the bone marrow as a cause of anemia in these patients. Circulating EPO is mainly produced by the kidney in the adult. The observed low EPO production suggests a defect of the EPO-producing cells, which could be related to either the presence of amyloid deposits in the renal interstitium or with other factors, such as circulating TTRV30M itself. The amyloid deposits present in renal biopsies were found to have no correlation with serum EPO levels, independently of the neuropathy score, the amount of amyloid deposition or the renal clinical manifestations [101]. Anemia in liver transplant recipients is usually due to the side-effects of immunosuppressive therapies, iron deficiency, renal failure and post-transplant lymphoproliferative disorders [102]. A study performed in cirrhotic patients showed that anemia affected 85% of these patients before liver transplantation. After liver transplantation, anemia decreased to 18% [103]. In contrast, in FAP patients the prevalence of anemia increased even after liver transplantation, and defective endogenous EPO production persisted [104], excluding an inhibitory effect of the circulating TTRV30M on the EPO-producing cells. Pro-inflammatory cytokines can inhibit EPO gene expression, contributing to the anemia of chronic disease [105]. In ATTRV30M amyloidosis, inflammation is observed, particularly with up-regulation of TNF-α, macrophage colony-stimulating factor and IL-1β [106-107], which could explain the low EPO levels in these patients. However, Beirão et al. analyzed 24 FAP patients and found no evidence of systemic inflammation, as no significant differences were found on interleukin-6, transferrin saturation, ferritin and hepcidin-25 [108]. 3.2. Ocular complications: vitreous opacities and glaucoma ATTRV30M amyloidosis is associated with several ocular manifestations such as lacrimal dysfunction, pupillary disturbances, changes in the conjunctiva, presbyopia [109], dry eye [110], vitreous opacities, which may occur before any other systemic manifestation of the disease [111], and, most seriously, severe glaucoma. Liver transplantation in FAP patients Introduction 17 proved unable to halt the progression of these ocular manifestations, probably due to the continued production of the mutated TTR by the retinal pigment epithelial cells [62]. A retrospective study of 477 symptomatic FAP patients was performed by Beirão et al. which showed that these patients have amyloid deposits in the iris, in the anterior lens capsule and in the vitreous. Vitrectomy with complete removal of the vitreous is usually carried out, but when the vitrectomy is incomplete new amyloid deposits are formed due to continuing deposition in the remaining vitreous [112]. Glaucoma can develop rapidly in FAP patients and, if not treated, may lead to blindness. A correlation was found between vitrectomy and glaucoma, with vitrectomy favoring the onset or worsening of glaucoma [113]. 3.2.1. Defective EPO production by the eye in ATTRV30M amyloidosis Glaucoma causes an increase in intraocular pressure, which leads to activation of neuroprotective mechanisms. Studies have reported upregulation of EPO expression and an increased intravitreal EPO concentration in some ocular disorders [114], which may reflect the cytoprotective function of EPO in response to hypoxia, ischemia, and inflammation [115]. Recently, our group found that EPO concentrations in the aqueous humor of glaucomatous eyes of non-FAP patients are significantly increased relatively to normal non-glaucomatous eyes, probably as a protective role. However, in glaucomatous eyes of FAP patients the EPO levels did not increase and maintained similar levels to those of control eyes [116]. These results show an inability of FAP patients to upregulate EPO production both systemically by the kidney and locally by the pigmented epithelium. Glaucoma is the second leading cause of blindness worldwide [117]. Vascular abnormalities and altered blood flow at the optic nerve head may lead to local hypoxia, accelerating neuronal cell death in patients. The hypoxia inducible factor 1 (HIF-1) is thought to be involved in the pathology of glaucoma, as increased activation of HIF-1 was found in glaucomatous eyes and localization of this protein was correlated with regions of visual field defects [118]. HIF is one of the main regulators of EPO expression, by inducing it in situations of hypoxia. In glaucoma, as a consequence of HIF activation, EPO levels are strongly elevated, probably as a cytoprotective response. The mechanisms responsible for the low expression of EPO in FAP patients, whether as a response to anemia, or as a response to ocular damage, as in the case of glaucoma, remain unexplained. What is certain is that these patients do not increase EPO levels in response to stimuli to which a non-PAF patient would respond with an increase in expression of this cytokine/hormone. Introduction 18 4. Erythropoietin Erythropoietin (EPO) is a hormone essential for red blood cell production. A moderate reduction in hemoglobin concentration is sufficient to increase EPO mRNA expression, which occurs within minutes of the onset of hypoxia, reaching a maximum after 6 hours [119-121]. Daily, it stimulates proliferation and differentiation of about 2x1011 erythroid progenitor cells in the bone marrow, contributing to the control of blood oxygen capacity throughout the body [122]. EPO is an endocrine, paracrine and autocrine hormone. Besides its hematopoietic function, EPO has been shown to be a cytoprotective hormone. Among other effects, EPO antagonizes the activity of pro-inflammatory cytokines, has neuroprotective functions and promotes healing through stimulation of angiogenesis and capillary growth [123]. 4.1. Erythropoietin structure Human EPO is a glycoprotein of 30.4 kDa encoded by 5 exons located in chromosome 7 as a single copy gene. Translation of the EPO gene results in a polypeptide chain of 193 amino acids that is cleaved posttranslationaly, both at the Nand C-terminal sites. The secreted protein has 165 amino acids [124]. About 40% of the molecular weight of EPO is due to carbohydrate chains. EPO has 4 glycosilated side chains that are important to its biological function by conferring thermal and structural stability, protection against free radicals, increased plasma half-life and selectivity [122, 125-126]. EPO has 4 α-helices and 2 dissulfide bonds with 3 asparagine N-glycosilation and 1 serine O-glycosilation sites. The sialic acid residues attached to the 4 carbohydrate chains are particularly important for the maintenance of in vivo half-life and biological activity [126-127]. 4.2. Sites of erythropoietin production During fetal life EPO is produced by the liver, whereas in the adult it is mainly produced by the kidney [128]. The molecular mechanisms underlying this switch are poorly understood, but are thought to involve the transcription factor GATA-4 [129], which is highly expressed by hepatocytes only in the fetal liver. Its inhibition leads to a dramatic reduction in Epo gene transcription in Hep3B cells. Introduction 19 Many efforts were done to identify the renal EPO-producing cells. Evidence has been provided for different locales including: renal glomeruli [130], peritubular interstitial or endothelial cells in anemic mouse [131-132], peritubular interstitial cells in hypoxic monkey [133], tubular epithelial cells [134-135] and proximal tubular cells [136-137]. In 2010, our group identified distal tubular cells and cortical collecting tubules as the major site of EPO production in normal adult human kidneys from patients with ATTRV30M amyloidosis with or without anemia [138]. In 2013 Bussolati et al. identified a subset of renal CD133(+)/CD73(+) progenitor cells isolated from the human renal inner medulla, with a mesenchymal phenotype, as a possible source of EPO under hypoxic conditions, via the prolyl hydroxylase-HIF-2α axis [139]. CD133+ progenitors have been identified along the renal nephron [140] which overlaps with the described localization of EPOproducing cells in different segments of the human nephron by in situ hybridization studies. In addition, Nagai et al. demonstrated recently in mice that EPO mRNA expression occurs in proximal convoluted tubules (PCTs), distal convoluted tubules (DCTs) and cortical collecting ducts (CCDs) under normoxic conditions and in peritubular cells in severe hypoxia [141]. These dissimilarities may result from inter-species differences. Additionally, it is likely that EPO production by different populations of renal cells depend on the varying hypoxic conditions used in the different experimental models. Apart from the kidney, EPO production has been found also in the brain [142], retina, lung, spleen, bone marrow, in the male and female reproductive organs [143-144], placenta [145] and also in numerous cancer cells [124]. Although EPO is mainly produced by the kidney in the adult, an adequate renal EPOproducing cell line is not available. So, most of the present knowledge of the O2 sensing mechanism that controls EPO expression has been based on in vitro studies using the human hepatoma cell lines Hep3B and HepG2, described in 1987 by Goldberg et al. as a constitutive and inducible EPO producer, in an oxygen-dependent manner [146]. 4.3. Erythropoietin functions: hematopoiesis and cellular protection Recognition that the EPO receptor (EPOR) was expressed in several cells other than the erythroid progenitor cells led to the discovery of the extra-hematopoietic functions of EPO. EPO is a member of the cytokine type I superfamily [147], and is both an endocrine, paracrine and autocrine hormone. Introduction 26 5. The transthyretin protein Transthyretin (TTR) is a plasma protein involved in the transport of retinol, in a complex with retinol binding protein (RBP), and of the thyroxine (T4) hormone [190-191], hence its name: trans (transport) thy (thyroxine) and retin (RBP). Formerly it was named prealbumin because it migrates just slightly ahead (anodal) of albumin in serum protein electrophoresis. Liver and choroid plexus are the most abundant sites of TTR synthesis in humans. Liver produces circulating TTR that is secreted to the plasma where it reaches a concentration of approximately 200-250 mg/L (ranging from 3-8 µM) [192-194]. Choroid plexus, in turn, has the highest concentration of TTR mRNA in the body as TTR accounts for 12% of all proteins synthesized [195]. The protein produced by the choroid plexus is secreted into the cerebrospinal fluid (CSF), where it reaches a concentration ranging from 5 to 20 mg/L (0,09-0,4 µM) [193]. In CSF, 80% of T4 is bound to TTR whereas only 15% of T4 is bound to TTR in the serum [196-197]. TTR is also found in the eye, as it is produced by the pigment epithelium (ciliar and retinal) [198], and in less extent in alpha-cells of pancreatic islets [199], yolk sac [200], placenta [201], and intestine [202]. 5.1. Transthyretin structure TTR is encoded by a single copy gene with 4 exons located at chromosome 18. Exon 1 codes mainly for a signal peptide of 20 aminoacids that is cleaved before secretion of mature TTR [203-204]. Structurally, TTR is a tetrameric protein with 4 identical subunits, each with 127 aminoacids and a molecular mass of approximately 14 kDa [205]. Each monomer contains 8 β-sheet strands (A-H) and a short helix between strands E and F [206]. The 4 monomers associate non-covalently to form the tetrameric protein, which has a molecular mass of approximately 55 kDa. Association of two dimers is subjacent to the tetrameric structure. The strength of the interactions between monomers (resulting in a dimer) and between dimers (forming a tetramer) suggests that the dimer rather than the monomer or tetramer is the most stable unit of the TTR structure [207]. Introduction 27 Figure 5 – Structure of human transthyretin monomer (A), dimer (B) and tetramer (C). From Protein Data Bank, PDB ID:s 1F41, 2PAB The tetramer has 2 identical binding sites for T4 in a central channel and 4 surface binding sites for the complex RBP/vitamin A. However, only one binding site is occupied by T4 under physiological conditions due to negative co-operativity [208]. Also, only one RBP molecule can bind to TTR due to steric hindrance [209]. 5.2. TTR function TTR main function is the transport of T4 and of retinol-binding protein (RBP), which in turn transports vitamin A, both in the plasma and in the CSF. However, TTR is not likely to be essential for life or developmentbecause no abnormalities are found in mice that have had the transthyretin gene inactivated [210]. There are studies that suggest that the complex TTR-T4 is endocytosed and internalised. This phenomenon was observed in hepatomas, primary hepatocytes and renal cells and involves an endocytic multi-ligand receptor of the LDL receptor family, megalin (LRP2) [211-213]. TTR also has a protease activity [214]. A fraction of plasma TTR circulates bound to Apolipoprotein A-1 (ApoA1) and acts as a protease, as it is able to cleave the carboxylterminal domain, after a phenylalanine residue, of ApoA1 [214]. TTR can also cleave full length β-amyloid to a smaller fragment, and play a protective role in AD [215]. A B C Introduction 28 5.3. Models of amyloidogenesis Various mechanisms have been proposed to explain amyloidogenesis. Mutations in several proteins lead to structural changes that promote unfolding and predispose for the formation of fibers, resulting in hereditary amyloidosis syndromes. In the case of TTR, several mutations alter the thermodynamics and kinetics of dissociation of the tetramer and favor the formation of intermediates that self assemble into amyloid fibers [216]. For example, the V30M mutation confers a moderate instability to the tetramer, while L55P is the most unstable and pathogenic variant. However, the presence of mutations is not essential for amyloidogenesis to be triggered. Examples are some localized amyloidoses, and again using the example of TTR, senile systemic amyloidosis is caused by deposition of amyloid fibers from wild-type TTR. Many factors can trigger destabilization of the protein structure: heat shock, oxidative stress or chemical modifications, alterations of intracellular macromolecular crowding, presence of suitable surfaces, absence of stabilizing ligands, impairment of intracellular quality control of protein folding, pH changes and others [217]. Most in vitro studies regarding the formation of amyloid fibers are based on lowering the pH, suggesting that mildly acidic pH (present for example in lysosomes) can induce rearrangement of the tetramer structure and dissociation into partially denatured monomeric amyloidogenic intermediates, that are then joined again to form amyloid fibrils [23]. However, it has been shown that normal TTR, as well as the variants V30M, L55P and the non-amyloidogenic T119M dissociate into monomeric species at physiological pH and ionic strength [30] and that the most unstable variants (V30Mand L55P-TTR) exist in a complex equilibrium between monomers, tetramers and aggregates of higher molecular weight [30]. Nowadays, it is thought that the critical step for amyloidogenesis is the destabilization of the structure and the formation of non-native intermediate species that have the ability to self-associate. These intermediate species seem to play an important role both in the formation of amyloid fibrils as in the mechanisms of toxicity subjacent to amyloid disease. Depending on conditions, during the process of amyloidogenesis different types of species can be formed: amorphous aggregates, soluble oligomers or amyloid fibrils [218221], and the process rarely results in a homogeneous product. Usually, heterogeneous mixtures containing several species of aggregates (amyloid fibrils, amorphous aggregates Introduction 29 or soluble oligomers) are observed. Amorphous aggregates are formed faster than oligomers or fibrils and usually result from partially unfolded proteins that precipitate out of solution with no special conformational prerequisite to occur. Soluble oligomers are formed more slowly and remain in solution even after high-speed centrifugation, indicating that are not insoluble fibrillar or aggregated species. Fibrils are the slowest formed species and require special conditions to be formed. Figure 6 – A schematic representation of the protein self-association process, highlighting the three major products of the aggregation reaction: amorphous aggregates, soluble oligomers (spheroidal and annular) and amyloid fibrils. From Uversky 2010 [222]. The three main mechanisms proposed for TTR amyloidogenesis are: template-dependent, template-independent and proteolisis mechanisms [218]. Template-dependent The aggregation of the protein results from a nucleation-polymerization process where addition of the monomers is thermodynamically unfavorable until a nuclear-core is formed [223]. Rather, the polymerization stage is thermodynamically favorable. The critic nuclear core is the oligomer with the minimum size cable of starting the elongation process [224]. In the case of TTR there is no evidence that amyloidogenesis proceeds this way [225]. Introduction 30 Template-independent Aggregation is a process that begins with the monomeric protein and ends with the formation of aggregates. In between, a number of steps occur, including the formation of different oligomers [222]. The model accepted for TTR is based on a conformational change, in which altered monomers are the building blocks for amyloid fibril formation [224, 226]. Proteolysis hypothesis In some types of amyloidosis, the protein precursor is a subproduct of partial proteolysis (eg, reactive systemic amyloidosis (AA), immunoglobulin amyloidosis (AL), Alzheimer's disease (Aβ), gelsolin, cystatin C and apoA-I amyloidoses. In the case of TTR, C-terminal fragments are found, in addition to the full-length protein, in amyloid deposits of SSA and FAP patients [227]. However, these fragments are not always present, and their role is still controversial. Recently, Pires et al. described the existence of distinct annular oligomeric intermediates formed during both the assembly and disassembly pathways of TTR protofibrils induced at acidic pH [228]. They suggested that annular oligomers undergo morphological transitions into spheroid oligomers and protofibrils, which can be reversed to annular oligomers at physiological pH. In the TTR assembly pathway, aggregation was induced by acidification. Within the first hours, the sample was mostly populated by monomeric particles as a result of tetramer dissociation. After a few hours, annular oligomers with circular shape and octameric symmetry, as well as more compact spheroid oligomers were observed. The spacing of each subunit of the annular oligomers was consistent with the dimensions of a single wildtype TTR monomer. These annular oligomers may associate laterally and form spheroid oligomers and short protofibrils. After 7 days of incubation, only protofibrils and a small population of monomers/dimers remained. So, the annular oligomers seem to be a transient intermediate along the protofibrillogenesis pathway. The disassembly pathway occurred when the TTR protofibrils that were formed in acidic conditions were exposed to physiological buffer. After 15 minutes exposed to physiological pH, there was dissociation of the protofibrils into annular oligomers, which were different from those observed during TTR assembly but may still serve as assembly blocks for another form of amyloid aggregation. Introduction 31 Figure 7 – Model of TTR protofibril assembly and disassembly. From Pires et al. [228]. So, in the protofibrillogenesis model proposed by Pires et al. the native TTR undergoes structural transitions upon acidification, and the tetramers dissociate into amyloidogenic monomers and, in smaller quantities, dimmers, from which annular oligomers with octameric symmetry assemble. A single annular oligomer may serve as a scaffold for the continuous addition of subunits up to the formation of an annular doublet. Spheroid oligomers coalesce in a dynamic equilibrium with the growing protofibrils which are abundant in the first week of incubation. Upon adjusting of pH to neutral, protofibril structure quickly reorganizes, and dissociation into an annular oligomeric species quite distinct from that seen in the assembly pathway proceeds in a time scale of minutes. The annular oligomeric species are probably toxic, as it has been seen that oligomers containing up to eight TTR monomers, but not more, are cytotoxic [229]. Introduction 32 5.4. Methods to induce the formation of oligomeric species in vitro In vitro amyloidogenesis and formation of oligomeric species can be achieved by several methodologies, as long as the conditions that promote fibril formation are established. It is a difficult endeavour because variability exists between protein batchs and has been plagued by poor reproducibility of experiments both within and among laboratories [230]. Factors such as the purity of the protein sample, the ion-pairing agent used in the protein purification process, for example HCl and trifluoroacetic acid (TFA) buffers, the existence of pre-formed aggregates that can behave as fibril seeds and lead to accelerated amyloid formation, can contribute to this variability by affecting the kinetics of aggregation. To promote fibril formation, the sample is incubated under a variety of different solution conditions (pH, temperature, and ionic strength). The most common methods are: incubating the protein with stirring at room temperature for 7 days [231] using low pH treatment of purified protein: diluting the protein in sodium acetate buffer at pH 3.6 and incubating at 37ºC [228]; diluting in acetate buffer, KCl and EDTA, pH varying from 4.8 to 6.0 [229, 232-233], followed by incubation at 37°C for a maximum of 7 days [229] incubating the protein for 48 h at room temperature in trifluoroethanol and sodium acetate, pH 5.5 [234]. also, different concentrations of the protein and pH solutions, using either acetate or citrate buffer, may be used [235]. The choice of the method may depend on the protein involved or on the previous experience of the laboratory. 5.5. Cytotoxicity induced by oligomeric and pre-fibrilar TTR species Mature amyloid fibrils were for long considered the responsible species for the pathogenic features of amyloid diseases, as they were the material commonly found in the pathological deposits. However, in later years, evidence has accumulated that the soluble oligomers and pre-fibrillar assemblies grown from the amyloidogenic proteins are the main or even the sole cytotoxic species [229, 236-239]. Most of these evidences come from cell culture-based experiments, as these soluble species have not been well characterized in vivo. Other circunstancial evidences, such as the fact that clinical manifestations of amyloidosis-related neurodegenerative diseases often precede detectable accumulation of the fibrillar protein aggregates, lend support to this theory. So, soluble oligomers are Introduction 33 considered very important players both in protein aggregation and in the related cytotoxicity. The existence of early non-fibrillar TTR aggregates in pre-symptomatic FAP patients is supported by studies conducted by Sousa et al. [107]. The presence of these putatively cytotoxic aggregates is associated with the expression of markers of oxidative stress, and induction of apoptosis related molecules and pro-inflammatory cytokines [107], processes that may be common to the toxicity of prefibrillar amyloid protein aggregates in all amyloidotic diseases [234]. The mechanisms that mediate the cytotoxicity of oligomeric TTR aggregates are not yet fully understood. However, several pathways, including ER stress, oxidative stress and inflammation, have been implicated. 5.5.1. ER stress response in FAP A connection between the endoplasmic reticulum (ER) stress response and FAP was demonstrated by Teixeira et al. [240]. Increased levels of the ER-resident chaperone BiP, a member of the heat-shock protein 70 family, were found both in human biopsies of FAP patients as well as in TTR transgenic mouse model [240]. It was also shown that extracellular TTR oligomers can induce BiP expression and activation of eIF2α in cell culture, involving the mobilization of Ca2+ from the ER to the cytosol. An N-glycosylated TTR fraction was identified in plasma of carriers of the V30M mutation, which was undetectable in plasma of normal individuals. This glycosylated TTR was secreted, escaping ER-associated degradation (ERAD) [241]. 5.5.2. Disruption of biological membranes Amyloidogenic species could be toxic by their capacity to interact and permeabilize the biological membranes, through mechanisms such as detergent, carpeting effects or pore formation [242]. Studies in vitro suggest that the TTR-mediated calcium permeability may be driven by binding of the misfolded protein directly to lipids of the plasma membrane, and that toxicity is correlated with increased membrane binding affinity, destabilisation of cell membrane fluidity and subsequent decrease in cell viability [243]. 5.5.2. Apoptosis Activation of caspase-3 was described in sciatic nerves of asymptomatic FAP individuals in which non-fibrillar TTR aggregates were observed [107], as well as in cells treated with pre-fibrilar aggregates [107]. The death receptor Fas as well as caspase-8 were found to Introduction 34 be up-regulated in tissues from animal models for FAP with non-fibrillar TTR deposition, as well as in salivary glands from FAP patients [244]. The TTRV30M protein itself may induce apoptosis and autophagy concomitant with the accumulation of reactive oxygen species (ROS) [245]. Nunes et al. found that TTRV30M decreases endothelial survival by inducing apoptosis [246]. In this study it was also shown that TTR could regulate angiogenesis, as endothelial cells seem to acquire different molecular identities when exposed to either wild-type TTR or TTRV30M proteins. 5.5.3. Inflammation The pro-inflammatory cytokines TNF-α, IL-1β and M-CSF (macrophage colony-stimulating factor) were found to be up-regulated in nerves from FAP patients [106-107]. IL-1β, in particular, was investigated as a possible therapeutic target in an FAP mouse model. Treatment with Anakinra, an IL-1 antagonist, prevented TTR extracellular deposition in sciatic nerve, protecting unmyelinated nerve fibers from aggregate-induced degeneration [247]. Moreover, Anakinra administration significantly inhibited apoptosis and nitrative stress, highlighting the relevance of the IL-1 signaling pathway in the pathophysiology of FAP. On the other hand, up-regulation of the anti-inflammatory cytokine IL-10 in axons and Schwann cells of FAP nerves, and its correlation with the presence of TTR amyloid deposits was also observed, suggesting a balance between proand anti-inflammatory mechanisms in FAP [248]. 5.5.4. Oxidative stress Oxidative stress markers were found to be increased in axons of FAP patients even at pre-symptomatic stages of disease [107], as well as the levels of iNOS [244, 249] and of 3-NT (3-nitrotyrosine), a marker of protein nitration, in FAP nerves, suggesting that deposited TTR was subjected to nitration [106]. 5.5.5. The receptor for advanced glycation end products (RAGE) It has been postulated that TTR aggregates can bind to the receptor for advanced glycation end-products (RAGE), which is highly expressed at the surface of different cell types in amyloid deposition sites, and activate extracellular signal-regulated kinase (ERK) cascades, leading to downstream increased nuclear transcription factor kB (NF-kB) activity, and activation of caspases [250-251]. The inflammatory cascades are consequently activated by NF-kB and in turn will promote NF-kB activation [252-253]. Several evidences suggest that both AGE and RAGE may have a common role in the progression of TTR amyloidosis, as amyloid proteins may directly bind to RAGE and Introduction 35 activate signaling pathways that result in cellular perturbations [251, 254]. However, conflicting results exist regarding the fact that the cytotoxic effects exerted by the oligomeric TTR may be triggered by activation of NF-kappaB or apoptosis [254]. 5.6. EPO deficiency and TTRV30M aggregate toxicity: common mechanisms As previously stated, TTR aggregates can bind to RAGE, which is highly expressed at the surface of different cell types in amyloid deposition sites, and activate ERK cascades, leading to downstream increased NF-kB activity, and activation of caspases. These mediators, as well as others involved in oxidative stress and apoptosis may influence EPO production. So, downstream events precipitated by the deposition of cytotoxic oligomeric species may be involved in the inhibition of EPO expression. Hypoxia Inducible Factor (HIF), GATA-2 and NF-kB are the main known regulators of EPO gene expression. HIF induces EPO expression under hypoxic conditions, while GATA-2 and NF-kB transcription factors suppress EPO promoter activity when up-regulated or activated by the proinflammatory cytokines IL-1 and TNF-α [168]. MATERIALS AND METHODS Materials and Methods 45 1. Cell culture models and human renal biopsies In this work we used both immortalized and primary cell lines in experimental models. Immortalized cell lines: Hep3B (human hepatocellular carcinoma) SH-SY5Y (human neuroblastoma cell line) HEK293T (human embryonic kidney 293 cell line) Primary cell lines: RPC (human renal progenitor cells) CD133+CD24+ RPE (human retinal pigment epithelial cells) Hep3B cells were kindly offered by Dra. Sandra Alves (Department of Human Genetics, INSA, Portugal). SH-SY5Y and HEK293T were kindly offered by Prof. Paola Romagnani’s group (Excellence Centre for Research, Transfer and High Education (DENOTHE), University of Florence, Italy). RPC cells were isolated from the glomeruli of adult human kidney, as described by Sagrinati et al. [255], by the collaborators of Prof. Paola Romagnani. RPE cells were isolated from the choroid of an eye of an adult human cadaveric donor from Centro Hospitalar do Porto, in the setting of a project approved by the institutional ethics committee. The procedure for RPE isolation was adapted from Engelmann et al. [261] and Chung et al. [262]. Briefly, the RPE/choroid complex was placed in RPMI supplemented with 10% FBS, 10 µg/mL gentamicin, 0,5 mg/mL colagenase I and 0,5 mg/mL colagenase IV, was triturated with a bistoury and incubated for 4 hours at 37ºC. After incubation, the cells and media were collected and centrifuged for 5 minutes at 1400 rpm. The pelleted cells were washed with PBS, centrifuged and resuspended in RPMI supplemented with 10% fetal bovine serum (FBS) and 20 µg/mL gentamicin. The pigmented cells were counted using a neubauer chamber and plated at a density of 350.000 cells in a 25 cm2 culture flask. Materials and Methods 46 The RPC cells were maintained in EGM-MV (Cambrex Bio Science) supplemented with 10% FBS (Hyclone). All the other cell types were maintained in RPMI 1640 supplemented with GlutaMax (Invitrogen, CA, USA) and 10% FBS (Invitrogen). All cells were incubated at 37ºC in a 5% CO2 atmosphere. Human renal biopsies In this study, 14 human formalin-fixed and paraffin embedded renal biopsies from renal cadaveric donors and ATTRV30M amyloidosis patients were used. Biopsies were provided by the Pathology Department of Santo António Hospital and informed consent for research use was obtained. Three renal biopsies from cadaveric donors whose kidney was successfully transplanted were selected as controls. No donor clinical data was available due to the anonymity rules imposed in organ collection procedures. Between 1995 and 2001, every FAP patient considered for liver transplantation was submitted to renal biopsy. Eleven biopsies from FAP patients were randomly selected, taking into consideration as inclusion criteria the absence of anemia, scarse deposition of amyloid in the kidney and normal renal function. FAP patients were on average 38 ±7 years old (29 – 56), with 4.5 ± 2.1 years (2 – 12) of symptomatic disease, and serum Hb of 13.2 ± 1.6 (10 – 16.2) g/dl. 2. Expression and purification of recombinant human TTRV30M The in vitro studies conducted in this work evaluated the effect of TTR aggregates in the cellular models described in section 1 of this chapter. To do so, it was of crucial importance to obtain TTR preparations of very good quality, and free of any kind of contamination, such as endotoxin, to ensure that the effects observed with cells exposed to these preparations for several days were only due to their presence. To this end, we produced under stringent conditions recombinant human TTRV30M, using an E. coli expression system. As a control for some experiments we also used a sample of commercially available wild-type TTR (Sigma). Recombinant Human TTRV30M was produced using a strain of chloramphenicol (cam) resistant BL21-RIL E. coli transformed with the made-to-order synthetic vector pJexpress401:34985–TTRV30M_v2_opt (pJT, DNA2.0 Inc.). Transformants were Materials and Methods 47 selected for resistance to cam and kanamycin (kan), and the presence of the correct vector was confirmed. Expression and purification of the recombinant protein was optimised. Briefly, a colony of BL21-RIL/TTRV30M clone (pJT) was grown overnight in 3 mL of LB medium with 50 μg/mL kan and 16 μg/mL cam at 37ºC and 220 rpm. The pre-culture was inoculated in 600 mL TB medium with the same antibiotics and grown for 4 h at 37ºC and 250 rpm. The expression of the protein was induced with 1 mM Isopropyl β-D-1-thiogalactopyranoside (IPTG), for 4 h at 30ºC and 250 rpm. The cells were collected by centrifugation and the pellet was stored at -20ºC. To extract the protein, the pellet was dissolved in lysis buffer (50 mM Tris.HCl, pH 7.6, 200 mM NaCl, 10 mM imidazole) and treated with 100 μg/mL lysozyme, on ice for 30 min. The mixture was sonicated 3 x 30 s in pulse mode (0,5 s “on” and 0,5 s “off”) and centrifuged. The supernatant was filtered, first through a 0,45 µm membrane and then through a 0,2 µm membrane, and applied to a prepacked Ni2+ sepharose high performance column (HisTrap HP, GE Heathcare) equilibrated in lysis buffer, an immobilized metal ion affinity chromatography (IMAC). After washing the column with 7 volumes of lysis buffer, a linear gradient of 10 - 400 mM imidazole was applied to elute the His-tagged protein. The elution peak fractions were pooled and their content verified by SDS-PAGE with Coomassie Blue or silver staining. The TTR containing fractions were dialysed overnight at 4ºC against H2O, quantified by absorbance spectroscopy at 280 nm (Thermo Scientific NanoDropTM 1000 Spectrophotometer), using the extinction coefficient of wild-type TTR (wt-TTR), ε280 = 7.76 x 104 M-1 [263], and stored at -20ºC. To remove the histidine tag, the chromatography product was cleaved with recombinant TEV protease, produced in our laboratory using the S219V TEV mutant, encoded in the pRK793 plasmid. This vector overproduces the catalytic domain of TEV protease in the form of an MBP fusion protein that cleaves itself in vivo to yield a TEV protease catalytic domain with an N-terminal His-tag and a C-terminal polyarginine tag, and was a kind gift from David Waugh (Addgene plasmid # 8827) [264]. It was expressed in BL21(DE3)-RIL cells, and TEV S219V purified by IMAC with a Ni2+ sepharose high performance column, as described for TTRV30M. The cleavage reaction was performed at room temperature for at least 4 h in 50 mM Tris.HCl pH 7.6, 0.5 mM EDTA, 1 mM DTT, using a ratio of 1 mg of TEV for 10 mg of TTRV30M. Following digestion, the reaction was dialysed against 50 mM Tris.HCl pH 7.6, 200 mM NaCl and TEV protease was removed via its own poly histidine tag by IMAC. These TTRV30M fractions were pooled, concentrated and buffer exchanged with 50 mM Tris.HCl pH 7.6, by ultrafiltration, using a centrifugal filter device with a molecular weight cut off of 10 kDa. Materials and Methods 48 An ion exchange chromatography was performed to further purify the recombinant TTRV30M. The protein was applied to a Q-Sepharose FF column equilibrated in 50 mM Tris.HCl pH 7.6. A linear gradient of 0 - 500 mM NaCl was applied to elute the TTR. Finally, the TTR containing fractions were concentrated by ultrafiltration and applied to a gel filtration Sephacryl S-200 column equilibrated with 10 mM phosphate buffer pH 7, 100 mM KCl and 1 mM EDTA. The appropriate fractions were pooled and concentrated by ultrafiltration. Purified soluble TTRV30M was detoxified using ActiClean Etox (Sterogene). The protein was eluted with PBS (Invitrogen), concentrated to a final concentration of 2 mg/mL and filter sterilized by passing through a 2 µm seringe filter. Final concentration was determined by measuring the absorvance with a NanoDrop 1000 spectrophotometer, using the extinction coefficient of wt-TTR, ε280 = 7.76 x 104 M-1. A limulus amebocyte lysate-based assay (E-Toxate Test, Sigma, MO, USA) was used to confirm that the protein was endotoxin-free. 3. Preparation and characterization of TTR amyloidogenic aggregates In the literature, several methods have been described for the production of oligomeric aggregates. Fibrillogenesis depends on various conditions such as temperature, protein concentration, ionic strength, agitation and pH [265]. The misfolding of the protein is an essential step. In the case of TTR, the most common method is based on the acidification of the protein solution to mildly acidic pH (~4-5) to mimic the lysosomal milieu, as it may be sufficient to initiate the assembly of TTR amyloid fibrils [266]. Complete unfolding of the protein into monomers with HCl pH 2.0 followed by refoldig with NaCl has also been described [267]. Another method of achieving oligomerization is by aging the protein, with or without stirring, at physiological pH [268]. In this work, three different conditions were tested to induce aggregation: moderate pH, unfolding and refolding (HCl and NaCl) and aging. Regardless the method used, we found that the fibrillation process is very complex and not always reproducible, whatever the method used. 3.1. TTR aggregation at mild pH (4.0-5.5) Aggregation was induced by lowering pH to 4.0 [269], using a sodium acetate solution, or to pH 5.5 using trifluoroethanol and sodium acetate, methods adapted from Reixach et al. [229] and Bucciantini et al. [238], respectively. Different aggregation times were tested. Materials and Methods 49 Briefly, 1,5 mg/mL TTRV30M was incubated with either 100 mM sodium acetate buffer, pH 4.0, as with a mixture of 10% trifluoroethanol and 50 mM sodium acetate pH 5.5, at room temperature and at 37ºC. At different time points, the amyloid formation process was followed by spectrofluorimetry with Thioflavin T (ThT). 3.2. TTR aggregation by unfolding with HCl and refolding with NaCl A 30 µM TTRV30M protein solution was dialysed for 96 h against 10 mM HCl (pH 2.0) to unfold the tetramer into monomers and the aggregation was induced with NaCl, as described by Lindgren et al. [267]. Final concentrations of 100 mM and 50 mM of NaCl, well as different incubation times, were tested. At different time points, the formation of amyloid was evaluated by spectrofluorimetry with ThT and cross-linking assays. 3.3. TTR aggregation at physiological pH followed by magnetic stirring A solution of 36 µM TTR V30M (2 mg/mL) in PBS pH 7.4 (Invitrogen) was filtered through 0,2 µm Anotop syringe filters (Whatman, Kent, UK) and incubated at 37ºC for 72 hours, followed by vigorous stirring for 5 minutes, using a magnetic stirring bar. The size of the molecular species in solution was evaluated by Dynamic Light Scattering (DLS). Oligomers from wild-type TTR (wt-TTR) were prepared by incubating a commercial protein (Sigma-Aldrich) in PBS, pH 7.4 for 96 hours at 37ºC, followed by vigorous stirring for 25 minutes. 4. Characterization of TTR amyloidogenic aggregates The formation of TTR oligomeric aggregates was evaluated by the following methods: 4.1. Thioflavin T assays ThT assay was adapted from the method described by Nilsson [230]. Briefly, a 12,5 mM ThT was prepared in 50 mM Tris pH 7.8 and filter sterilized. Aliquots from the TTR aggregation reaction at different time points were mixed at a final concentration of 0,02 mg/mL with the assay buffer at 25 µM ThT. The fluorescence intensity for each sample was measured by excitation at 440 nm and emission 482 nm, using a Spectramax Gemini XS Reader. An aliquot of untreated protein solution and of Tris buffer were also added to the ThT assay buffer and intensity measured, to serve as control sample and negative control, respectively. A measured intensity above the control sample is indicative of the presence of amyloid fibrils. Materials and Methods 50 4.2. Chemical cross-linking assays Chemical cross-linking, followed by SDS-PAGE and silver staining, was performed to visualize the formation of TTR oligomers of different size according to the aggregation time [229]. Briefly, aliquots from the TTR aggregation reaction at different time points were mixed at a final concentration of 0,4 mg/mL with a 2.5% solution of glutaraldehyde, and incubated for 5 min at 37ºC. The reaction was terminated by addition of 1 M Tris-HCl, pH 8.0 to a final concentration of 100 mM. The samples were evaluated by SDS-PAGE and silver staining. 4.3. Dynamic light scattering (DLS) DLS measurements were performed at 25ºC in a Malvern Zetasizer Nano ZS (Malvern, Worcestershire, UK) [268] to characterize tetrameric (time zero) and oligomeric TTR V30M preparations. Each sample was measured three times; average distributions are presented. 5. Cell toxicity, apoptosis and caspases 3/7 assays It has been postulated that oligomeric species, rather than mature fibrils, induce citotoxicity in vitro. Mainly neuronal cells (neuroblastoma) [106, 229, 250, 270-272], and primary human umbilical vein endothelial cells (HUVECs) [246] have been used in these cytotoxicity studies. In this work we used both immortalized and primary cell lines to evaluate the cytotoxic effect of the oligomeric TTR aggregates in these cells in terms of cell viability, apoptosis and activity of caspases 3/7. Three distinct aggregation methods were tested in order to establish the best conditions. 5.1. MTT and MTS cell viability assays Cells were plated into 96-well plates in complete cell medium (EGM-MV supplemented with 10% FBS for RPC or RPMI 1640 supplemented with GlutaMax and 10% FBS for all the other cell types) at a density of 3 x 103 cells per well and incubated overnight at 37ºC and 5% CO2. The next day the medium was removed and 100 μL of fresh medium with 0.5% FBS and 2 μM of each protein preparation added. The non-radioactive cell proliferation assays MTT (Promega, Wisconsin, USA) or MTS (Promega, Wisconsin, USA) were performed after 24, 48 and 72h of incubation, according to the manufacturer's instructions. The absorvance was read in a microplate reader at 580 nm for the MTT assay and 490 nm for the MTS. The result was expressed as “percentage of living cells” Materials and Methods 51 relative to that seen in control using the expression 100 × (ODsample - ODblank)/(ODcontrol - ODblank). The ODblank was established from the average of the wells containing only medium. Average values and SD were calculated from triplicate determinations. 5.2. Annexin V apoptosis assays Renal progenitor cells (RPC) were plated at a density of 100.000 cells/well in a 6-well plate in EGM-MV with 20% FBS and incubated overnight at 37ºC and 5% CO2. The cell medium was changed to EBM without serum and the cells incubated for 7 h. Protein preparations (2 μM) diluted in EBM with 0.5% FBS were added to each well and the cells incubated for 48 and 72 h. The cells were detached, washed with PBS and resuspended in FACS Buffer (10 mM HEPES, 140 mM NaCl, 2 mM CaCl2, pH 7.4) at 1 x 106 cells/mL. Cells were stained with Annexin V-APC (eBioscience BMS306APC/100) and propidium iodide (PI) (Invitrogen) at a final concentration of 5 µg/mL and incubated for 15 min at room temperature, then kept on ice. Cell were analysed by flow cytometry in a BD LSRII flow cytometer (BD Biosciences). 5.3. Caspases 3/7 assays SH-SY5Y, Hep3B and RPE cells were plated as for the MTS assay. 2 μM of each protein preparation, diluted in RPMI 1640 with 0,5% FBS was added to the respective wells. After 24 h of incubation at 37°C, the Caspase-Glo 3/7 Assay (Promega, Wisconsin, USA) was performed to determine the activity of caspases 3/7, according to the manufacturer's instructions. Luminescence was read in a Vitor3 spectrophotometer (Perkin Elmer) and the activity of caspases 3/7 was expressed in luminescence units. Average values and SD were calculated from triplicate determinations. 6. Influence of TTR oligomeric aggregates in the cell cycle and differentiation capacity of renal progenitor cells 6.1. Cell Cycle analysis RPC were plated at 1 x 105 cells/well in a 6-well plate the day before the experiment. Solutions of 2 μM TTRV30M diluted in EBM with 0,5% FBS were added to each well. The cells incubated for 48 and 72 h with the stimuli, then were detached, washed and resuspended in 100 µL 50% FBS in PBS, then fixed at 4ºC for 1 hour with 300 µL cold 70% ethanol added dropwise, washed, stained with Propidium Iodide (50 µg/mL) in Results and Discussion 59 1. Production and evaluation of TTRV30M aggregates Recombinant TTRV30M protein was obtained using an E. coli expression system. The synthetic vector pJexpress401:34985–TTRV30M_v2_opt (pJT, DNA2.0 Inc.) codes for a fusion protein containing the human TTR with the Met30 mutation and a poly-histidine tag that enables its purification by IMAC, and that can be subsequently removed. The recombinant protein accumulates inside the cell, in inclusion bodies, a process that was minimized by decreasing the incubation temperature to 30ºC at the time of protein expression induction with IPTG. After lysing the cells, the protein was released to the medium in soluble form and the supernatant was directly applied to the Ni2+ sepharose column. A linear gradient of imidazole was run through the column, and the TTRV30M protein was eluted with a concentration of imidazole of approximately 300 mM, in an almost pure form (figure 10). On average we obtained 40 mg of TTRV30M-HIS tagged / L of cell culture, which was equivalent to approximately 2 mg of TTRV30M-HIS tagged per gram of bacterial pellet. Figure 10 – IMAC elution profile of the rh-TTRV30M; lane 1: MW ladder; lanes 2-5: flowthrough; lanes 6-15: linear gradient of imidazole; lanes 10-14: fractions containing TTRV30M. After TEV protease treatment, the histidine tags and protease were removed in the end by passing the reaction mixture through a second Ni2+ sepharose column, to obtain a relatively pure cleaved TTRV30M (figure 11). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Results and Discussion 60 Figure 11 – Proteolysis of his-tagged TTRV30M by TEV protease (A) and cleaved TTRV30M after IMAC purification and concentration by ultrafiltration (B); lanes 1 and 9: MW ladder; lane 2: TEV793; lanes 3 and 6: his-tagged TTRV30M; lanes 4 and 5: completed proteolysis reaction, before IMAC purification; lane 7: peptides and TEV that remained attached to the column, eluted with 400 mM imidazole; lanes 8 and 10: purified TTRV30M; lane 11: TTRV30M purified and concentrated by ultrafiltration. To further increase the purity of the protein, two more chromatographic steps were performed: an ion exchange chromatography, using Q-Sepharose FF, followed by a polishing gel filtration chromatography, to eliminate possible protein aggregates that may be formed during the purification process. The last step was the chromatographic removal of LPS. A limulus amebocyte lysate-based assay (E-Toxate Test, Sigma, MO, USA) was used to confirm that the final protein batch was endotoxin-free. Due to the extensive purification process, involving two IMAC chromatographies, one ion exchange chromatography, a gel filtration chromatography and finally detoxification, the yield of purified TTRV30M was approximately 45% of the initial TTRV30M-HIS tagged pool. We obtained an average of 18 mg TTRV30M / L of cell culture. Although the yield is relatively low, we obtained a highly pure protein to be used on the cell culture assays. A B 1 2 3 4 5 6 7 8 9 10 11 Results and Discussion 61 2. Evaluation of TTRV30M amyloidogenic aggregates Throughout the work, three different methods were used to produce the TTRV30M aggregates, according to the literature: acidification of the protein solution to mildly acidic pH (~ 4-5) [266], complete unfolding of the protein into monomers with HCl pH 2.0 followed by refoldig with NaCl [267], and aging the protein at 37ºC in physiological pH, followed by 5 minutes magnetic stirring, a method that we adapted from the procedure described by Ferreira et al. [268]. The reason for testing different methods was the poor reproducibilitty of the oligomerization process. Regardless the method used, we found that the fibrillation process to be very complex and difficult to reproduce. Fibrillation of TTR seems to proceed by disruption of the tetramer into monomers, which undergo a conformational change, and are the building blocks for the formation of fibrils [224, 275-278]. Other models implicate the dimer [279] or the oligomers [280] as the building blocks. Irrespective of the model proposed, there are common evidences regarding which species are responsible for the cytotoxicity induced in the tissues. The intermediate species or small soluble oligomeric aggregates that are formed during the fibrillation process are toxic to cells in a free radical dependent manner [271], while the amyloid fibrils themselves are not toxic to the cells. Amyloid fibrils have been proposed to be protective, by storing the misfolded proteins in non-toxic fibrils and thereby removing toxic protofilaments from circulation [271]. To evaluate the formation of amyloidogenic TTRV30M aggregates, Thioflavin T (ThT), chemical crosslinking or dynamic light scattering (DLS) were used. Upon addition of thioflavine T to suspensions of TTRV30M incubated with either 100 mM Acetate Buffer, pH 4.0, with a mixture of 10% trifluoroethanol (TFE) and 50 mM sodium acetate pH 5.5, or with HCl pH 2.0 and 100 mM NaCl, the fluorescence intensity increases (figure 12), indicating formation of amyloidogenic species. Results and Discussion 62 In the case of acetate buffer at pH 4, the intensity of fluorescence increased slowly over time (figure 12A). On the other hand, for the TTRV30M incubated with either a mixture of TFE and sodium acetate pH 5.5 (figure 12B) or with HCl and NaCl (figure 12C), the aggregation reaction started immediately. Figure 13 shows dissociation of the TTRV30M tetramer into monomers as a function of pH and the rapid formation of larger species upon addiction of NaCl. These methods, based on acidification of the protein solution to induce amyloid formation are described in the literature and used to produce cytotoxic aggregates. We tested these aggregated TTRV30M preparations on cell culture assays. Despite positive ThT results, we had no significant cytotoxic effect with the solutions prepared with acetate buffer pH 4.0 or with acetate buffer pH 5.5 and TFE. On the other hand, we obtained a moderate cytotoxic effect with the protein solution prepared by acidification Figure 12 – Time course of TTRV30M aggregation followed by ThT fluorescence, induced by acetate buffer pH 4.0 (A), acetate buffer pH 5.5 and 10% TFE (B) or HCl pH 2.0 followed by 100 mM NaCl (C). Figure 13 – Cross-linking with glutaraldehyde followed by SDS-PAGE with silver staining showing TTRV30M tetramer dissociation with HCl at pH 6 (2), 5 (3), 4 (4), 3 (5), 2 (6) and formation of aggregates at pH 2 and incubation for 5 minutes with 100 mM NaCl (7). 1 2 3 4 5 6 7 kDa 75 31 19 17 Results and Discussion 63 with HCl and induction of aggregation with NaCl for 5 minutes. Interestingly, the solutions prepared with the acetate buffers and TFE usually precipitated immediately, indicating formation of large amorphous aggregates. On the other hand, the solution of TTRV30M in HCl was completely transparent, even after the addiction of NaCl, probably due to the formation of soluble oligomers. Although the ThT assays done did indicate the presence of amyloid-like species, no correlation with the cytotoxicity induced by the protein solutions was apparent. As the aim of our work was to use the TTRV30M oligomeric preparations on cell culture assays, and to more accurately approximate the physiological environment, we also tested the method based on aging the protein solution while maintaining the physiological pH, as described by Ferreira et al [268]. We made a slight change, by aging the TTRV30M solution at 37ºC under stagnant conditions and then stirring vigorously for only 5 minutes. DLS was used to evaluate the size of the TTRV30M species in solution. Soluble tetrameric TTRV30M, diluted in PBS at pH 7.4, at time 0 showed a single peak at approximately 7 nm (figure 14A). To induce the formation of oligomers, TTRV30M was incubated for 72 hours at 37ºC, followed by 5 minutes of vigorous stirring. At the end of the 72 h incubation, the presence of oligomeric aggregates with 150 nm was noted (figure 14B), that progressed very rapidly to larger 300 nm aggregates after vigorous stirring, representing then the majority of the TTR preparation (figure 14C). A preparation stirred for 7 days showed a predominance of species of approximately 600 nm (figure 14D). However, this solution did not induce cytotoxicity, unlike the solutions containing species of 150 and 300 nm. Results and Discussion 64 Figure 14 - DLS analysis of TTR V30M aggregation for 0 (A), 3 days at 37º C under stagnant conditions (B), 3 days at 37º C under stagnant conditions followed by 5 minutes of magnetic stirring (C), 7 days of magnetic stirring (D). Preparation represented in (C) has a high percentage of small oligomeric species (78%) so was the one used for the cell culture assays. TTRV30M is a tetrameric protein with an extensive β-sheet conformation, prone to dissociation into intermediate molecular species during amyloid fibrillogenesis. Pires et al. reported that distinct annular oligomeric intermediates formed along both the assembly and disassembly pathways of TTR, that undergo morphological transitions into spheroid oligomers and protofibrils [228]. We induced protein fibrillation in vitro by aging a solution of TTRV30M at physiological pH and temperature, which is evidence for the natural Results and Discussion 65 instability of the quaternary structure of the protein and its tendency to aggregate and form amyloidogenic fibers. We saw that after 72 hours of incubation under stagnant conditions, species of molecular size corresponding to early oligomers are already in solution, albeit in small amounts relative to the predominant tetrameric form. Stirring this solution for 5 minutes was sufficient to trigger the rapid formation of larger species. TTRV30M preparations obtained by aging at physiological pH followed by magnetic stirring for 5 minutes (figure 14C) and preparations obtained by dissociation of the tetramer into monomers at pH 2.0 followed by refolding with NaCl (figure 12C) were the final preparations of oligomeric TTRV30M used for the cell culture assays, as we saw that these were the most cytotoxic species. Results and Discussion 66 3. TTRV30M oligomeric aggregates compromise cell viability of both immortalized SH-SY5Y, Hep3B and HEK293T cell lines, as well as of primary RPE and RPC cells The immortalized cell lines Hep3B, SH-SY5Y and HEK293T, the primary RPE cells and the renal progenitor cells RPC were exposed to TTR preparations in order to evaluate cell viability. Oligomeric preparations were prepared by unfolding with HCl and refolding with NaCl, or by aging the protein solution at pH 7.4 and 37ºC followed by 5 minutes of stirring. Hep3B and SH-SY5Y were exposed to oligomers produced by both methods. HEK293T and RPC were exposed only to oligomers produced by unfolding and refolding. RPE cells were exposed only to oligomers produced by aging the TTRV30M at physiological pH and temperature. 3.1. Cell viability assays A non-radioactive cell proliferation assay (MTT or MTS) was used to evaluate the cytotoxicity of TTRV30M preparations. The results obtained for each cell type were similar regardless of the method used, so we calculated the average of all the experiments and respective replicates in which these oligomers were cytotoxic, for each cell type. Preliminary experiments were performed with RPC cells exposed to protein concentrations of 0.5, 1.5 and 3 µM and with an incubation time of 72 hours. A concentration dependent reduction in cell viability was observed (figure 15). Results and Discussion 67 Figure 15 - Non-radioactive Cell Proliferation Assay in the human RPC exposed for 72 hours to concentrations of 0.5, 1.5 and 3 µM of oligomeric TTRV30M. The results are represented as percentage of living cells for each condition relatively to control (cell medium). Values are means of triplicates. Error bar= SD. In order to reach a compromise between the effect of the protein preparations on the cells and the quantity of protein spent on each experiment, a concentration of 2 µM was chosen for all the subsequent experiments throughout the work. A modest but statistically significant reduction in cell viability was observed after 24, 48 and 72 hours in the presence of oligomeric TTRV30M when compared to the control (cell media alone), in the five cell types evaluated (figures 16-18 and tables 3-5). Detailed information about the percentages of living cells relatively to controls, and respective standard deviations and student’s t-test values are presented in tables 3-5. Comparing cell viability between the oligomeric and tetrameric TTR preparations, the 24 hours incubation time gave more consistent results, showing a statistically significant reduction of cell viability for all cell types in the presence of the oligomeric TTRV30M preparations. Despite being a modest reduction (varying from 7% for RPC to 13% for Hep3B and RPE), oligomeric TTR aggregates reduced cell viability consistently and to a similar amount independently of the cell type, after 24 hours of incubation. Less consistent results were obtained for longer incubation times. Results and Discussion 74 The caspase assay is rather more sensitive than the MTS assay, therefore it is not surprising that some activity is detected after 24 hours of incubation at 37ºC with tetrameric TTRV30M in the caspase assay, but not with MTS. As already mentioned for the viability assays, tetrameric protein solution may begin to form monomers or oligomers that even in small quantities could be sufficient to activate caspases. Our assumption is supported by the fact that DLS of TTRV30M after 72 hours incubation at 37ºC showed that this protein has a tendency form oligomeric species, and was able to induce a reduction of cell viability (figures 14B and 19). Early oligomeric aggregates have been shown to be more cytotoxic than mature amyloid fibrils [229, 270-272]. Similar reduction levels of cell viability induced by the oligomeric TTRV30M preparations after 24 hours of incubation were observed for the five different cell types evaluated: human hepatoma cell line Hep3B, human neuroblastoma cell line SH-SY5Y, human embrionic kidney HEK293T, human renal progenitor cells (RPC) and primary retinal pigment epithelial cells (RPE). For three of these cell types: Hep3B, SHSY5Y and RPE, a caspase assay was performed. Together with the reduction of cell viability, a concomitant increase in caspase 3/7 activity was seen in the cells exposed to oligomeric TTRV30M aggregates. These results agree with previous studies performed using other cell types, mainly neuronal cells (neuroblastoma) [106, 229, 250, 270-272], and primary human umbilical vein endothelial cells (HUVECs) [246], and implicate apoptosis in loss of viability. Interestingly, this cytotoxic effect induced by aggregates seems to be independent of the cell type, contrary to what was reported by Cecchi et al. Using aggregates of HypF-N, a prokaryotic peptide domain, that display the same properties of aggregates produced by disease-associated peptides and proteins, they showed that different cell lines have different susceptibilities to damage and apoptosis, and this susceptibility was inversely related to membrane content in cholesterol, to total anti-oxidant capacity of the cell and to the ability of the cell to maintain the balance of intracellular free Ca2+ [282]. However, the same group reported later that such differences are more pronounced in longer exposure times due to different recovering capacity from cellular damage [281]. In fact we saw a similar cytotoxic effect between the various cell types in the shortest exposure time to the aggregates. Increasing the incubation time, the results became less reproducible, an increased toxic effect being observed in some cases but a reduction of cytotoxicity in most of them. This may be due to continuing fibrillogenesis, leading to the formation of larger species which may no longer be toxic. According to Cecchi et al, the reduction of cytotoxicty at longer incubation times may also be due to recovery from cellular damage. Results and Discussion 75 In vivo, it is known that TTR aggregates preferentially deposit in perypheral nerves, causing myelin destruction, and also in the heart, kidney and vitreous rather than in the liver. However, the mechanisms involved in this preference are not yet understood. In view of our citotoxicity results we can speculate that, at least in early disease stages, TTR aggregates may influence cell viability regardless of cell type. Oligomers generated from wild-type TTR were tested on Hep3B cells and also induced a reduction of cell viability. This result is in agreement with our expectations, as wild-type TTR is also amyloidogenic and responsible for senile systemic amyloidosis (SSA), causing mainly cardiac complications. With these results we can say that both V30M and wild-type TTR oligomeric preparations were cytotoxic, a property that may be common to all prefibrillar amyloid protein aggregates and apparently independent of the target cell type. Results and Discussion 76 4. Renal progenitor cell proliferation is inhibited by TTRV30M oligomeric aggregates but maintain their capacity to differentiate into podocytes in vitro A subset of renal progenitor cells (RPC), expressing the surface markers CD133 and CD24, with self-renewal and multidifferentiation potential exist at the urinary pole of Bowman’s capsule in the adult human kidney [255]. In Familial Amyloidotic Polyneuropathy, amyloid deposition of mutant TTRV30M leads to renal complications, including nephrotic syndrome and end-stage kidney failure [283]. Here we assessed whether RPC are vulnerable, in vitro, to TTRV30M oligomers in order to understand if their regenerative potential could be compromised. We showed (section 3.1.) that oligomeric TTRV30M aggregates reduce RPC viability by 9±4% after 24 hours. This reduction was significant relatively to both the unexposed and tetrameric TTRV30M exposed controls. Apoptosis, necrosis and alterations in cell cycle progression were assessed by FACS analysis on cells treated for 48 and 72 hours. TTRV30M oligomers did not induce apoptosis, necrosis or alterations in cell cycle progression to the RPC cells in a significant extent, neither at 48 or 72 h (figure 21). Figure 21 - FACS analysis to evaluate apoptosis and necrosis (A) and alterations in cell cycle (B) in RPC treated with 2 µM oligomeric TTRV30M for 48 and 72 h. A B Results and Discussion 77 CD133+CD24+ RPC have an inherent capacity to differentiate into podocytes when cultivated in VRAD medium. The influence of oligomeric TTRV30M on the regenerative capacity of these progenitor cells was assessed. Podocytes express nephrin so the relative expression level of this marker was evaluated by real-time pcr to confirm differentiation. The results varied between experiments, but on average we obtained similar results between the controls and the cells exposed to TTRV30M aggregates, both on differentiation and non-differentiation media (Figure 22). Figure 22 - Differentiation of renal progenitor cells into podocytes. Assessment by quantitative RT-PCR of fold increase mRNA levels for the podocyte marker nephrin and the housekeeping gene GAPDH. TTRV30M aggregates reduced proliferation of renal progenitor cells. However, in the apoptosis and cell cycle progression assays no significant alterations were observed. Also, in the differentiation experiments we saw no alterations in the expression level of the nephrin gene, a marker for podocytes, relatively to controls, so the inherent capacity of these progenitor cells to differentiate into podocytes was apparently not affected by the oligomers. These results are in agreement with a study of Petrakis et al. They saw, with a transgenic mouse model of TTRV30M-related amyloidosis, that TTRV30M deposition has deleterious effects on glomerular basement membrane thickness and podocyte foot process width, but without affecting nephrin and podocin gene expression [284]. A study from Neri et al, using murine embryonic stem (ES) cells and haematopoietic progenitor (HP) cells exposed to oligomers of Aβ42 showed that ES, but not HP, displayed some impaired viability but their differentiation was not affected by these oligomers [285], which is in agreement with our results. Results and Discussion 78 From this study we can say that TTRV30M oligomers inhibit RPC proliferation but do not influence their capacity to differentiate into functionally mature podocytes, and thus should not compromise tissue regeneration. These resident progenitor cells can also induce regeneration of tubular structures of different portions of the nephron, which can be critical for preventing irreversible renal failure, and could be useful in cell therapies, particularly in FAP. We have previously shown that EPO expression occurs in distal tubular cells, podocytes and cells of the cortical collecting ducts, in normoxic FAP patients [138]. Although there is no data regarding the ability of RPC to differentiate into EPO-producing cells, they can differentiate into podocytes and tubular cells, which could be important to maintain EPO production. Results and Discussion 79 5. Oligomeric TTR V30M aggregates reduce erythropoietin mRNA expression The principal aim of this study was to evaluate the effects of oligomeric TTRV30M aggregates on EPO gene expression, in a cell culture model of EPO production. Due to the lack of an adequate renal EPO-producing cell line, we used the human hepatoma Hep3B cell line, described in 1987 by Goldberg et al. as a constitutive and inducible EPO producer, in an oxygen-dependent manner [273]. In a second phase, retinal pigment epithelial cells (RPE) were also evaluated for endogenous EPO gene expression. 5.1. Oligomeric TTR V30M aggregates reduce EPO expression in Hep3B cells EPO mRNA expression in Hep3B cells exposed to 2 µM TTRV30M was evaluated by realtime PCR. Threshold cycle (Ct) values obtained for EPO in normoxia were 32.68±0.08 for control media, 32.37±0.28 for tetrameric TTRV30M and 33.30±0.06 for oligomeric TTRV30M. In mimicking hypoxia using DMOG, a cell permeable panhydroxylase inhibitor, Ct values for EPO were significantly lower: 28.81±0.10 for control media, 28.58±0.40 for tetrameric TTRV30M and 28.37±0.13 for oligomeric TTRV30M. The housekeeping gene TBP was used as an internal control. Ct values for TBP did not differ significantly between samples (25.16±0.19). Relative quantification of EPO mRNA levels was calculated taking as reference a standard curve for EPO and the results were normalized to TBP, using a ΔCT model [286]. EPO expression after 24 hours was reduced by 50.3±2.8% (p=0,0092) in normoxic Hep3B cells treated with oligomeric TTRV30M when compared to the tetrameric form of the protein (figure 23A). With DMOG, EPO gene expression levels increased 13 to 22 times. In these simulated hypoxia experiments, cells did not show a significant difference in EPO expression levels using either tetrameric TTRV30M or oligomeric TTRV30M aggregates (figure 23B). Hep3B cells were also exposed to 2 µM tetrameric and oligomeric wild-type TTR for 24 hours in normoxia. EPO expression was reduced by 22.0±9.1% (p=0,010) in Hep3B cells treated with oligomeric wild-type TTR when compared to the tetrameric form. Results and Discussion 80 Figure 23 - Relative expression of erythropoietin in Hep3B cells exposed to TTR V30M for 24 hours. (A) Relative expression levels of EPO, normalized for TBP, in normoxia. (B) Relative expression levels of EPO both in normoxia and simulated hypoxia (induced by DMOG). p value was calculated relatively to tetrameric TTR V30M experiments, using student’s t-test. *p<0,05. Values are means of duplicates. Error Bar= SD. The EPO gene has a hypoxia responsive element at the EPO 3’ enhancer site to which hypoxia inducible factor (HIF) can bind [287-289]. In normoxia, the α-chain of HIF is hydroxylated and degraded by the proteasome. In hypoxia the prolyl hydroxylases (PHD) and aspaginyl hydroxylase factor inhibiting HIF (FIH) are inhibited, the α-chain of HIF is not degraded and binds to the β-chain, which is constitutively expressed, and HIF accumulates, inducing gene expression [290-291]. To mimic hypoxia, we used the cell permeable panhydroxylase inhibitor dimethyloxalylglycine (DMOG). We found that exposure of normoxic Hep3B cells to oligomeric TTR V30M significantly reduces EPO mRNA expression. However, Hep3B cells treated with DMOG did not show significant differences in EPO mRNA levels when exposed to tetrameric or oligomeric TTR V30M. Under these simulated hypoxia conditions, Hep3B cells up-regulated EPO expression 13 to 20 times. It is possible that this large increase in EPO expression masks any effect of exposure to TTR aggregates on the gene expression level, or, alternatively, different regulatory mechanisms of EPO production could be involved in normoxia and hypoxia, altering the cellular response to the presence of oligomeric aggregates. Our previous work using patient kidney biopsies evaluated EPO expression in normoxia, and suggested that differences in the regulation of this gene in normoxia and hypoxia are Results and Discussion 81 possible [138]. Recently, Nagai et al. explored this subject and demonstrated that, in mice, EPO mRNA expression occurs in renal tubular cells in normoxia, as we have shown for the human ATTRV30M amyloidosis kidney, while in hypoxia it occurs largely in peritubular cells, strengthening the case for different regulatory mechanisms for EPO expression in normoxia and hypoxia [141]. Anemia in FAP patients sometimes precedes overt clinical disease and persists even after liver transplantation. There is evidence that after liver transplantation, the wild-type TTR can still form amyloid deposits and that there is progression and maybe even acceleration of amyloid deposition [59, 292]. In order to understand the persistence of low EPO production after liver transplantation, we produced oligomeric aggregates from normal TTR. We saw that Hep3B cells exposed to oligomeric wild-type TTR aggregates are also capable of reducing EPO mRNA expression when compared to tetrameric TTR controls. Although these studies were performed in vitro with a hepatoma cell line, they support the hypothesis that cytotoxic oligomeric species may be involved in the genesis of anemia in FAP patients. Besides this evidence, we recently showed that EPO concentrations in the aqueous humor of glaucomatous eyes of non-amyloidotic patients are significantly increased relatively to normal non-glaucomatous eyes, probably with a protective role [116]. However, in glaucomatous eyes of FAP ATTRV30M patients the EPO levels did not increase and maintained similar levels to those of control eyes [116]. These results show an inability of these patients to upregulate EPO production also locally in the retina. 5.2. Oligomeric TTR V30M aggregates reduce EPO expression in RPE cells Beyond using Hep3B cells, a tumoral cell line, primary cell cultures of retinal pigment epithelial cells (RPE) were also evaluated for EPO gene expression. RPE cells when are cultivated acquire a hexagonal conformation and are pigmented (figure 24). As the cells divide, they gradually lose their pigment and start differentiating, acquiring a more elongated and fibroblast-like conformation (figures 25 and 26). Preliminary results showed that RPE isolated from the eye of an adult human cadaveric donor express EPO mRNA, but only on early passages (P0 and P1). Results and Discussion 82 Figure 24 – Day 1 of a culture of retinal pigment epithelial cells (RPE) at passage 1 (P1), cultured on 25 cm3 flasks, visualized under light microscope with original magnifications x100 (A and B) and x200 (C-H). Note the abundance of pigment in the cytoplasm of these cells. Figure 25 – Day 3 (A) and day 8 (B and C) of a culture of retinal pigment epithelial cells (RPE) at passage 1 (P1), cultured on 25 cm3 flasks, visualized under light microscope with original magnifications x100 (A) and x200 (B-C). As the cells divide they start loosing their pigment. A B C D E F G H A B C Results and Discussion 83 Figure 26 – RPE cells at passage 2 (P2), cultured on 25 cm3 flasks, visualized under light microscope with original magnifications x100. At this passage the cells lost most of their pigment and aquired a more fibroblast-like conformation. RPE cells on passage 1 were exposed to TTRV30M preparations. As already shown in sections 3.1. and 3.2., oligomeric TTRV30M aggregates reduced RPE cell viability with concomitant increase in caspase 3/7 activity. Real-time PCR showed that, similarly to what happened for Hep3B cells, EPO expression after 24 hours was reduced by 48.3±17.1% (p= 7.0x10-3) in normoxic RPE cells treated with oligomeric TTRV30M when compared with tetrameric form experiments (figure 27). Figure 27 - Relative expression levels of erythropoietin in normoxic RPE cells exposed to TTRV30M for 24 hours. Values were normalized for TBP. p value was calculated relatively to tetrameric TTRV30M, using student’s t-test. *p<0,05 with respect to TTRV30M. Values are means of duplicates. Error Bar= SD. A B C Results and Discussion 90 sequester them away from the transcription machinery, resulting in transcriptional regulation [182-183, 299-302]. Xue et al propose that the formation of a complex between the ATF3, c-Jun and Sp1 explains the increase of EPO promoter activity induced by PDGF-BB [182]. They showed that the EPO promoter region does not contain an ATF3 binding site, suggesting that ATF3 may not act directly on the EPO promoter [182]. Elevated expression of PDGF has been observed in RPE cells after retinal detachments or retinal laser treatment in murine model systems [184], in in vitro wounded human RPE cell cultures [185] and in epi-retinal membranes isolated from proliferative vitreoretinopathy and proliferative diabetic retinopathy patients [186]. Besides Sp1 and c-Jun, ATF3 may also interact with NF-kB, p53 and p73. It may also inhibit transcription of pro-inflammatory cytokines, by interacting with NF-κB at the promoter of these target genes [301]. In our TTRV30M exposure Hep3B cell model, it appears that oligomeric TTRV30M aggregates inhibit EPO expression by decreasing EPO promoter activity. However, when the cells over-express ATF3 the inhibitory effect of these aggregates is reverved. This could be due either to increased activity of promoter inhibitors or to a decreased activity of ATF3, preventing it from sequestering these inhibitors away from the EPO promoter. This issue deserves further study. ATF3 has a basic region-leucine zipper (bZip) DNA binding domain. However, several isoforms derived from alternative splicing have been described (ATF3ΔZip, ATF3ΔZip2a, ATF3ΔZip2b, ATF3ΔZip2c, ATF3ΔZip3, and ATF3b), lacking the leucine zipper region [302]. The role of these alternative splicing forms needs also to be explored. CONCLUSIONS AND FUTURE PERSPECTIVES Conclusion and Future Perspectives 93 Conclusions and Future Perspectives In this project we carried out in vitro studies aiming to explore the role of amyloid deposition in ATTRV30M amyloidosis, especially regarding the role of early non-fibrillar aggregates on cell differentiation and erythropoietin production. First, the production of recombinant human TTRV30M and their oligomeric aggregates were optimized. Obtaining a pure and LPS-free protein for the cellular assays was crucial to ensure that the effects on the cells were only due to the protein preparations. The poor reproducibility of preparation of cytotoxic oligomeric solutions made it necessary to test several methods described in the literature, for optimization. Mild acidification (pH 4 to 5.5) of the protein solution, acidification to pH 2.0 followed by NaCl or aging under physiological pH and temperature followed by vigorous stirring were the methods chosen. In the case of acidification to mild pH we obtained insoluble material, probably amorphous aggregates that, despite being positive in ThT assays, were not toxic to cells. The method that induced the formation of most cytotoxic species and with the most reproducible results was the last one tested, aging the protein at physiological pH and 37ºC for 72 hours followed by 5 minutes of stirring. This TTR preparation contained mainly oligomers of approximately 300 nm when analysed by dynamic light scattering, and was the one used for the subsequent cell culture assays. Different cell culture models were used to evaluate the influence of oligomeric TTR aggregates on cell viability: using immortalized human hepatoblastoma Hep3B, neuroblastoma SH-SY5Y and embryonic kidney cells HEK293T, as well as primary retinal pigment epithelial cells RPE and human renal progenitor cells RPC. The proliferation of the cells exposed for 24 hours to oligomeric TTRV30M aggregates was moderately reduced when compared to the cells exposed to the tetrameric protein, independently of the cell type. Oligomers generated from the wild-type TTR also reduced cell proliferation in comparison to the soluble tetrameric form. Both oligomeric TTR and TTRV30M preparations were cytotoxic, a process that may be common to all prefibrillar amyloid protein aggregates. Concomitantly to the reduction in cell proliferation, a significant increase in caspases 3/7 activity occurred in cells exposed to oligomeric TTRV30M aggregates, implicating apoptosis in the loss of viability. These results are in agreement with other studies performed mainly in neuronal cells (neuroblastoma) [106, 229, 250, 270-272]. Interestingly, the cytotoxic effect induced by the aggregates seems to be independent of Conclusions and Future Perspectives 94 the cell type, since both immortalized, as well as primary and progenitor cells were shown to be susceptible to damage and apoptosis driven by aggregate exposure. Renal progenitor cells (RPC) express the stem cell markers CD133+CD24+ and may potentially differentiate into podocytes and tubular cells. We evaluated the influence of the oligomeric aggregates on the viability and differentiation capacity of the RPC, in order to understand if the misfolded protein could compromise a potential therapeutic use of these cells to treat renal damage. Proliferation of RPC in vitro was impared by the oligomeric TTRV30M aggregates comparatively to the tetrameric form, but their differentiation capacity into mature podocytes was not significantly affected, so renal tissue regeneration should not be compromised by the presence of early oligomeric species. ATTRV30M amyloidosis patients develop anemia with low EPO levels. The impaired EPO production is not related to the extent or pattern of congophilic renal deposition or with the presence of circulating mutant TTRV30M. The role of oligomeric aggregate cytotoxicity on EPO production had not been studied yet. In this work we used the EPO producing cell line Hep3B to evaluate the influence of oligomeric TTRV30M aggregates on EPO gene expression, as well as on pathways linked to the regulatory regions of the EPO gene. Normoxic Hep3B exposed for 24 hours to early oligomeric TTRV30M aggregates showed a 2-fold reduction in EPO mRNA expression. In accordance to this reduction, there was a 2-fold reduction in EPO promoter activity in transfected Hep3B cells. These results support involvement of early oligomeric TTR aggregates in decreased EPO gene expression and could explain the anemia and low EPO levels in FAP patients, seen even in the pre-symptomatic phase of the disease. Interestingly, a preparation of wild-type TTR oligomers also reduced EPO mRNA expression in normoxia, although to a less extent than the TTRV30M oligomers, which may explain the continuing low EPO production after liver transplantation seen in FAP patients. The inhibition of EPO expression caused by exposure of the cells to cytotoxic oligomeric aggregates seems to be mediated by decreased activity of the EPO promoter. Transcription factors such as NF-kB and GATA-2 bind to the promoter and inhibit transcription of the EPO gene, mainly in inflammatory states. Here, we obtained inconclusive results about the influence of oligomeric aggregates on NF-kB or GATA-2 activation. However, we believe this issue deserves further consideration, as well as other signaling pathways that may be activated by the oligomeric species and that may influence EPO expression. Conclusion and Future Perspectives 95 FAP patients are also unable to up-regulate EPO production in the glaucomatous eye, as a neuroprotective response. Besides the possible involvement of TTR aggregates in EPO expression and anemia, a much wider role for these aggregates in disease progression could be considered. Recent studies point to a potentially major role of EPO in neurodegenerative diseases, including Alzheimer's disease, Parkinson's disease, epilepsy, multiple sclerosis and motor neuron diseases [151, 303-311]. It has been already shown in another model of amyloid-related neurodegeneration, in which PC12 cells are exposed to Aβ25-35 aggregates, that apoptosis is counteracted by EPO signaling [312-313]. EPO has also been found to increase the resistance of neurons to damage induced by pro-inflammatory agents [314]. It could be postulated that deposition of TTR aggregates in the peripheral nervous system down-regulates local EPO expression, depriving neurons from needed neuroprotective signals, contributing thus for the onset and/or progression of the disease, an hypothesis that we feel deserves further study. Retinal pigment epithelial cells (RPE) produce both EPO and TTR in the eye. In order to find a more suitable cell culture model to explore the direct effect of the amyloidogenic TTRV30M on EPO production, we isolated RPE cells from an eye of a human cadaveric donor. Preliminary results showed that, at early passages, these cells express EPO in normoxia and respond to hypoxia simulated with DMOG. Besides the reduction of cell proliferation and increased activity of caspases 3/7, EPO gene expression in normoxic RPE treated for 24 hours with oligomeric TTRV30M aggregates was reduced by approximately 2-fold comparatively to the cells exposed to tetrameric TTRV30M, similarly to what was seen in Hep3B cells. This is an important result as it shows a direct influence of the oligomeric species on a primary cell line that is physiologically responsible for EPO production in the eye. As future perspectives, and taking into consideration these preliminary results, we believe that RPE cells could become a suitable cell culture model to further explore the mechanisms responsible for EPO inhibition in ATTRV30M amyloidosis, ideally, using RPE cells from FAP patients. The blockage of EPO production observed both in the kidney as in the eye of FAP patients also happens in Diabetic Nephropathy (DN). Oxidative stress, overproduction of reactive oxygen species (ROS), production of AGEs, activation of the receptor for advanged glycation end products (RAGE), upregulation of the NF-κB and proinflammatory cytokines seem to be common factors between FAP and DN. ROS may upregulate GATA-2. In this work we used immunofluorescence to study the activation of Conclusions and Future Perspectives 96 NF-kB and GATA-2 by exposure to oligomeric aggregates and we could not see significant differences. However, it would be important to use more sensitive techniques, such as Chromatin immunoprecipitation or electrophoretic mobility shift assays (EMSA), which can show protein-DNA interaction, to confirm if the inhibition of EPO expression by cytotoxic aggregates is indeed related to activation of these two transcription factors. ATF3 is other transcription factor that may be involved in EPO gene regulation. In our cell model, the over-expression of ATF3 reversed the inhibitory effect of the oligomeric TTR aggregates on the EPO promoter activity. This is a preliminary observation that deserves further study. It would also be interesting to use DN patients for comparison. Tafamidis is a drug already approved for the treatment of eligible FAP patients. We would like to know if Tafamidis, a TTR tetramer stabilizer, is able to prevent blockage of EPO production. In summary, although we have shown here that exposure to cytotoxic oligomeric TTR aggregates inhibits EPO expression in vitro, much remains to be discovered about the mechanisms underlying this inhibition and possible therapeutic targets. REFERENCES References 99 References 1 Cohen AS, Calkins E. Electron microscopic observations on a fibrous component in amyloid of diverse origins. Nature. 1959 Apr 25;183(4669):1202-3. 2 Glenner GG, Eanes ED, Bladen HA, Linke RP, Termine JD. Beta-pleated sheet fibrils. A comparison of native amyloid with synthetic protein fibrils. J Histochem Cytochem. 1974 Dec;22(12):1141-58. 3 Cohen AS, Shirahama T, Sipe JD, Skinner M. Amyloid proteins, precursors, mediator, and enhancer. Lab Invest. 1983 Jan;48(1):1-4. 4 Sipe JD, Benson MD, Buxbaum JN, Ikeda S, Merlini G, Saraiva MJ, Westermark P. Nomenclature 2014: Amyloid fibril proteins and clinical classification of the amyloidosis. Amyloid. 2014 Dec;21(4):221-4. 5 Sipe JD, Cohen AS. Review: history of the amyloid fibril. J Struct Biol. 2000 Jun;130(2-3):88-98. 6 Kyle RA. Amyloidosis: a convoluted story. Br J Haematol. 2001 Sep;114(3):529-38. 7 Westermark GT, Norling B, Westermark P. Fibronectin and basement membrane components in renal amyloid deposits in patients with primary and secondary amyloidosis. Clin Exp Immunol. 1991 Oct;86(1):150-6. 8 Benditt EP. What role(s) may extracellular matrix, particularly heparan sulfate, play in amyloid of Alzheimer's disease. Neurobiol Aging. 1989 Sep-Oct;10(5):506; discussion 510-2. 9 Wisniewski T, Golabek A, Matsubara E, Ghiso J, Frangione B. Apolipoprotein E: binding to soluble Alzheimer's beta-amyloid. Biochem Biophys Res Commun. 1993 Apr 30;192(2):359-65. 10 de Beer MC, de Beer FC, McCubbin WD, Kay CM, Kindy MS. Structural prerequisites for serum amyloid A fibril formation. J Biol Chem. 1993 Sep 25;268(27):20606-12. 11 Moss J, Shore I, Woodrow D. AA glomerular amyloid. An ultrastructural immunogold study of the colocalization of heparan sulphate proteoglycan and P component with amyloid fibrils together with changes in distribution of type IV collagen and fibronectin. Histopathology. 1994 May;24(5):427-35. 12 Pepys MB. Pathogenesis, diagnosis and treatment of systemic amyloidosis. Philos Trans R Soc Lond B Biol Sci. 2001 Feb 28;356(1406):203-10; discussion 210-1. References 106 Antibodies to human serum amyloid P component eliminate visceral amyloid deposits. Nature. 2010 Nov 4;468(7320):93-7. 89 Sasaki H, Sakaki Y, Matsuo H, Goto I, Kuroiwa Y, Sahashi I, Takahashi A, Shinoda T, Isobe T, Takagi Y. Diagnosis of familial amyloidotic polyneuropathy by recombinant DNA techniques. Biochem Biophys Res Commun. 1984 Dec 14;125(2):636-42. 90 Ando Y, Ueda M. Diagnosis and therapeutic approaches to transthyretin amyloidosis. Curr Med Chem. 2012;19(15):2312-23. 91 Ando Y, Ohlsson PI, Suhr O, Nyhlin N, Yamashita T, Holmgren G, Danielsson A, Sandgren O, Uchino M, Ando M. A new simple and rapid screening method for variant transthyretin-related amyloidosis. Biochem Biophys Res Commun. 1996 Nov 12;228(2):480-3. 92 Yamashita T, Ando Y, Bernt Suhr O, Nakamura M, Sakashita N, Ohlsson PI, Terazaki H, Obayashi K, Uchino M, Ando M. A new diagnostic procedure to detect unknown transthyretin (TTR) mutations in familial amyloidotic polyneuropathy (FAP). J Neurol Sci. 2000 Feb 15;173(2):154-9. 93 Bergquist J, Andersen O, Westman A. Rapid method to characterize mutations in transthyretin in cerebrospinal fluid from familial amyloidotic polyneuropathy patients by use of matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Clin Chem. 2000 Sep;46(9):1293-300. 94 Lacerda PC, Moreira L, Vitorino R, Costa PP. Use of MALDI-TOF Mass Spectrometry to Assay the Transthyretin V30M Mutation in Serum From a Liver Transplant Donor: A Case Report. Transplantation (in press). 95 Beirão JM, Moreira LV, Lacerda PC, Vitorino RP, Beirão IB, Torres PA, Costa PP. Inability of mutant transthyretin V30M to cross the blood-eye barrier. Transplantation. 2012 Oct 27;94(8):e54-6. 96 Beirão I, Lobato L, Costa PM, Fonseca I, Mendes P, Silva M, Bravo F, Cabrita A, Porto G. Kidney and anemia in familial amyloidosis type I. Kidney Int. 2004 Nov;66(5):2004-9. 97 Beirão I, Lobato L, Moreira L, Mp Costa P, Fonseca I, Cabrita A, Porto G. Long-term treatment of anemia with recombinant human erythropoietin in familial amyloidosis TTR V30M. Amyloid. 2008 Sep;15(3):205-9. 98 World Health Organization. Nutritional anaemias: Report of a WHO scientific group. Geneva, Switzerland; 1968 Jan 1. 99 Olofsson BO, Grankvist K, Boman K: Evaluation of the anemia in familial amyloidotic polyneuropahy. Eur J Int Med. 1990; 1: 425-429 References 107 100 Asahara K, Ando Y, Tanaka Y, Yi S, Yamashita T, Ando M. Secondary hypoplastic anemia in patients with familial amyloidotic polyneuropathy. Acta Haematol. 1993;90(3):130-5. 101 Beirão I, Moreira L, Porto G, Lobato L, Fonseca I, Cabrita A, Costa PM. Low erythropoietin production in familial amyloidosis TTR V30M is not related with renal congophilic amyloid deposition. A clinicopathologic study of twelve cases. Nephron Clin Pract. 2008;109(2):c95-9. 102 Faquin WC, Schneider TJ, Goldberg MA. Effect of inflammatory cytokines on hypoxia-induced erythropoietin production. Blood. 1992 Apr 15;79(8):1987-94. 103 Vasilopoulos S, Hally R, Caro J, Martin P, Westerberg S, Moritz M, Jarrell B, Muñoz S. Erythropoietin response to post-liver transplantation anemia. Liver Transpl. 2000 May;6(3):349-55. 104 Beirão I, Lobato L, Costa PM, Fonseca I, Silva M, Bravo F, Cabrita A, Porto G. Liver transplantation and anemia in familial amyloidosis ATTR V30M. Amyloid. 2007 Mar;14(1):33-7. 105 Means RT Jr, Krantz SB. Progress in understanding the pathogenesis of the anemia of chronic disease. Blood. 1992 Oct 1;80(7):1639-47. 106 Sousa MM, Du Yan S, Fernandes R, Guimaraes A, Stern D, Saraiva MJ. Familial amyloid polyneuropathy: receptor for advanced glycation end products-dependent triggering of neuronal inflammatory and apoptotic pathways. J Neurosci. 2001;21(19):7576-7586. 107 Sousa MM, Cardoso I, Fernandes R, Guimarães A, Saraiva MJ. Deposition of transthyretin in early stages of familial amyloidotic polyneuropathy: evidence for toxicity of nonfibrillar aggregates. Am J Pathol. 2001 Dec;159(6):1993-2000. 108 Beirão I, Almeida S, Swinkels D, Costa PM, Moreira L, Fonseca I, Freitas C, Cabrita A, Porto G. Low serum levels of prohepcidin, but not hepcidin-25, are related to anemia in familial amyloidosis TTR V30M. Blood Cells Mol Dis. 2008 SepOct;41(2):175-8. 109 Beirão M, Matos E, Beirâo I, Costa PP, Torres P. Anticipation of presbyopia in Portuguese familial amyloidosis ATTR V30M. Amyloid. 2011 Sep;18(3):92-7. 110 Beirão JM, Matos ME, Beirão IB, Costa PP, Torres PA. Topical cyclosporine for severe dry eye disease in liver-transplanted Portuguese patients with familial amyloidotic polyneuropathy (ATTRV30M). Eur J Ophthalmol. 2013 MarApr;23(2):156-63. 111 Benson MD. The hereditary amyloidoses. Best Pract Res Clin Rheumatol. 2003 Dec;17(6):909-27. References 108 112 Beirão NM, Matos E, Beirão I, Costa PP, Torres P. Recurrence of vitreous amyloidosis and need of surgical reintervention in Portuguese patients with familial amyloidosis ATTR V30M. Retina. 2011 Jul-Aug;31(7):1373-7. 113 Beirão NM, Matos ME, Meneres MJ, Beirão IM, Costa PP, Torres PA. Vitreous surgery impact in glaucoma development in liver transplanted familial amyloidosis ATTR V30M Portuguese patients. Amyloid. 2012 Sep;19(3):146-51. 114 Hernández C, Simó R. Erythropoietin produced by the retina: its role in physiology and diabetic retinopathy. Endocrine. 2012 Apr;41(2):220-6. 115 Genc S, Koroglu TF, Genc K. Erythropoietin as a novel neuroprotectant. Restor Neurol Neurosci. 2004;22(2):105-19. 116 Beirão JM, Moreira LM, Oliveira JC, Menéres MJ, Pessoa BB, Matos ME, Costa PP, Torres PA, Beirão IB. Aqueous humor erythropoietin levels in open-angle glaucoma patients with and without TTR V30M familial amyloid polyneuropathy. Mol Vis. 2014 Jul 2;20:970-6. 117 Grimm C, Willmann G. Hypoxia in the eye: a two-sided coin. High Alt Med Biol. 2012 Sep;13(3):169-75. 118 Tezel G, Wax MB. Hypoxia-inducible factor 1alpha in the glaucomatous retina and optic nerve head. Arch Ophthalmol. 2004 Sep;122(9):1348-56. 119 Thomas M, Tsalamandris C, MacIsaac R, Jerums G. Anaemia in diabetes: an emerging complication of microvascular disease. Curr Diabetes Rev. 2005 Feb;1(1):107-26. 120 Eckardt KU. Biology of erythropoietin production. Nephrol Dial Transplant. 1995;10(9):1572-4. 121 Fandrey J, Bunn HF. In vivo and in vitro regulation of erythropoietin mRNA: measurement by competitive polymerase chain reaction. Blood. 1993 Feb 1;81(3):617-23. 122 Zhang Y, Wang L, Dey S, Alnaeeli M, Suresh S, Rogers H, Teng R, Noguchi CT. Erythropoietin action in stress response, tissue maintenance and metabolism. Int J Mol Sci. 2014 Jun 10;15(6):10296-333. 123 Brines M, Cerami A. Discovering erythropoietin's extra-hematopoietic functions: biology and clinical promise. Kidney Int. 2006; 70(2):246-250. 124 Bunn HF. Erythropoietin. Cold Spring Harb Perspect Med. 2013 Mar 1;3(3):a011619. 125 Konstantinopoulos PA, Karamouzis MV, Papavassiliou AG. Selective modulation of the erythropoietic and tissue-protective effects of erythropoietin: Time to reach the full therapeutic potential of erythropoietin. Biochemica et Biophysica Acta 2007; 1776: 1-9. References 109 126 Witzig TE, Silberstein PT, Loprinzi CL, Sloan JA Phase III, randomized, double-blind study of epoietin alpha compared with placebo in anemic patients receiving chemotherapy. J. Clin. Oncol. 2005; 23: 2606-2617. 127 Lee JS, Ha TK, Lee SJ, Lee GM. Current state and perspectives on erythropoietin production. Appl Microbiol Biotechnol. 2012 Sep;95(6):1405-16. 128 Dame C, Fahnenstich H, Freitag P, Hofmann D, Abdul-Nour T, Bartmann P, Fandrey J. Erythropoietin mRNA expression in human fetal and neonatal tissue. Blood. 1998 Nov 1;92(9):3218-25. 129 Dame C, Sola MC, Lim KC, Leach KM, Fandrey J, Ma Y, Knöpfle G, Engel JD, Bungert J. Hepatic erythropoietin gene regulation by GATA-4. J Biol Chem. 2004 Jan 23;279(4):2955-61. 130 Burlington H, Cronkite EP, Reincke U, Zanjani ED. Erythropoietin production in cultures of goat renal glomeruli. Proc Natl Acad Sci U S A. 1972 Dec;69(12):354750. 131 Koury ST, Bondurant MC, Koury MJ. Localization of erythropoietin synthesizing cells in murine kidneys by in situ hybridization. Blood. 1988 Feb;71(2):524-7. 132 Lacombe C, Da Silva JL, Bruneval P, Fournier JG, Wendling F, Casadevall N, Camilleri JP, Bariety J, Varet B, Tambourin P. Peritubular cells are the site of erythropoietin synthesis in the murine hypoxic kidney. J Clin Invest. 1988 Feb;81(2):620-3. 133 Fisher JW, Koury S, Ducey T, Mendel S. Erythropoietin production by interstitial cells of hypoxic monkey kidneys. Br J Haematol. 1996 Oct;95(1):27-32. 134 Loya F, Yang Y, Lin H, Goldwasser E, Albitar M. Transgenic mice carrying the erythropoietin gene promoter linked to lacZ express the reporter in proximal convoluted tubule cells after hypoxia. Blood. 1994 Sep 15;84(6):1831-6. 135 Maxwell AP, Lappin TR, Johnston CF, Bridges JM, McGeown MG. Erythropoietin production in kidney tubular cells. Br J Haematol. 1990 Apr;74(4):535-9. 136 Horiguchi H, Oguma E, Kayama F. Cadmium and cisplatin damage erythropoietinproducing proximal renal tubular cells. Arch Toxicol. 2006 Oct;80(10):680-6. 137 Shanks JH, Hill CM, Lappin TR, Maxwell AP. Localization of erythropoietin gene expression in proximal renal tubular cells detected by digoxigenin-labelled oligonucleotide probes. J Pathol. 1996 Jul;179(3):283-7. 138 Beirão I, Moreira L, Barandela T, Lobato L, Silva P, Gouveia CM, Carneiro F, Fonseca I, Porto G, Pinho e Costa P. Erythropoietin production by distal nephron in normal and familial amyloidotic adult human kidneys. Clin Nephrol. 2010;74(5):327335. References 110 139 Bussolati B, Lauritano C, Moggio A, Collino F, Mazzone M, Camussi G. Renal CD133(+)/CD73(+) progenitors produce erythropoietin under hypoxia and prolyl hydroxylase inhibition. J Am Soc Nephrol. 2013 Jul;24(8):1234-41. 140 Bussolati B, Bruno S, Grange C, Buttiglieri S, Deregibus MC, Cantino D, Camussi G. Isolation of renal progenitor cells from adult human kidney. Am J Pathol. 2005 Feb;166(2):545-55. 141 Nagai T, Yasuoka Y, Izumi Y, Horikawa K, Kimura M, Nakayama Y, Uematsu T, Fukuyama T, Yamazaki T, Kohda Y, Hasuike Y, Nanami M, Kuragano T, Kobayashi N, Obinata M, Tomita K, Tanoue A, Nakanishi T, Kawahara K, Nonoguchi H. Reevaluation of erythropoietin production by the nephron. Biochem Biophys Res Commun. 2014;449(2):222-228. 142 Masuda S, Okano M, Yamagishi K, Nagao M, Ueda M, Sasaki R. A novel site of erythropoietin production. Oxygen-dependent production in cultured rat astrocytes. J Biol Chem. 1994; 269(30): 19488-19493. 143 Yasuda Y, Masuda S, Chikuma M, Inoue K., Nagao M., Sasaki R. Estrogen dependent production of erythropoietin in uterus and its implication in uterine angiogenesis. J Biol Chem 1998; 273:25381-25387. 144 Magnanti M, Gandini O, Giuliani L, Gazzaniga P, Marti HH, Gradilone A, Frati L, Aglianò AM, Gassmann M. Erythropoietin expression in primary rat Sertoli and peritubular myoid cells. Blood 2001; 98:2872-2874. 145 Conrad KP, Benyo DF, Westerhausen-Larsen A, Miles TM. Expression of erythropoietin by the human placenta. FASEB J 1996; 10(7): 760-768. 146 Goldberg MA, Glass GA, Cunningham JM, Bunn HF. The regulated expression of erythropoietin by two human hepatoma cell lines. Proc Natl Acad Sci U S A. 1987 Nov;84(22):7972-6. 147 Ghezzi P, Brines M. Erythropoietin as an antiapoptotic, tissue-protective cytokine. Cell Death Differ. 2004 Jul;11 Suppl 1:S37-44. 148 Udupa K.B. Functional Significance of Erythropoietin Receptor on Tumor Cells. World J Gastroenterol 2006; 12(46): 7460-7462. 149 Pregi N, Wenker S, Vittori D, Leirós CP, Nesse A. TNF-alpha-induced apoptosis is prevented by erythropoietin treatment on SH-SY5Y cells. Exp Cell Res. 2009 Feb 1;315(3):419-31. 150 Broxmeyer HE. Erythropoietin: multiple targets, actions, and modifying influences for biological and clinical consideration. J Exp Med. 2013 Feb 11;210(2):205-8. 151 Bahcekap.l. N, Akgun-Dar K, Albeniz I, Kapucu A, Kandil A, Ya..z O, Uzum G. Erythropoietin pretreatment suppresses seizures and prevents the increase in References 111 inflammatory mediators during pentylenetetrazole-induced generalized seizures. Int J Neurosci. 2014 Feb 6. 152 Martinez F, Pallet N. When erythropoietin meddles in immune affairs. J Am Soc Nephrol. 2014 Sep;25(9):1887-9. 153 Brines M, Cerami A. The receptor that tames the innate immune response. Mol Med. 2012 May 9;18:486-96. 154 Madonna R, Shelat H, Xue Q, Willerson JT, De Caterina R, Geng YJ. Erythropoietin protects myocardin-expressing cardiac stem cells against cytotoxicity of tumor necrosis factor-alpha. Exp Cell Res. 2009 Oct 15;315(17):2921-8. 155 Duran-Salgado MB, Rubio-Guerra AF. Diabetic nephropathy and inflammation. World J Diabetes. 2014 Jun 15;5(3):393-8. 156 Oster HS, Neumann D, Hoffman M, Mittelman M. Erythropoietin: the swinging pendulum. Leuk Res. 2012 Aug;36(8):939-44. 157 Jelkmann W. Physiology and pharmacology of erythropoietin. Transfus Med Hemother. 2013 Oct;40(5):302-9. 158 Jelkmann W. Molecular biology of erythropoietin. Intern Med. 2004 Aug;43(8):64959. 159 Jelkmann W. Erythropoietin. J Endocrinol Invest. 2003 Sep;26(9):832-7. 160 Jelkmann W. Regulation of erythropoietin production. J Physiol. 2011;589(Pt 6):1251-1258. 161 Storti F, Santambrogio S, Crowther LM, Otto T, Abreu-Rodríguez I, Kaufmann M, Hu CJ, Dame C, Fandrey J, Wenger RH, Hoogewijs D. A novel distal upstream hypoxia response element regulating oxygen-dependent erythropoietin gene expression. Haematologica. 2014 Apr;99(4):e45-48. 162 McGary EC, Rondon IJ, Beckman BS. Post-transcriptional regulation of erythropoietin mRNA stability by erythropoietin mRNA-binding protein. J Biol Chem. 1997;272(13):8628-8634. 163 Barbosa C, Romão L. Translation of the human erythropoietin transcript is regulated by an upstream open reading frame in response to hypoxia. RNA. 2014;20(5):594608. 164 Stockmann C, Fandrey J. Hypoxia-induced erythropoietin production: a paradigm for oxygen-regulated gene expression. Clin Exp Pharmacol Physiol. 2006;33(10): 968979. 165 Wang GL, Semenza GL. General involvement of hypoxia-inducible factor 1 in transcriptional response to hypoxia. Proc Natl Acad Sci USA. 1993;90(9): 43044308. References 112 166 Imagawa S, Yamamoto M, Miura Y. Negative regulation of the erythropoietin gene expression by the GATA transcription factors. Blood. 1997;89(4):1430-1439. 167 Obara N, Suzuki N, Kim K, Nagasawa T, Imagawa S, Yamamoto M. Repression via the GATA box is essential for tissue-specific erythropoietin gene expression. Blood. 2008;111(10):5223-5232. 168 La Ferla K, Reimann C, Jelkmann W, Hellwig-Bürgel T. Inhibition of erythropoietin gene expression signaling involves the transcription factors GATA-2 and NFkappaB. FASEB J. 2002;16(13):1811-1813. 169 Semenza GL, Wang GL. A nuclear factor induced by hypoxia via de novo protein synthesis binds to the human erythropoietin gene enhancer at a site required for transcriptional activation. Mol Cell Biol. 1992 Dec;12(12):5447-54. 170 Semenza GL. Involvement of oxygen-sensing pathways in physiologic and pathologic erythropoiesis. Blood. 2009 Sep 3;114(10):2015-9. 171 Haase VH. Regulation of erythropoiesis by hypoxia-inducible factors. Blood Rev. 2013 Jan;27(1):41-53. 172 Kapitsinou PP, Liu Q, Unger TL, Rha J, Davidoff O, Keith B, Epstein JA, Moores SL, Erickson-Miller CL, Haase VH. Hepatic HIF-2 regulates erythropoietic responses to hypoxia in renal anemia. Blood. 2010 Oct 21;116(16):3039-48. 173 Rosenberger C, Mandriota S, Jürgensen JS, Wiesener MS, Hörstrup JH, Frei U, Ratcliffe PJ, Maxwell PH, Bachmann S, Eckardt KU. Expression of hypoxiainducible factor-1alpha and -2alpha in hypoxic and ischemic rat kidneys. J Am Soc Nephrol. 2002 Jul;13(7):1721-32. 174 Paliege A, Rosenberger C, Bondke A, Sciesielski L, Shina A, Heyman SN, Flippin LA, Arend M, Klaus SJ, Bachmann S. Hypoxia-inducible factor-2alpha-expressing interstitial fibroblasts are the only renal cells that express erythropoietin under hypoxia-inducible factor stabilization. Kidney Int. 2010 Feb;77(4):312-8. 175 http://www.astellas.com/en/ir/pdf/4503Astellas_Roxadusutat.pdf 176 http://www.ukmi.nhs.uk/applications/ndo/record_view_open.asp?newDrugID=5878 177 Tabata M, Tarumoto T, Ohmine K, Furukawa Y, Hatake K, Ozawa K, Hasegawa Y, Mukai H, Yamamoto M, Imagawa S. Stimulation of GATA-2 as a mechanism of hydrogen peroxide suppression in hypoxia-induced erythropoietin gene expression. J Cell Physiol. 2001 Feb;186(2):260-7. 178 Imagawa S, Yamamoto M, Ueda M, Miura Y. Erythropoietin gene expression by hydrogen peroxide. Int J Hematol. 1996 Oct;64(3-4):189-95. 179 Imagawa S, Suzuki N, Ohmine K, Obara N, Mukai HY, Ozawa K, Yamamoto M, Nagasawa T. GATA suppresses erythropoietin gene expression through GATA site in mouse erythropoietin gene promoter. Int J Hematol. 2002 May;75(4):376-81. References 113 180 Tarumoto T, Imagawa S, Ohmine K, Nagai T, Higuchi M, Imai N, Suzuki N, Yamamoto M, Ozawa K. N(G)-monomethyl-L-arginine inhibits erythropoietin gene expression by stimulating GATA-2. Blood. 2000 Sep 1;96(5):1716-22. 181 Souma T, Yamazaki S, Moriguchi T, Suzuki N, Hirano I, Pan X, Minegishi N, Abe M, Kiyomoto H, Ito S, Yamamoto M. Plasticity of renal erythropoietin-producing cells governs fibrosis. J Am Soc Nephrol. 2013 Oct;24(10):1599-616. 182 Xue Y, Lim S, Yang Y, Wang Z, Jensen LD, Hedlund EM, Andersson P, Sasahara M, Larsson O, Galter D, Cao R, Hosaka K, Cao Y. PDGF-BB modulates hematopoiesis and tumor angiogenesis by inducing erythropoietin production in stromal cells. Nat Med. 2011 Dec 4;18(1):100-10. 183 Hai T, Wolfgang CD, Marsee DK, Allen AE, Sivaprasad U. ATF3 and stress responses. Gene Expr. 1999;7(4-6):321-35. 184 Seo MS, Okamoto N, Vinores MA, Vinores SA, Hackett SF, et al. Photoreceptorspecific expression of platelet-derived growth factor-B results in traction retinal detachment. Am J Pathol. 2000;157:995-1005. 185 Campochiaro PA, Hackett SF, Vinores SA, Freund J, Csaky C, et al. Plateletderived growth factor is an autocrine growth stimulator in retinal pigmented epithelial cells. J Cell Sci. 1994;107(Pt 9):2459-69. 186 Vinores SA, Henderer JD, Mahlow J, Chiu C, Derevjanik NL, et al. Isoforms of platelet-derived growth factor and its receptors in epiretinal membranes: immunolocalization to retinal pigmented epithelial cells. Exp Eye Res. 1995;60:60719 187 Makita T, Duncan SA, Sucov HM. Retinoic acid, hypoxia, and GATA factors cooperatively control the onset of fetal liver erythropoietin expression and erythropoietic differentiation. Dev Biol. 2005 Apr 1;280(1):59-72. 188 Dame C, Kirschner KM, Bartz KV, Wallach T, Hussels CS, Scholz H. Wilms tumor suppressor, Wt1, is a transcriptional activator of the erythropoietin gene. Blood. 2006 Jun 1;107(11):4282-90. 189 Sánchez-Elsner T, Ramírez JR, Sanz-Rodriguez F, Varela E, Bernabéu C, Botella LM. A cross-talk between hypoxia and TGF-beta orchestrates erythropoietin gene regulation through SP1 and Smads. J Mol Biol. 2004 Feb 6;336(1):9-24. 190 Peterson PA. Studies on the interaction between prealbumin, retinol-binding protein, and vitamin A. J Biol Chem. 1971 Jan 10;246(1):44-9. 191 Oppenheimer JH. Role of plasma proteins in the binding, distribution and metabolism of the thyroid hormones. N Engl J Med. 1968 May 23;278(21):1153-62. 192 Stabilini R, Vergani C, Agostoni A, Agostoni RP. Influence of age and sex on prealbumin levels. Clin Chim Acta. 1968 May;20(2):358-9. References 114 193 Vatassery GT, Quach HT, Smith WE, Benson BA, Eckfeldt JH. A sensitive assay of transthyretin (prealbumin) in human cerebrospinal fluid in nanogram amounts by ELISA. Clin Chim Acta. 1991 Feb 28;197(1):19-25. 194 Schussler GC. The thyroxine-binding proteins. Thyroid. 2000 Feb;10(2):141-9. 195 Aleshire SL, Bradley CA, Richardson LD, Parl FF. Localization of human prealbumin in choroid plexus epithelium. J Histochem Cytochem. 1983 May;31(5):608-12. 196 Schreiber G. The evolutionary and integrative roles of transthyretin in thyroid hormone homeostasis. J Endocrinol. 2002 Oct;175(1):61-73. 197 Herbert J, Wilcox JN, Pham KT, Fremeau RT Jr, Zeviani M, Dwork A, Soprano DR, Makover A, Goodman DS, Zimmerman EA, et al. Transthyretin: a choroid plexusspecific transport protein in human brain. The 1986 S. Weir Mitchell award. Neurology. 1986 Jul;36(7):900-11. 198 Getz RK, Kennedy BG, Mangini NJ. Transthyretin localization in cultured and native human retinal pigment epithelium. Exp Eye Res. 1999 May;68(5):629-36. 199 Jacobsson B, Collins VP, Grimelius L, Pettersson T, Sandstedt B, Carlström A. Transthyretin immunoreactivity in human and porcine liver, choroid plexus, and pancreatic islets. J Histochem Cytochem. 1989 Jan;37(1):31-7. 200 Gitlin D, Perricelli A. Synthesis of serum albumin, prealbumin, alpha-foetoprotein, alpha-1-antitrypsin and transferrin by the human yolk sac. Nature. 1970 Dec 5;228(5275):995-7. 201 McKinnon B, Li H, Richard K, Mortimer R. Synthesis of thyroid hormone binding proteins transthyretin and albumin by human trophoblast. J Clin Endocrinol Metab. 2005 Dec;90(12):6714-20. 202 Loughna S, Bennett P, Moore G. Molecular analysis of the expression of transthyretin in intestine and liver from trisomy 18 fetuses. Hum Genet. 1995 Jan;95(1):89-95. 203 Sasaki H, Yoshioka N, Takagi Y, Sakaki Y. Structure of the chromosomal gene for human serum prealbumin. Gene. 1985;37(1-3):191-7. 204 Tsuzuki T, Mita S, Maeda S, Araki S, Shimada K. Structure of the human prealbumin gene. J Biol Chem. 1985 Oct 5;260(22):12224-7. 205 Kanda Y, Goodman DS, Canfield RE, Morgan FJ. The amino acid sequence of human plasma prealbumin. J Biol Chem. 1974 Nov 10;249(21):6796-805. 206 Blake CC, Geisow MJ, Oatley SJ, Rérat B, Rérat C. Structure of prealbumin: secondary, tertiary and quaternary interactions determined by Fourier refinement at 1.8 A. J Mol Biol. 1978 May 25;121(3):339-56. References 115 207 Robbins J, Cheng SY, Gershengorn MC, Glinoer D, Cahnmann HJ, Edelnoch H. Thyroxine transport proteins of plasma. Molecular properties and biosynthesis. Recent Prog Horm Res. 1978;34:477-519. 208 Wojtczak A, Cody V, Luft JR, Pangborn W. Structures of human transthyretin complexed with thyroxine at 2.0 A resolution and 3',5'-dinitro-N-acetyl-L-thyronine at 2.2 A resolution. Acta Crystallogr D Biol Crystallogr. 1996 Jul 1;52(Pt 4):758-65. 209 Monaco HL, Rizzi M, Coda A. Structure of a complex of two plasma proteins: transthyretin and retinol-binding protein. Science. 1995 May 19;268(5213):1039-41. 210 Episkopou V, Maeda S, Nishiguchi S, Shimada K, Gaitanaris GA, Gottesman ME, Robertson EJ. Disruption of the transthyretin gene results in mice with depressed levels of plasma retinol and thyroid hormone. Proc Natl Acad Sci U S A. 1993 Mar 15;90(6):2375-9. 211 Sousa MM, Saraiva MJ. Internalization of transthyretin. Evidence of a novel yet unidentified receptor-associated protein (RAP)-sensitive receptor. J Biol Chem. 2001 Apr 27;276(17):14420-5. 212 Sousa MM, Norden AG, Jacobsen C, Willnow TE, Christensen EI, Thakker RV, Verroust PJ, Moestrup SK, Saraiva MJ. Evidence for the role of megalin in renal uptake of transthyretin. J Biol Chem. 2000 Dec 8;275(49):38176-81. 213 Divino CM, Schussler GC. Receptor-mediated uptake and internalization of transthyretin. J Biol Chem. 1990 Jan 25;265(3):1425-9. 214 Liz MA, Faro CJ, Saraiva MJ, Sousa MM. Transthyretin, a new cryptic protease. J Biol Chem. 2004 May 14;279(20):21431-8. 215 Geneste A, Guillaume YC, Magy-Bertrand N, Lethier L, Gharbi T, André C. The protease activity of transthyretin reverses the effect of pH on the amyloid-ß protein/heparan sulfate proteoglycan interaction: a biochromatographic study. J Pharm Biomed Anal. 2014 Aug;97:88-96. 216 Quintas A, Saraiva MJ, Brito RM. The amyloidogenic potential of transthyretin variants correlates with their tendency to aggregate in solution. FEBS Lett. 1997 Dec 1;418(3):297-300. 217 Chiti F, Stefani M, Taddei N, Ramponi G, Dobson CM. Rationalization of the effects of mutations on peptide and protein aggregation rates. Nature. 2003 Aug 14;424(6950):805-8. 218 Bhak G, Choe YJ, Paik SR. Mechanism of amyloidogenesis: nucleation-dependent fibrillation versus double-concerted fibrillation. BMB Rep. 2009 Sep 30;42(9):541-51. 219 Kodali R, Wetzel R. Polymorphism in the intermediates and products of amyloid assembly. Curr Opin Struct Biol. 2007 Feb;17(1):48-57. References 122 293 Takino J, Yamagishi S, Takeuchi M. Glycer-AGEs-RAGE signaling enhances the angiogenic potential of hepatocellular carcinoma by upregulating VEGF expression. World J Gastroenterol. 2012 Apr 21;18(15):1781-8. 294 Yoshida T, Yamagishi S, Nakamura K, Matsui T, Imaizumi T, Takeuchi M, Koga H, Ueno T, Sata M. Telmisartan inhibits AGE-induced C-reactive protein production through downregulation of the receptor for AGE via peroxisome proliferatoractivated receptor-gamma activation. Diabetologia. 2006 Dec;49(12):3094-9. 295 Stevens PE. Anaemia, diabetes and chronic kidney disease: where are we now? J Ren Care. 2012 Feb;38 Suppl 1:67-77. 296 Brennan E, McEvoy C, Sadlier D, Godson C, Martin F. The genetics of diabetic nephropathy. Genes (Basel). 2013 Nov 5;4(4):596-619. 297 Schmid H, Boucherot A, Yasuda Y, Henger A, Brunner B, Eichinger F, Nitsche A, Kiss E, Bleich M, Gröne HJ, Nelson PJ, Schlöndorff D, Cohen CD, Kretzler M; European Renal cDNA Bank (ERCB) Consortium. Modular activation of nuclear factor-kappaB transcriptional programs in human diabetic nephropathy. Diabetes. 2006 Nov;55(11):2993-3003. 298 Forbes JM, Cooper ME. Glycation in diabetic nephropathy. Amino Acids. 2012 Apr;42(4):1185-92. 299 Liang Y, Jiang H, Ratovitski T, Jie C, Nakamura M, Hirschhorn RR, Wang X, Smith WW, Hai T, Poirier MA, Ross CA. ATF3 plays a protective role against toxicity by Nterminal fragment of mutant huntingtin in stable PC12 cell line. Brain Res. 2009 Aug 25;1286:221-9. 300 Chen BP, Liang G, Whelan J, Hai T. ATF3 and ATF3 delta Zip. Transcriptional repression versus activation by alternatively spliced isoforms. J Biol Chem. 1994 Jun 3;269(22):15819-26. 301 Thompson MR, Xu D, Williams BR. ATF3 transcription factor and its emerging roles in immunity and cancer. J Mol Med (Berl). 2009 Nov;87(11):1053-60. 302 Stephen J McConoughey, Chris C Wolford, Tsonwin Hai. Atf3. UCSD Molecule Pages. Jun 2011, doi:10.6072/H0.MP.A003217.01 303 Chateauvieux S, Grigorakaki C, Morceau F, Dicato M, Diederich M. Erythropoietin, erythropoiesis and beyond. Biochem Pharmacol. 2011;82(10):1291-1303. 304 Chong ZZ, Shang YC, Mu Y, Cui S, Yao Q, Maiese K. Targeting erythropoietin for chronic neurodegenerative diseases. Expert Opin Ther Targets. 2013;17(6):707720. 305 Shang YC, Chong ZZ, Wang S, Maiese K. Prevention of β-amyloid degeneration of microglia by erythropoietin depends on Wnt1, the PI 3-K/mTOR pathway, Bad, and Bcl-xL. Aging (Albany NY). 2012;4(3):187-201. References 123 306 Arabpoor Z, Hamidi G, Rashidi B, Shabrang M, Alaei H, Sharifi MR, Salami M, Dolatabadi HR, Reisi P. Erythropoietin improves neuronal proliferation in dentate gyrus of hippocampal formation in an animal model of Alzheimer's disease. Adv Biomed Res. 2012;1:50. 307 Jia Y, Mo SJ, Feng QQ, Zhan ML, Ouyang LS, Chen JC, Ma YX, Wu JJ, Lei WL. EPO-Dependent Activation of PI3K/Akt/FoxO3a Signalling Mediates Neuroprotection in In Vitro and In Vivo Models of Parkinson's Disease. J Mol Neurosci. 2014;53(1):117-124. 308 Merelli A, Czornyj L, Lazarowski A. Erythropoietin: a neuroprotective agent in cerebral hypoxia, neurodegeneration, and epilepsy. Curr Pharm Des. 2013;19(38):6791-6801. 309 Najmi Varzaneh F, Najmi Varzaneh F, Azimi AR, Rezaei N, Sahraian MA. Efficacy of combination therapy with erythropoietin and methylprednisolone in clinical recovery of severe relapse in multiple sclerosis. Acta Neurol Belg. 2014 Mar 7. 310 Cervellini I, Ghezzi P, Mengozzi M. Therapeutic efficacy of erythropoietin in experimental autoimmune encephalomyelitis in mice, a model of multiple sclerosis. Methods Mol Biol. 2013;982:163-173. 311 Nagańska E, Taraszewska A, Matyja E, Grieb P, Rafałowska J. Neuroprotective effect of erythropoietin in amyotrophic lateral sclerosis (ALS) model in vitro. Ultrastructural study. Folia Neuropathol. 2010;48(1):35-44. 312 Ma R, Huang C, Hu J, Wang M, Xiang J, Li G. JAK2/STAT5/Bcl-xL signaling is essential for erythropoietin-mediated protection against PC12 cells apoptosis induced by Aβ25-35. Br J Pharmacol. 2014 Mar 6. 313 Ma R, Xiong N, Huang C, Tang Q, Hu B, Xiang J, Li G. Erythropoietin protects PC12 cells from beta-amyloid(25-35)-induced apoptosis via PI3K/Akt signaling pathway. Neuropharmacology. 2009;56(6-7):1027-1034. 314 Wenker SD, Chamorro ME, Vittori DC, Nesse AB. Protective action of erythropoietin on neuronal damage induced by activated microglia. FEBS J. 2013;280(7):16301642. Annex Papers published by the author of this thesis during the PhD, that have not been used on the results section, but are in the context of theme developed here Glaucoma is a progressive optic nerve neuropathy and the major cause of preventable and irreversible blindness worldwide. It is characterized by visual field defects and nerve head cupping due to the loss of retinal ganglion cells [1]. Despite its multifactorial genesis [2-4], the major risk factor for glaucoma progression is the elevated intraocular pressure (IOP) [5,6], which compresses the retinal ganglion cells at the optic nerve head [7]. The only treatment that slows glaucoma progression involves lowering the IOP [8]. Familial amyloid polyneuropathy (FAP) is caused by the extracellular deposition of amyloid fibrils of mutant transthyretin (TTR) V30M in various tissues and organs [9-11]. TTR V30M mutation is the most common form of transthyretin amyloidosis (ATTR) variant in Portugal as well as in the world [12]. The main clinical expression of FAP disease is a sensorimotor and autonomic neuropathy, but other manifestations, such as nephropathy and hematologic and ocular abnormalities can occur. Among the reported ocular FAP complications [13-15], glaucoma is the major cause of irreversible vision loss and is often difficult to control [16]. Erythropoietin (EPO) was identified as a hematopoietic cytokine that promotes proerythroblast survival and maturation [17]. Recently, EPO was recognized as a member of the cytokine type 1 superfamily with multiple functions outside the bone marrow [18]. It provides direct protection against hypoxia by its anti-apoptotic, anti-oxidative, and anti-inflammatory properties and for its angiogenic capacity that allows the oxygen supply to ischemic tissues. Several studies have found that EPO protects photoreceptor cells, retinal ganglion cells, and retinal pigment epithelial cells from apoptosis [19-26]. Hernandez et al. [27] suggested that EPO is produced locally in the retina. Muller cells and retinal pigment epithelium were identified by Fu et al. [28] and Garcia-Ramírez et al. [29], respectively, as the cells responsible for EPO production in the eye. Previous studies have shown a significantly increased EPO concentration in the aqueous humor of eyes with glaucoma [30-32]; this is probably a defence mechanism against Molecular Vision 2014; 20:970-976 <http://www.molvis.org/molvis/v20/970> Received 12 December 2013 | Accepted 30 June 2014 | Published 2 July 2014 © 2014 Molecular Vision 970 Aqueous humor erythropoietin levels in open-angle glaucoma patients with and without TTR V30M familial amyloid polyneuropathy João M. Beirão,1,2,3 Luciana M. Moreira,3,4 João C. Oliveira,5 Maria J. Menéres,1,3 Bernardete B. Pessoa,1 Maria E. Matos,3 Paulo P. Costa,3,4 Paulo A. Torres,1,3 Idalina B. Beirão2,3 1Ophthalmology, Hospital de Santo António, Porto; 2Unidade Clínica de Paramiloidose, Hospital de Santo António, Porto; 3UMIB, ICBAS, Instituto de Ciências Biomédicas Abel Salazar, Porto; 4INSA Dr. Ricardo Jorge, Porto; 5Biochemistry Service, Hospital Santo António, Porto Purpose: Glaucoma is the leading cause of irreversible blindness in familial amyloidotic polyneuropathy (FAP) patients. Erythropoietin (EPO) is a cytokine that has been shown to play a role in neuroprotection and is endogenously produced in the eye. EPO levels in the aqueous humor are increased in eyes with glaucoma. In this study, we evaluated the EPO concentration in the aqueous humor of FAP and non-FAP patients, with and without glaucoma. Methods: Undiluted aqueous humor samples were obtained from 42 eyes that underwent glaucoma surgery, phacoemulsification, or vitrectomy. EPO concentration in the aqueous humor and blood were measured using the Immulite 2000 Xpi using an automatic analyzer (Siemens Healthcare Diagnostics). Results: The mean EPO concentration in the aqueous humor of non-FAP glaucoma eyes group 2 (75.73±13.25 mU/ml) was significantly higher than non-FAP cataract eyes (17.22±5.33 mU/ml; p<0.001), FAP glaucoma eyes (18.82±10.16 mU/ml; p<0.001), and FAP nonglaucoma eyes (20.62±6.22 mU/ml; p<0.001). There was no statistically significant difference between FAP nonglaucoma eyes versus non-FAP cataract eyes (p = 0.23) and FAP glaucoma eyes versus FAP nonglaucoma eyes (p = 0.29). In the glaucoma groups, there was no correlation between the aqueous humor EPO concentration and the ocular pressure (p = 0.95) and mean deviation (p = 0.41). There was no correlation between the EPO serum concentration and EPO aqueous humor concentration in our patients (p = 0.77). Conclusions: Unlike other glaucomatous patients, FAP patients with glaucoma do not show increased and potentially neuroprotective endocular EPO production in the aqueous humor and may need more aggressive glaucoma management. Correspondence to: Melo Beirão, Serviço de Oftalmologia, Hospital de Santo António, Largo Prof. Abel Salazar, 2, 4099-001 Porto, Portugal; Phone: +351966009826; FAX: + 351 22 606 61 06; email: [email protected] Molecular Vision 2014; 20:970-976 <http://www.molvis.org/molvis/v20/970>© 2014 Molecular Vision 971 glaucomatous damage [33] caused by hypoxia, ischemia, oxidative stress, and reduced pro-inflammatory cytokine production [34-39]. Although hypoxia/ischemia is the major stimulus for endocular and systemic EPO production [21,4043], other incompletely understood factors may be involved [27,29]. FAP patients and even presymptomatic carriers have an inappropriately low EPO production [44]. In vitro studies suggest that the dissociated mutant TTR that polymerizes into misfolding amyloidogenic intermediates, protofilaments, and nonfibrillar aggregates of TTR rather than mature amyloid fibrils may induce cellular toxicity [45,46]. We propose that these amyloid precursors may be toxic to EPO-producing cells. This study was performed to evaluate the ocular EPO response in FAP patients with glaucoma. METHODS It was recruited 42 eyes of 42 patients (18 females) with a mean age of 56.8±7.4 years. A prospective, controlled, nonrandomized, nonblind comparative study was conducted from January 2008 to December 2011 at the Ophthalmic and Clinical Chemistry Departments from Centro Hospitalar do Porto, Porto. Written informed consent was obtained from all patients. This study was performed in accordance with the Declaration of Helsinki of the World Medical Association and was approved by the Ethics Committee of the Centro Hospitalar do Porto. Presurgical assessment included Snellen best-corrected visual acuity (Snellen chart, Takagi chart projector CP-30, calibrated for approximately 6 m), slit-lamp biomicroscopy, intraocular pressure (IOP) measurement by Goldman applanation tonometry (same person with AT-900 tonometer; Haag-Streit, Koniz, Switzerland), fundoscopy (90 D noncontact slit-lamp lens; Volk Optical, Mentor, OH), Humphrey perimetry (Humphrey Field Analyzer; Humphrey Instruments, San Leandro, CA), and the cup/disc ratio. All examinations were performed within 2 weeks before the surgical procedure. Exclusion criteria for all groups were: previous laser and/or intraocular surgery; history of systemic (e.g., diabetes mellitus, kidney disease, cardiovascular disorders, anemia, immune disease, except FAP in groups 1 and 3) or any ocular disorders (e.g., age-related macular degeneration); history of medications that could influence the level of EPO (e.g., iron preparations, chemotherapeutic agents, granulocyte colony-stimulating factor, or systemic therapy with EPO), and patients with any type of glaucoma except open-angle glaucoma, such as angle-closure, pigmented, exfoliation, normotensive, and neovascular glaucomas, or ocular hypertension. To clarify the relationship between aqueous EPO production and circulating blood EPO levels, we compared the aqueous and serum concentrations of EPO. Aqueous humor samples were obtained from each eye before the beginning of surgery (trabeculectomy, phacoemulsification, or vitrectomy). The standard procedure involved collecting undiluted aqueous humor samples (50–150 µl) through a paracentesis, using a 30-gauge needle on a tuberculin syringe under an operating microscope. Samples were obtained carefully to avoid touching intraocular tissues or blood contamination. All samples were carefully protected from light and were sent immediately to the laboratory for EPO measurement. At the same time, 9 ml of venous blood samples were collected in EDTA tubes from an antecubital vein immediately before sugery. The blood was immediately centrifuged and the blood serum put on the automatic analyzer. Serum samples were obtained from the centrifugation of the blood sample. The samples of aqueous humor and serum had the same processing routine analysis. Serum and aqueous humor EPO concentrations were measured by a chemiluminescent method in an automatic Xpi Immulite 2000 analyzer (Siemens Healthcare Diagnostics, Siemens AG, Munich, Germany). Statistical analysis: Statistical analysis was performed using nonparametric tests. The Kruskal–Wallis test was used to compare the groups in relation to age, and the chi-square test was used in relation to gender. The Mann–Whitney U test was used to compare the nonglaucoma, glaucoma, and FAP groups in relation to aqueous humor EPO and serum EPO levels. The relation between EPO and serum was evaluated by Spearman correlation. Values of p<0.05 were considered statistically significant. Data analysis was performed using IBM SPSS Statistics software version 20. RESULTS A total of 21 glaucomatous eyes from 21 patients and 21 control eyes (21 patients) were enrolled in the study. The demographic characteristics of the patients are summarized in Table 1. Of the glaucoma eyes, ten were from FAP patients (group 1, mean age 55.4±10.0 years mean and standard deviation; five females) and 11 were from non-FAP patients (group 2, mean age 55.8±7.0 years mean and standard deviation; four females). Of the 21 control eyes, nine were from FAP patients with an indication for vitrectomy due to amyloid deposition (group 3, 55.9±8.5 years mean and standard deviation; four females) and 12 were from non-FAP patients awaiting phacoemulsification and intraocular lens implantation (group 4, 58.8±4.9 years mean and standard deviation; five females). Groups 1 and 2 presented indications for trabeculectomy, Molecular Vision 2014; 20:970-976 <http://www.molvis.org/molvis/v20/970>© 2014 Molecular Vision 972 had uncontrolled IOP (defined as IOP higher than the target pressure with maximally topical antiglaucoma medications: prostaglandin + beta blocker + anhydrase carbonic inhibitor + alpha-2 agonist), abnormal visual field test results, and abnormal cup/disc ratio. The ages and gender distribution of the patients were similar between groups (Kruskal–Wallis test, p = 0.56; chi-square test, p = 0.94). All FAP patients had received an orthotopic liver transplant. As summarized in Table 2, the mean EPO concentration in the aqueous humor of nonglaucomatous eyes (group 3 versus group 4) was not significantly different between FAP and non-FAP patients (20.62±6.22 mU/ml in group 3 and 17.22±5.33 mU/ml in group 4, p = 0.23) and corresponds presumably to the basal ocular production of EPO. In the presence of glaucoma, EPO concentrations in the aqueous humor showed a significant increase in the non-FAP group (group 2, 75.73±13.25 mU/ml; group 1, 18.82±10.16 mU/ml; p<0.001), and when we compared the non-FAP glaucoma group (group 2) with the nonglaucoma groups (FAP group 3 and non-FAP group 4), a similar finding was observed (p<0.001) (Table 2). In the FAP groups (group 1 and group 3), we observed no significant difference between the mean EPO values of patients with or without uncontrolled glaucoma (p = 0.29). As listed in Table 3, FAP patients with glaucoma (group 1) and non-FAP patients with glaucoma (group 2) were comparable in terms of the IOP (p = 0.39) and mean deviation (p = 0.75). The correlation between the IOP and the aqueous humor EPO was not significant in group 1 (mean IOP 26.20±1.93 mmHg; rs = 0.02, p = 0.95) and group 2 (mean IOP 26.82±1.72 mmHg; rs = 0.27, p = 0.41). There was also no significant correlation between the mean deviation and the aqueous humor EPO in group 1 (rs = –0.48, p = 0.16) or group 2 (rs = –0.07, p = 0.83). Serum EPO levels among patient groups were not significantly different when multiple testing was taken into account (Bonferroni correction). No statistically significant correlation between the values of EPO in the serum and in the aqueous humor was observed in any patient (Spearman correlation coefficient r = 0.047, p = 0.77). Table 1. Demographic of The groups. Age/Sex Group 1 FAP glaucoma Group 2 Non-FAP glaucoma Group 3 FAP non-glaucoma Group 4 Non-FAP non-glaucoma Age (year, mean±SD) 55.4±10.0 55.8±7.0 55.9±8.5 58.8±4.9 Female/Male 5 / 5 4 / 7 4 / 5 5 / 7 Table 2. epo in aqueous humor anD serum of The groups. Aqueous humor/ serum EPO level Group 1 FAP glaucoma Group 2 Non-FAP glaucoma Group 3 FAP non-glaucoma Group 4 Non-FAP non-glaucoma Aqueous humor EPO level (mU/ml) 18.82±10.16 75.73±13.25 20.62±6.22 17.22±5.33 Serum EPO level (mU/ml) 13.44±4.82 9.99±2.84 15.04±5.87 8.73±4.12 Aqueous humor EPO (Mann–Whitney U test): Group 3 versus Group 4 p=0.23 ; Group 1 versus Group 2 p<0.001; Group 1 versus Group 3 p=0.29; Group 2 versus Group 4 p<0.001 Table 3. inTraocular pressure anD mean DeviaTion in glaucoma groups. IOP/mean deviation Group 1 FAP Glaucoma Group 2 Non FAP Glaucoma P value Mann– Whitney test N10 11 IOP, mmHg, mean±SD 26.20±1.93 26.82±1.72 0.39 Mean deviation, dB, mean±SD −8.92±3.30 −8.26±3.63 0.75 Molecular Vision 2014; 20:970-976 <http://www.molvis.org/molvis/v20/970>© 2014 Molecular Vision 973 DISCUSSION Glaucoma is a manifestation of a heterogeneous group of diseases with a very complex and multifactorial pathophysiology [8]. Although hypotensive therapy is today the only possible therapeutic intervention, neuroprotective treatment strategies are emerging as a result of the advances in the comprehension of the pathophysiological mechanisms of glaucoma. In the future, neuroprotective agents will probably be part of the therapeutic arsenal available for the treatment of glaucoma. EPO has been shown to have a protective effect on ganglion cells against acute ischemia injury [28,47] and has been proposed as a potential neuroprotective treatment. In this study we confirmed that the aqueous humor EPO level is higher in glaucomatous eyes than in nonglaucomatous eyes with cataracts, as previously reported [30-32,48,49]. This increase in aqueous humor EPO levels could be a result of local production and/or active transport through the blood– ocular barrier. This observation lends support to the hypothesis that EPO acts as an endogenous neuroprotector of retinal ganglion cells [19]. In spite of the inappropriately low renal EPO production reported in FAP ATTR V30M [44], its basal level in the aqueous humor of FAP patients was not significantly altered. However, FAP patients seemed to be unable to increase endocular EPO production in the presence of glaucoma. In previous studies, we showed an inappropriate secretion of renal EPO in FAP and an inability to increase EPO production in response to decreased serum hemoglobin levels, leading to a high incidence of anemia in these patients. The lack of response to glaucoma in FAP patients could be the ocular counterpart of the stunted renal EPO production in FAP in response to anemia. It has been suggested that inhibition of EPO production could be caused by the toxicity of prefibrillar aggregates of TTR V30M [44,50,51]. These oligomers induce the expression of oxidative stress, pro-inflammatory cytokines, and apoptosis-related molecules [52,53] through the binding of TTR aggregates to the receptor for advanced glycation end products, activation of extracellular signal-regulated kinase cascades, and nuclear transcription factor kB [52-56], suppressing the EPO production. All our FAP patients had previously received an orthotopic liver transplant to eliminate their main source of mutant TTR, their own liver [57]. After liver transplantation, mutant TTR is removed from systemic circulation; however, its local production in the eye remains presumably unaffected. Therefore, the ocular pathology related to FAP, which includes glaucoma, continues to progress after liver transplantation; presumably there is also continuing deposition of cytotoxic prefibrillar TTR aggregates. Garcia-Ramirez found that other factors besides hypoxiainducible factors (HIF)-mediated hypoxia might be important in the upregulation of EPO. Hypoxia, ischemia, elevated reactive oxygen species, or increases in glutamate and nitric oxide caused by glaucomatous damage are probably the cause of elevated aqueous humor EPO concentration in chronic glaucoma [30]. The pro-inflammatory cytokines interleukin (IL)-1, IL-6, interferon-γ, and tumor necrosis factor (TNF)-α inhibit EPO production [58,59], but despite being increased in the aqueous humor of glaucoma eyes, as is especially the case for TNF-α [60], these cytokines do not prevent an increase in EPO levels. Increased levels of TTR in the aqueous humor of glaucoma patients have been documented [61-63]. If glaucoma leads to an increase expression of TTR in the aqueous humor, an increased concentration of the unstable TTR V30M in FAP patients’ eyes could contribute to the increased development of a mechanical barrier to the outflow of the aqueous humor [64], resulting in worsening the glaucoma. The association of open-angle glaucoma with autonomic nervous system dysfunction suggests that this could also play a role in the pathogenesis of the disease [65]. Patients with systemic sympathetic and parasympathetic neuropathies have a higher incidence of open-angle and normal-pressure glaucoma [66-69]. Because FAP patients have an early onset neuropathy with markedly autonomic involvement, it is likely that autonomic dysfunction plays a role in glaucoma pathophysiology. Other possible contributing factors are the hemodynamic instability often presented in FAP patients due to vascular deregulation and abnormal blood pressure that may compound the harmful effects of glaucoma, particularly during sleep [65]. In the groups with glaucoma, there was no correlation between the aqueous humor EPO concentration and the values of IOP and mean deviation. It seems that the concentration of EPO in the aqueous humor is not related to the IOP in eyes with glaucoma or previous eye injury caused by glaucoma. In this study, patients with pseudoexfoliative and uveitic glaucomas were excluded because some studies pointed to blood–aqueous humor barrier breakdown in these situations [70,71]. EPO can cross the blood–brain barrier and blood– retina barrier [41]. We did not found a significant correlation between aqueous humor and serum EPO concentrations as other authors have found [30,31]. The elevation of the aqueous humor EPO level in glaucoma was not associated with a parallel increase in blood EPO levels, corroborating the role Molecular Vision 2014; 20:970-976 <http://www.molvis.org/molvis/v20/970>© 2014 Molecular Vision 974 of local EPO production as already proposed by Fu [28] and Garcia-Ramirez [29]. In conclusion, our study confirmed that the level of EPO is increased in aqueous humor of open-angle glaucomatous eyes, as found by other authors. This increase was not observed in FAP patients. With the increased survival of transplanted FAP patients, glaucoma prevalence is expected to increase dramatically with increased suvival of the transplanted patients. We showed lower endogenous neuroprotection in glaucomatous eyes of FAP patients, emphasizing the need for more aggressive glaucoma management to maintain vision through life. REFERENCES 1. Agar A, Yip SS, Hill MA, Coroneo MT. Pressure related apoptosis in neuronal cell lines. J Neurosci Res 2000; 60:495503. [PMID: 10797552]. 2. Flammer J, Orgül S. Optic nerve blood-flow abnormalities in glaucoma. Prog Retin Eye Res 1998; 17:267-89. [PMID: 9695795]. 3. Cioffi GA. Ischemic model of optic nerve injury. Trans Am Ophthalmol Soc 2005; 103:592-613. [PMID: 17057819]. 4. Tezel G, Wax MB. Hypoxia-inducible factor 1alpha in the glaucomatous retina and optic nerve head. Arch Ophthalmol 2004; 122:1348-56. [PMID: 15364715]. 5. Evans DW, Hosking SL, Gherghel D, Bartlett JD. Contrast sensitivity improves after brimonidine therapy in primary open angle glaucoma: a case for neuroprotection. Br J Ophthalmol 2003; 87:1463-5. [PMID: 14660453]. 6. Gordon MO, Beiser JA, Brandt JD, Heuer DK, Higginbotham EJ, Johnson CA, Keltner JL, Miller JP, Parrish RK 2nd, Wilson MR, Kass MA. The Ocular Hypertension Treatment Study: baseline factors that predict the onset of primary open-angle glaucoma. Arch Ophthalmol 2002; 120:714-20. [PMID: 12049575]. 7. Izzotti A, Bagnis A, Saccà SC. The role of oxidative stress in glaucoma. Mutat Res 2006; 612:105-14. [PMID: 16413223]. 8. Leske MC, Heijl A, Hussein M, Bengtsson B, Hyman L, Komaroff E. Early Manifest Glaucoma Trial Group. Factors for glaucoma progression and the effect of treatment: the early manifest glaucoma trial. Arch Ophthalmol 2003; 121:48-56. [PMID: 12523884]. 9. Saraiva MJ, Costa PP, Goodman DS. Studies on plasma transthyretin (prealbumin) in familial amyloidotic polyneuropathy, Portuguese type. J Lab Clin Med 1983; 102:590603. [PMID: 6311926]. 10. Saraiva MJ, Costa PP, Birken S, Goodman DS. Presence of an abnormal transthyretin (prealbumin) in Portuguese patients with familial amyloidotic polyneuropathy. Trans Assoc Am Physicians 1983; 96:261-70. [PMID: 6208668]. 11. Saraiva MJ, Birken S, Costa PP, Goodman DS. Amyloid fibril protein in familial amyloidotic polyneuropathy, Portuguese type. Definition of molecular abnormality in transthyretin (prealbumin). J Clin Invest 1984; 74:104-19. [PMID: 6736244]. 12. Ando Y, Araki S, Ando M. Transthyretin and familial amyloidotic polyneuropathy. Intern Med 1993; 32:920-2. [PMID: 8204970]. 13. Ando E, Ando Y, Okamura R, Uchino M, Ando M, Negi A. Ocular manifestations of familial amyloidotic polyneuropathy type I: long-term follow up. Br J Ophthalmol 1997; 81:295-8. [PMID: 9215058]. 14. Sandgren O, Kjellgren D, Suhr OB. Ocular manifestations in liver transplant recipients with familial amyloid polyneuropathy. Acta Ophthalmol (Copenh) 2008; 86:520-4. [PMID: 18435819]. 15. Beirão M, Matos E, Beirão I, Costa PP, Torres P. Anticipation of presbyopia in Portuguese familial amyloidosis ATTR V30M. Amyloid 2011; 18:92-7. [PMID: 21591979]. 16. Kimura A, Ando E, Fukushima M, Koga T, Hirata A, Arimura K, Ando Y, Negi A, Tanihara H. Secondary glaucoma in patients with familial amyloidotic polyneuropathy. Arch Ophthalmol 2003; 121:351-6. [PMID: 12617705]. 17. Jelkmann W. Erythropoietin: structure, control of production, and function. Physiol Rev 1992; 72:449-89. [PMID: 1557429]. 18. Ghezzi P, Brines M. Erythropoietin as an antiapoptotic, tissueprotective cytokine. Cell Death Differ 2004; 11:Suppl 1S3744. [PMID: 15243580]. 19. Becerra SP, Amaral J. Erythropoietin - an endogenous retinal survival factor. N Engl J Med 2002; 347:1968-70. [PMID: 12477950]. 20. Dreixler JC, Hagevik S, Hemmert JW, Shaikh AR, Rosenbaum DM, Roth S. Involvement of erythropoietin in retinal ischemic preconditioning. Anesthesiology 2009; 110:774-80. [PMID: 19322943]. 21. Inomata Y, Hirata A, Takahashi E, Kawaji T, Fukushima M, Tanihara H. Elevated erythropoietin in vitreous with ischemic retinal diseases. Neuroreport 2004; 15:877-9. [PMID: 15073535]. 22. Sakanaka M, Wen TC, Matsuda S, Masuda S, Morishita E, Nagao M, Sasaki R. In vivo evidence that erythropoietin protects neurons from ischemic damage. Proc Natl Acad Sci USA 1998; 95:4635-40. [PMID: 9539790]. 23. Tsai JC, Song BJ, Wu L, Forbes M. Erythropoietin: a candidate neuroprotective agent in the treatment of glaucoma. J Glaucoma 2007; 16:567-71. [PMID: 17873720]. 24. Tsai JC, Wu L, Worgul B, Forbes M, Cao J. Intravitreal administration of erythropoietin and preservation of retinal ganglion cells in an experimental rat model of glaucoma. Curr Eye Res 2005; 30:1025-31. [PMID: 16282136]. 25. Villa P, Bigini P, Mennini T, Agnello D, Laragione T, Cagnotto A, Viviani B, Marinovich M, Cerami A, Coleman TR, Brines M, Ghezzi P. Erythropoietin selectively attenuates