- Email: [email protected]
Pathogenesis of Fabry nephropathy: The pathways leading to fibrosis
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Molecular Genetics and Metabolism journal homepage: www.elsevier.com/locate/ymgme
Regular Article
Pathogenesis of Fabry nephropathy: The pathways leading to fibrosis ⁎
Paula Adriana Rozenfelda, , María de los Angeles Bollab, Pedro Quietoc, Antonio Pisanid, Sandro Feriozzie, Pablo Neumanf, Constanza Bondara a
IIFP, Universidad Nacional de La Plata, CONICET, Facultad de Ciencias Exactas, Departamento de Ciencias Biológicas, 47 y 115 (1900), La Plata, Argentina Servicio de Anatomía Patológica, Hospital San Martin, Calle 1 y 70 (1900), La Plata, Argentina Servicio de Nefrología, Hospital Rodolfo Rossi, Calle 37 N° 193 (1900), La Plata, Argentina d Chair of Nephrology, Department of Public Health, University Federico II of Naples, Italy e Nephrology and Dialysis Unit, Belcolle Hospital, Viterbo, Italy f Servicio de Diálisis y Nefrologia, IPENSA, Calle 59 N°434 (1900), La Plata, Argentina b c
A R T I C LE I N FO
A B S T R A C T
Keywords: Fabry Nephropathy TGF-β Kidney biopsy Fibrosis
Background: Kidney is one of the main target organs in Fabry disease, a lysosomal X-linked genetic disorder. Renal involvement is characterized by proteinuria and progressive chronic kidney disease leading to end-stage renal disease. Pathogenic mechanisms in the progression of renal damage in Fabry disease are not thoroughly known yet. The lysosomal Gb3 deposition is the first step of complex pathological pathways resulting in renal sclerosis/fibrosis. Our hypothesis is that Fabry disease associated cellular alterations in tubular cells induce the production of TGF-β1, which mediate transdifferentiation of renal cells into myofibroblasts resulting in fibrosis of renal tissue. Objectives: The aim of this work is to study the mechanisms leading to fibrosis in kidney from human Fabry patients. Methods: Fifteen renal biopsies from naïve Fabry patients were included. Histological and immunohistochemical analysis was carried out. Results: Positive staining for TGF-β1 was found in tubular epithelial cells in biopsies from Fabry patients. Apoptosis was determined by active caspase 3 staining in tubular and mesangial glomerular cells. Due to TGF-β1 is the main profibrotic cytokine and induces accumulation of myofibroblasts, we performed a study for its marker α-smooth muscle actin (α-SMA). This study revealed expression of α-SMA on pericytes surrounding peritubular capillaries, smooth muscle cells of blood vessels, mesangial cells and periglomerular zone. TGF-β1 is produced mainly by tubular cells in Fabry kidney biopsies probably inducing cellular trans-differentiation of renal cells into myofibroblasts. A positive staining for a marker of myofibroblasts was present, affirming the presence of those profibrotic cells. Conclusions: These results show for the first time that TGF-β1 is expressed in human renal tissue from Fabry patients, and that this profibrotic cytokine is mainly produced by proximal tubular cells. In addition both, peritubular interstitium and glomeruli, presented cells positive for myofibroblasts markers.
1. Introduction Fabry disease is a X-linked genetic lysosomal disorder caused by pathogenic mutations in GLA gene, resulting in deficient alfa galactosidase A activity leading to globotriaosylceramide (Gb3) deposits [1] . One of the main target organs is the kidney. Due to chronic and progressive nature of this disorder, manifestations in the male with the classical form [2], may start in infancy with microalbuminuria, and then become evident with proteinuria. Around the 3rd-4th decade of
life glomerular function starts to be affected, with development of chronic kidney disease (CKD) and ultimately end-stage renal disease (ESRD) within their fifth decade [3]. In the late onset variant of the disease and in the females the clinical features appear later and the rate of progression is slower [2,4]. Enzyme replacement therapy (ERT) is a specific targeted treatment, and it has been available for more than 15 years. The results show that ERT slows renal deterioration, and that the outcome depends on the renal function and proteinuria at baseline of the patients [5,6]. Early intervention is crucial because ERT is less
⁎ Corresponding author at: Departamento de Ciencias Biológicas, Facultad de Ciencias Exactas, Universidad Nacional de La Plata-CONICET (CCT-La Plata), 47 y 115, La Plata 1900, Argentina. E-mail address: [email protected] (P.A. Rozenfeld).
https://doi.org/10.1016/j.ymgme.2019.10.010 Received 27 April 2019; Received in revised form 25 October 2019; Accepted 26 October 2019 1096-7192/ © 2019 Elsevier Inc. All rights reserved.
Please cite this article as: Paula Adriana Rozenfeld, et al., Molecular Genetics and Metabolism, https://doi.org/10.1016/j.ymgme.2019.10.010
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
effective in more advanced disease with irreversible damage [7,8]. Recently, positive results have been also reported with chaperone therapy in patients with amenable mutations [9]. It is well known that histological damage is present in renal biopsies of Fabry patients for many years and usually decades before proteinuria and/or elevated serum creatinine is produced [10]. Light microscopy of renal biopsy specimens shows cellular vacuolization due to intracytoplasmic glycosphingolipid accumulation affecting podocytes, epithelial cells of tubules, endothelial cells of peritubular capillaries and vascular intima. On electron microscopy the deposits appear as typical osmiophilic inclusion bodies in the cytoplasm of all kinds of renal cells and show a characteristic ‘myelin figures’ and ‘zebra´ appearance. These pathological features are also evident in heterozygous females. Deposits occur before the development of renal impairment [11,12]. In overt disease renal biopsies show focal and segmental glomerulosclerosis (FSGS), vascular changes and interstitial fibrosis [13,14]. The pathogenic link between the metabolic abnormality at cell level and tissue injury is still unclear. The initial metabolic derangement may promote the production of secondary mediators of injury that lead to inflammation, parenchymal cell loss, and fibrosis. Different types of damage to renal tissue leads ultimately to CKD. Regardless of the initial insult(s), CKD is characterized by kidney injury responses seen histopathologically as glomerulosclerosis (glomerular fibrosis and associated capillary loop destruction), interstitial fibrosis, tubular atrophy, peritubular capillary rarefaction, and inflammation [15]. Renal fibrosis, characterized by excessive deposition of extracellular matrix (ECM), is recognized as a common pathological feature of CKD which associates to the development of ESRD. Transforming growth factor β (TGF-β1) is known as the most potent profibrogenic cytokine; it can be produced by all types of renal resident cells and regulates cell differentiation, proliferation, ECM production, and apoptosis [16,17]. TGF-β1 directly induces the production of ECM, including upregulation of expression of collagen I and fibronectin [18]. TGF-β1 is believed to play critical roles in the transdifferentiation toward myofibroblast of several types of cells, including epithelial cells, endothelial cells, bone marrow derived-mesenchymal stem cells and pericytes, although the origin of myofibroblast in kidney is still undefined [19,20]. Ferrari et al suggested that TGF-β1 activates fibroblast growth factor 2 (FGF-2) expression in endothelial cells, which then promotes vascular-endothelial growth factor (VEGF) production [21]. VEGF is expressed in glomerular podocytes and tubular epithelial cells. Moreover, these authors suggest apoptosis via caspase activation signaling may be induced by the increased expression of TGF-β1 and VEGF in the kidney in the Fabry disease mouse model [22]. Understanding of Fabry kidney pathology has begun with the use of animal models and in vitro assays. A study using the Fabry mouse model, GLA gene knockout, revealed higher thrombospondin 1, VEGF and TGF-β1 expression levels in homogenate of kidneys from Fabry mice when compared to wild-type littermates [22]. In vitro assays using normal podocytes, tubular and mesangial cells exposed to Gb3 or LysoGb3 showed the production of TGF-β1 by these cells, and the consequent deposition of ECM proteins [23–25]. To the best of our knowledge, the pathogenic mechanisms leading to renal damage in human specimens from Fabry patients have not yet been revealed. Our hypothesis is that Fabry disease associated cellular alterations in tubular cells induce the production of TGF-β1, which mediates transdifferentiation of renal cells into myofibroblasts. These myofibroblasts secrete extracellular matrix proteins resulting in fibrosis of renal tissue. TGF-β1 could also mediate apoptosis of renal cells through the induction of FGF-2, VEGF and caspase-3.
Table 1 Fabry patients: demographic and clinical data. Proteinuria is expressed as mg/ 24hs. GFR was determined by creatinine clearance (ml/min) (marked with &) or calculated by CKD-EPI formulae (marked with #) from serum creatinine (ml/ min/1.73 m2). #
Gender
Mutation
Classic or Variant
Age at diagnosis
Age at biopsy
Proteinuria
GFR
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
M M F F F M F F M M M F F M M
c.618_619delTT c.1079G > A c.1079G > A c.1066C > T c.1066C > T c.1066C > T c.1066C > T c.667 T > G c.1066C > T c.1066C > T c.1066C > T c.658C > T c.520 T > G c.1024C > T c.145C > G
C V V C C C C V C C C C V C C
23 53 52 43 50 33 34 58 45 28 48 26 54 35 41
23 53 54 43 51 33 34 64 46 28 48 27 54 35 41
800 500 800 200 350 200 250 200 270 3000 2500 160 190 1270 2470
105& 68& 76& 115# 80# 103# 149# 66# 96# 52# 46# 110# 197# 24# 75#
Table 2 Non-Fabry individuals: demographic and clinical data. #
Gender
Age at biopsy
Diagnosis/renal diagnosis
1
M
55
2 3 4 5
F M F M
35 32 67 68
Clear cell renal cell carcinoma - Biopsy was taken from normal tissue Focal segmental glomerulosclerosis Membranous glomerulonephritis Focal segmental glomerulosclerosis Diabetes - Focal segmental glomerulosclerosis
Table 3 Renal biopsies: histological findings. Fabry
Number of total glomeruli
Global GS
FSGS
Inflammatory infiltration
Interstitial fibrosis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
16 20 8 16 17 6 2 5 25 13 20 22 36 10 25
0 0 0 0 0 0 0 1 1 5 1 0 1 1 3
− − − − − − − − − − − − − + +
mild no no no no no no no no moderate mild mild mild moderate moderate
moderate no no mild mild mild no moderate moderate moderate mild mild mild moderate moderate
0 17 2 1 2
− + − + +
no moderate moderate mild moderate
no moderate moderate mild moderate
Non-Fabry 1 70 2 31 3 7 4 9 5 12
The table displays the number of total glomeruli, the number of global glomerular sclerosis (global GS), presence (+) or absence (−) of focal and segmentary glomerular sclerosis (FSGS), the score of inflammatory infiltration (no, mild or moderate) and the score of interstitial inflammation (no, mild or moderate) from Fabry and non-Fabry biopsies.
2. Materials and methods 2.1. Patients and samples A total of 15 renal biopsies from treatment naïve Fabry patients (7 2
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
Fig. 1. Histological analysis of kidney biopsies. 1A and 1B) Representative images of H&E staining of kidney biopsies from Fabry patients showing vacuolization of mesangial cells and podocytes. Picture B is a higher magnification, where the black arrows point to mesangial cells and the blue arrow point to podocyte cell. C) Mallory trichromic reaction in a Fabry biopsy showing mild interstitial fibrosis. D) H&E staining of a Fabry biopsy showing global glomerular sclerosis and interstitial inflammatory lymphocyte infiltration. E) Representative image of H&E staining of kidney biopsies from a Fabry patient showing thickening of interstitial sector. F) Representative image of H&E staining of kidney biopsies from a Fabry patient showing thickening of the arteriolar walls (yellow arrow). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
experienced nephropathologist; in addition, interstitial fibrosis (using Mallory's trichrome reaction) and inflammatory infiltration (using hematoxylin-eosin staining) were scored by light microscopy in mild or moderate when present.
female and 8 male) were included in this study. Patients' age ranged from 23 to 64 years (median age 43). Patient demographic and clinical data are shown in Table 1. Renal biopsies from five non Fabry patients were included as controls (Table 2). Ages from these individuals ranged from 32 to 68 (median age 55). The procedures followed were in accordance with the ethical standards of the Ethical Committee of Framingham (La Plata, Argentina) and with the Helsinki Declaration of 1975, as revised in 2013. All patients provided written informed consent to participate in this study.
2.3. Immunohistochemistry for Gb3 (CD77) Paraffin-embedded kidney biopsies from Fabry patients and nonFabry controls were used. Sections of 5 μm were rehydrated and heatinduced antigen retrieval was performed. Non-specific binding was blocked with 5% bovine serum albumin (Sigma) in PBS for 1 h at room temperature. The specimens were then incubated with a monoclonal antiGb3/CD77 1:500 (Becton Dickinson, Franklin Lakes, NJ, USA) for at least 1 h at room temperature. For detection, HRP-conjugated goat antihuman Ig antibody was used.
2.2. Renal biopsies for histological investigations We received twenty renal biopsies as paraffin blocks for histological study. Serial sections (3 μm) were stained with hematoxylin-eosin and Mallory's trichrome reaction. Glomerular sclerosis was analyzed by an 3
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
Fig. 2. Gb3 immunohistochemistry of kidney biopsies. Representative images of kidney biopsies from a Fabry patient showing Gb3 deposits in epithelial tubular cells, epithelial Bowman cells, endothelial glomerular cells, podocytes and mesangial cells (C) and from a non-Fabry individual with FSGS showing mild Gb3 expression at tubular epithelial cells (D).
sclerosis was present in 7 of 15 Fabry patients (46,7%) (Fig. 1D). The presence of interstitial fibrosis was identified in 12 of 15 (80%) kidney biopsies assayed from Fabry individuals. Interstitial fibrosis was scored as mild in 6 patients, and as moderate in the other 6 patients (Fig. 1C and E). At interstitial level, inflammatory infiltration was found in 7 of 15 patients (46,7%) (Fig. 1D). This infiltration was scored as mild in 26,7% and moderate in 20% of them (Table 3). Sixty seven percent of patients had hyaline thickening of arteriolar wall (Fig. 1F).
2.4. Immunofluorescence confocal microscopy Paraffin-embedded kidney biopsies from Fabry patients and nonFabry controls were used. Sections of 5 μm were rehydrated and heatinduced antigen retrieval was performed. Sections were blocked with goat serum and overnight incubations were performed using the following primary antibodies, all from Abcam: mouse anti-human FGF2 (cat. ab181), rabbit anti-human active caspase 3 (cat. ab49822), rabbit anti-human TGF-β1 (cat. ab66043), rabbit anti-human VEGFA (cat. ab52917) or rabbit anti-human SMA (cat. ab5694) and mouse antihuman CD31 (cat. ab9498). For detection, secondary antibodies were incubated for one hour. FGF2 staining was performed with Alexa Fluor 488-conjugated goat anti-mouse IgG antibody (Invitrogen, cat A11020). For active caspase 3, TGF-β1 and VEGFA staining, Alexa Fluor 488conjugated goat anti-rabbit IgG (Invitrogen, cat A11008) antibody was used. In the double staining protocol, Alexa Fluor 647-conjugated goat anti-rabbit IgG (Invitrogen, cat A21246) and Alexa Fluor 488-conjugated goat anti-mouse IgG antibodies were used for the detection of SMA and CD31, respectively. The nuclei were stained with propidium iodide (1 mg/ml) for 15 min (Sigma, cat P4170). The samples were mounted using fluorescent mounting medium (DakoCytomation, cat S3023) and visualized in a TCS SP5 Leica confocal microscope. The images were obtained using the Leica LAS AF software.
3.3. Gb3 Immunohistochemical analysis Immunohistochemical staining of Gb3 was performed in kidney biopsies using a monoclonal antibody specific for Gb3. This analysis showed Gb3 expression in epithelial tubular cells, epithelial Bowman cells, endothelial glomerular cells, podocytes and mesangial cells from Fabry samples (Fig. 2A). This study confirmed the Gb3 nature of the deposits causing vacuolization of cells. To compare, we also performed the staining in non-Fabry samples, where mild Gb3 expression could be observed only at the epithelial tubular cells (Fig. 2B), as it has been shown by previous reports [26]. 3.4. Immunofluorescence analysis 3.4.1. -TGF-β1 immunofluorescence Previous reports have shown an increment of deposition of extracellular matrix in mesangium and interstitium in kidney biopsies from Fabry patients. It is known that TGF-β1 is the most potent profibrogenic cytokine and it can be potentially produced by all types of renal resident cells [17,18]. In order to unravel the fibrotic mechanism in Fabry nephropathy, we decided to carry out an immunofluorescence analysis. Positive staining for TGF-β1 was found in proximal tubular epithelial cells in all the biopsies from Fabry patients. Occasionally, there was positive staining for TGF-β1 in the vascular smooth muscle of blood vessels. (3/15). The glomeruli did not show any positivity for TGF-β1 staining. As a control of this assay we stained a biopsy from a diabetic non-Fabry patient, which shows TGF-β1 staining both in tubular epithelium and mesangium. We also analyzed a kidney biopsy taken from a section of undamaged tissue of a patient with renal cancer. This tissue presented low or no level of staining for TGF-β1 (Fig. 3). Similarly, the others non-Fabry controls included in this work showed expression at tubular cells and no expression in glomeruli (not shown).
3. Results 3.1. Demographic and clinical data of Fabry patients Demographic and clinical data of naïve Fabry patients included in the study are shown in Table 1. We included 8 males and 7 females with Fabry disease. Eleven of the patients have classical pathogenic mutations (age range: 23–48 years; median age: 35) while the rest are late onset ones (age range:33–64 years; median age: 52). All Fabry patients have pathological proteinuria (Table 1). Renal function expressed as estimated glomerular filtration rate (CKD-EPI formula) is highly variable, with CKD stages from 1 to 4. Non Fabry controls were also included for comparison (3 Males and 2 females). Non Fabry controls were biopsied because of kidney affection other than Fabry disease (Described in Table 2). 3.2. Histological examinations
3.4.2. -FGF-2 and VEGF immunofluorescence It has been suggested that TGF-β1 activates FGF-2 expression in endothelial cells, which then promotes VEGF production [22]. We detected FGF-2 expression in 75% of the stained biopsies from Fabry patients, in tubular cells, mesangial glomerulus cells and blood vessels.
Histological findings from renal biopsies from Fabry and non-Fabry individuals are detailed in Table 3. Histopathological examination revealed enlarged and vacuolated podocytes and mesangial cells in all samples from Fabry patients (Fig. 1A and B). Global glomerular 4
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
Fig. 3. TGF-β1 expression in renal tissue. TGF-β1 analysis was performed by immunofluorescence staining and confocal microscopy in kidney biopsies. Nuclei are shown in red while TGF- β is shown in green. Representative images of kidney biopsies from Fabry patients (A-C) and non-Fabry individuals (D-F) are shown. A) Fabry patients showed positive TGF-β1 staining in proximal tubular cells but not in glomeruli. B) In addition to renal tubules, Fabry patients occasionally presented TGF-β1 expression in the vascular smooth muscle of blood vessels. C) Higher magnification showing lack of expression of TGF- β in Fabry glomerulus. D) Undamaged tissue from an individual with renal cancer showed no expression for TGF- β at the glomerulus and low level of expression in tubular cells. E and F) TGF- β staining was observed in the glomeruli and tubules from a diabetic patient presenting FSGS. Staining was stronger in tubular cells. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
5
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
Fig. 4. FGF2 and VEGFA expression in renal tissue. The analysis was performed by immunofluorescence staining and confocal microscopy in kidney biopsies from Fabry and control individuals. Nuclei are shown in red while FGF2 or VEGFA are shown in green. Representative images are shown. A) FGF2 was found in tubular, mesangial and endothelial cells from most of kidney biopsies taken from Fabry patients. B) Similarly, the expression of FGF2 was found in the whole tissue section obtained from a non-Fabry patient with FSGS. C) VEGFA was expressed at the glomerular level in Fabry patients. D) A non-Fabry individual showing VEGFA expression in the glomeruli and blood vessels. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
control individuals tested in this work, showing a similar pattern of expression (tubules and glomeruli) (Fig. 5).
We also evaluated FGF-2 expression in a non-Fabry patient with FSGS and we found a similar pattern of expression, where the glomerulus, the tubular cells and blood vessels showed to express this protein (Fig. 4A and B). By contrast, we only found FGF-2expression in tubule from undamaged tissue from an individual with renal cancer (data not shown). Positive staining for VEGF was detected in glomerulus and blood vessel cells from Fabry kidney biopsies, as well as from a non-Fabry control with FSGS (Fig. 4C and D).
3.6. Detection of myofibroblasts TGF-β1 is the main profibrotic cytokine, it induces accumulation of myofibroblasts which are the primary cells to synthesize and deposit pathological components of fibrillar matrix characteristic of fibrosis. For this reason, we decided to analyze the staining for a marker of myofibroblasts, the αSMA. We also included CD31 staining to evidence endothelial cells. In Fabry kidneys we could detect positive staining for αSMA on pericytes surrounding peritubular capillaries, mesangial cells and periglomerular zone. On the other hand, non-Fabry controls showed different patterns of expression for αSMA. The undamaged tissue taken from a patient with renal cancer showed positive staining only on peritubular capillaries. Samples from non-Fabry FSGS and membranous glomerulonephritis tissues showed αSMA expression on peritubular capillaries, periglomerular zone and, in the last case, also in peritubular space. (Fig. 6).
3.5. Apoptosis determination Overexpression of TGF-β1, FGF2 and VEGF has been associated to increased apoptosis in kidneys of a Fabry mouse model [22]. For this reason, we aimed to analyze a marker of apoptosis in these kidney specimens. Active caspase 3 staining in biopsies included in this study revealed induction of apoptosis in tubules from all the Fabry biopsies and in the mesangial glomerular cells in 80% of them. Besides, a small percentage (18%) of Fabry specimens displayed this marker in periglomerular zone. Active caspase-3 staining was also present in the 6
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
Fig. 5. Active caspase 3 staining. Immunofluorescence assays revealed the expression of active caspase 3 in kidney biopsies. Nuclei are shown in red while active Caspase 3 is shown in green. Representative images are shown. A) All Fabry patients showed the expression of this proapoptotic protein at tubular cells; part of them also presented active caspase 3 positive cells in the glomeruli. B) A similar pattern of expression was observed in a non-Fabry individual presenting FSGS. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
4. Discussion
endothelial cells, which then promotes VEGF production [22]. VEGF is physiologically expressed by podocytes and tubular cells [30]. Overexpression of VEGF, as happens in diabetes, can cause thickening of the glomerular basement membrane, glomerular enlargement and mesangial proliferation in addition to foot process effacement [31]. The expression of FGF-2 and VEGF in Fabry kidney specimens was proved in this work. We found VEGF expression in glomerulus and blood vessel cells, but not in tubules. This expression pattern was also observed in diabetic nephropathy [32]. It was reported that VEGF in conjunction with TGF-β1 induces podocyte apoptosis [33]. Indeed, casapase-3 positive staining was detected in Fabry kidneys from our work. A proapoptotic role for Gb3 and a higher apoptotic state in Fabry PBMCs has been previously reported [34]. Moreover, apoptosis of cardiomyocytes from hearts of Fabry patients was recently reported [35]. In agreement with these observations, our results strongly suggest that renal cells in Fabry patients undergo apoptosis. Fibrosis is a prominent feature in CKD, including Fabry disease. We and others have shown fibrosis in Fabry biopsies. Fibrosis implies pathological deposition of extracellular matrix proteins, by myofibroblasts. Formation and proliferation of myofibroblasts is induced by TGF-β1, the main profibrotic cytokine. In the present work we could detect positive staining for a marker of myofibroblasts, affirming the presence of those profibrotic cells in human Fabry kidneys. There is still some controversy about the origin of myofibroblasts in injured kidneys. At the interstitium, the pericytes attached to peritubular capillaries could transdifferentiate into myofibroblasts. In the glomerulus, mesangial cells, which are a glomerular pericyte, would be the major source of glomerular fibrosis [12]. However, recent fate-tracing studies have implicated resident fibroblasts and/or pericytes as the likely precursor cells to the myofibroblast [36]. In this work, we detected αSMA, a marker of myofibroblasts, in Fabry kidneys. Particularly, positive staining for αSMA was found in pericytes attached to peritubular capillaries of interstitium, mesangial cells and periglomerular zone. This location may explain the presence of fibrosis both in glomerulus (“glomerulosclerosis”) and insterstitium (“tubulointerstitial fibrosis”) in Fabry kidneys. The results of this work may provide additional evidence to explain the renal damage progression in Fabry nephropathy, beside the vision of the podocyte as the main cell affected leading to kidney problems in Fabry. The proximal tubular kidney cell is the one that produces TGF-
Various insults to the kidney are known to lead to chronic kidney disease. Regardless of the etiology, chronic disease is characterized by kidney damage seen histopathologically as glomerulosclerosis, interstitial fibrosis and inflammation [15]. Kidney is one of the target organs affected in Fabry disease and kidney biopsies from patients with overt nephropathy may display the same features of chronic kidney disease of different etiologies. The primary insult in Fabry disease is the glycolipid deposits in almost every glomerular cell type, with massive deposits in podocytes, but also in mesangial, endothelial cells and parietal epithelial cells. Not only those deposits appear in glomerular cells, they are also detected in tubular cells and peritubular capillary endothelium [11,27]. In this work, the classical histological analysis revealed similar results, with all kidney cells affected by glycolipid deposits. Another interesting feature revealed by histological analysis was the prominent interstitial fibrosis, shown in 80% of Fabry kidney biopsies assayed. Glomerular sclerosis was also found although with a minor frequency. Podocytes are reported to be the cell type with the highest score of Gb3 inclusions and vacuolization. In a small study, patients treated with ERT showed clearance of the podocytes and concomitant reduction of podocyte effacement [27]. The massive deposits in podocytes have identified podocytes as the target cell type whose affection could correlate to pathophysiology of chronic kidney disease. Moreover, an in vitro study has shown the production of TGF-β1 and extracellular matrix proteins when normal podocytes are exposed to lyso-Gb3 [23]. Extensive evidence in experimental animal models of both, type 1 and type 2 diabetes, implicates TGF-β1 as an important mediator of diabetic kidney disease, playing a pathophysiological role in promoting glomerulosclerosis, interstitial fibrosis, and the decline in GFR [28]. TGFβ1 mRNA and protein levels are increased in both the glomerular and tubular compartments of various models of experimental diabetes in rats and mice [29]. Considering the above data, we evaluated the production of TGF-β1 in kidney biopsies from Fabry patients in order to identify the pathological mechanism associated with nephropathy and fibrosis in Fabry disease and the main cell types involved. The immunofluorescence analysis carried out in this work revealed that TGF-β1 is produced mainly by proximal tubular cells in all Fabry kidney biopsies studied, but not by glomerular cells, including podocytes. It has been suggested that TGF-β1 activates FGF-2 expression in 7
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
Fig. 6. Myofibroblasts markers in kidney biopsies. Immunofluorescence staining of αSMA (green) and CD31 (red), and confocal microscopy in kidney biopsies was performed. Nuclei are shown in blue. Representative images are shown. Kidney biopsies from Fabry patients (A and B), a non-Fabry patient with renal cancer (D), and a non-Fabry patient with FSGS (E) are shown. C and F shows glomeruli staining with higher magnification for a Fabry patient (C) and a non-Fabry control with membranous glomerulonephritis (F). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
8
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
β1, the main profibrotic cytokine. Therefore, these could be the cells in which the pathological profibrotic changes are initiated. Moreover, the presence of cells with myofibroblast markers in the glomeruli may play a role in producing renal fibrosis. Of course, it does not rule out other pathways stimulated by Gb3 deposition. Podocytes also may release pro-inflamamtory and/or profibrotic cytokines as many experimental papers have demonstrated. In addition it has been reported that glycolipids could induce other cellular mechanisms that could mediate fibrosis. It was shown Gb3 deposition is associated with a reduction in the activity of endothelial nitric oxide synthase resulting in a decreased nitric oxide bioavailability or enzyme uncoupling determining the production of reactive species and an increase of adhesion molecule [37]. Another work showed that lyso-Gb3 deposits stimulate the proliferation of smooth muscle cells with a re-modeling of the intima-media layer of the blood vessels [38]. Both mechanisms could coexist and can cause the release of cytokines with tissue damage in all organs. Moreover, the vascular damage can impair the blood blow to the organs with the following histological damage. A dysregulation of autophagy could also contribute to tissue damage. Gb3 deposition is associated with inhibition of mTOR kinase activity, a negative regulator of the autophagic machinery [39]. Our data suggest that Fabry disease associated cellular alterations, likely Gb3 deposition, is followed by the activation of biological processes resulting in tissue fibrosis. A recent article by Braun et al. has demonstrated in a model of podocyte cell culture, that Gb3 deposition triggers autophagy alterations and stimulate pro-fibrotic processes. Moreover, the addition of recombinant alpha-galactosidase determined a complete clearance of Gb3, however the pathologic pathways continued to be active. This evidence demonstrates that the deposition of Gb3 activates pathogenic biological pathways that can become independent from the Gb3 deposition itself. Recently, new molecules (pegunigalsidasealfa, moss-made agalsidasealfa, lucerastat, venglustat) have been proposed for treatment of Fabry disease and some of them might have a role in modulating fibrosis processes; particularly lucerastat and venglustat that, inhibiting the glucosylceramidesynthetase (GCS), are able to reduce the synthesis of precursors and the levels of the metabolites that accumulate. The reduction of the substrate deposition might contribute to slow down the processes involved in fibrosis pathways [40]. This study has a limitation, associated to the relatively small number of patients included, and the less representation of variant patients as compared to higher number of classical ones; and the inclusion of both, males and females. Therefore, more studies including more biopsies should be carried out in order to validate these results. In conclusion, our results show for the first time that TGF- β1 is expressed in human renal tissue from Fabry patients, and that this profibrotic cytokine is mainly produced by proximal tubular cells. In addition both, peritubular interstitium and glomeruli, presented cells positive for myofibroblasts markers such as αSMA, suggesting a role for pro-fibrotic cells in the pathogenic pathways leading to Fabry nephropathy. On the other hand, active caspase 3 expression at the tubular epithelium suggest that apoptosis of the tubular cell is also involved in tissue damage. Through these mechanisms the pathological process of fibrosis characteristic of Fabry nephropathy would be initiated and propagated.
Research program under IIR-ARG-001540 (Japan) and by Universidad Nacional de La Plata [X679 to PR] (Argentina). None of the funders was involved in the study design, data collection, data analysis, data interpretation, writing of the report, or decision to submit for publication. References [1] R.O. Brady, A.E. Gal, R.M. Bradley, et al., Enzymatic defect in Fabry’s disease Ceramidetrihexosidase deficiency, N. Engl. J. Med. 276 (1967) 1163–1167. [2] M. Arends, C. Wanner, D. Hughes, et al., Characterization of classical and nonclassical Fabry disease: a multicenter study, J. Am. Soc. Nephrol. 28 (2017) 1631–1641. [3] B. Najafian, M. Mauer, R.J. Hopkin, et al., Renal complications of Fabry disease in children, Pediatr. Nephrol. 28 (2013) 679–687. [4] M. Arends, M. Biegstraaten, D.A. Hughes, et al., Retrospective study of long-term outcomes of enzyme replacement therapy in Fabry disease: analysis of prognostic factors, PLoS One 12 (8) (2017 Aug 1) e0182379. [5] M. Beck, D. Hughes, C. Kampmann, et al., Long term outcomes with agalsidase alfa enzyme replacement therapy: analysis using deconstructed composite events, Mol. Genet. Metab. Rep 14 (2018) 31–35. [6] D.P. Germain, J. Charrow, R.J. Desnick, et al., Ten-year outcome of enzyme replacement therapy with agalsidase beta in patients with Fabry disease, J. Med. Genet. 52 (2015) 353–358. [7] M. Beck, D. Hughes, C. Kampmann, et al., Long-term outcomes with agalsidase alfa enzyme replacement therapy: analysis using deconstructed composite events, Mol. Genet. Metab. Rep. 14 (2017 Nov 9) 31–35. [8] S. Waldek, S. Feriozzi, Fabry nephropathy: a review - how can we optimize the management of Fabry nephropathy? BMC Nephrol. 15 (2014) 72. [9] D.P. Germain, K. Nicholls, R. Giugliani, et al., Efficacy of the pharmacologic chaperone migalastat in a subset of male patients with the classic phenotype of Fabry disease and migalastat-amenable variants: data from the phase 3 randomized, multicenter, double-blind clinical trial and extension study, Genet Med. (2019 Feb) 6, https://doi.org/10.1038/s41436-019-0451-z. [10] B. Najafian, E. Svarstad, L. Bostad, et al., Progressive podocyte injury and globotriaosylceramide (GL-3) accumulation in young patients with Fabry disease, Kidney Int. 79 (2011) 663–670. [11] C. Tøndel, L. Bostad, A. Hirth, et al., Renal biopsy findings in children and adolescents with Fabry disease and minimal albuminuria, Am. J. Kidney Dis. 51 (5) (2008) 767–776. [12] A. Sessa, M. Meroni, G. Battini, et al., Evolution of renal pathology in Fabry disease, Acta Paediatr. Suppl. 92 (2003) 6–8 discussion 5. [13] E. Svarstad, L. Bostad, O. Kaarbøe, et al., Focal and segmental glomerular sclerosis (FSGS) in a man and a woman with Fabry’s disease, Clin. Nephrol. 63 (2005) 394–401. [14] A.B. Fogo, L. Bostad, E. Svarstad, et al., Scoring system for renal pathology in Fabry disease: report of the international study group of Fabry Nephropathy (ISGFN), Nephrol. Dial. Transplant. 25 (2010) 2168–2177. [15] G. Campanholle, G. Ligresti, S.A. Gharib, et al., Cellular mechanisms of tissue fibrosis. 3. Novel mechanisms of kidney fibrosis, Am. J. Phys. Cell Phys. 304 (2013) C591–C603. [16] X.-M. Meng, P.M.-K. Tang, J. Li, et al., TGF-β1/Smad signaling in renal fibrosis, Front. Physiol. 6 (2015) 82. [17] B. Sutariya, D. Jhonsa, M.N. Saraf, TGF-β1: the connecting link between nephropathy and fibrosis, Immunopharmacol. Immunotoxicol. 38 (2016) 39–49. [18] R. Samarakoon, J.M. Overstreet, P.J. Higgins, TGF-β1 signaling in tissue fibrosis: redox controls, target genes and therapeutic opportunities, Cell. Signal. 25 (2013) 264–268. [19] C.-F. Wu, W.-C. Chiang, C.-F. Lai, et al., Transforming growth factor β-1 stimulates profibrotic epithelial signaling to activate pericyte-myofibroblast transition in obstructive kidney fibrosis, Am. J. Pathol. 182 (2013) 118–131. [20] Y.B.Y. Sun, X. Qu, G. Caruana, J. Li, The origin of renal fibroblasts/myofibroblasts and the signals that trigger fibrosis, Differentiation. 92 (3) (2016) 102–107. [21] G. Ferrari, G. Pintucci, G. Seghezzi, et al., VEGF, a prosurvival factor, acts in concert with TGF-beta1 to induce endothelial cell apoptosis, Proc. Natl. Acad. Sci. U. S. A. 103 (2006) 17260–17265. [22] M.H. Lee, E.N. Choi, Y.J. Jeon, et al., Possible role of transforming growth factor-β1 and vascular endothelial growth factor in Fabry disease nephropathy, Int. J. Mol. Med. 30 (2012) 1275–1280. [23] M.D. Sanchez-Niño, A.B. Sanz, S. Carrasco, et al., Globotriaosylsphingosine actions on human glomerular podocytes: implications for Fabry nephropathy, Nephrol. Dial. Transplant. 26 (2011) 1797–1802. [24] Y.J. Jeon, N. Jung, J.-W. Park, et al., Epithelial-mesenchymal transition in kidney tubular epithelial cells induced by Globotriaosylsphingosine and Globotriaosylceramide, PLoS One 10 (2015) e0136442. [25] Y.-J. Shin, Y.J. Jeon, N. Jung, et al., Substrate-specific gene expression profiles in different kidney cell types are associated with Fabry disease, Mol. Med. Rep. 12 (2015) 5049–5057. [26] C. Valbuena, D. Leitão, F. Carneiro, et al., Immunohistochemical diagnosis of Fabry nephropathy and localisation of globotriaosylceramide deposits in paraffin-embedded kidney tissue sections, Virchows Arch. 460 (2012) 211–221. [27] C. Tøndel, L. Bostad, K.K. Larsen, et al., Agalsidase benefits renal histology in young patients with Fabry disease, J. Am. Soc. Nephrol. 24 (2013) 137–148. [28] A.S. Chang, C.K. Hathaway, O. Smithies, et al., Transforming growth factor-β1 and diabetic nephropathy, Am. J. Physiol. Ren. Physiol. 201 (310) (2016) F689–F696.
Acknowledgements We acknowledge the physicians who included their patients in the study: Segundo Fernández, Laura Gil, Sebastian Calabrese and Monica Calvo. Funding This project was funded by Shire Human Genetic Therapies Inc., now part of the Takeda group of companies, Investigator Initiated 9
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
[29] F.N. Ziyadeh, Mediators of diabetic renal disease: the case for TGF-Beta as the major mediator, J. Am. Soc. Nephrol. 15 (Suppl. 1) (2004) S55–S57. [30] A. Tufro, D. Veron, VEGF and podocytes in diabetic nephropathy, Semin. Nephrol. 32 (2012) 385–393. [31] D. Veron, K.J. Reidy, C. Bertuccio, et al., Overexpression of VEGF-A in podocytes of adult mice causes glomerular disease, Kidney Int. 77 (2010) 989–999. [32] M.T. Lindenmeyer, M. Kretzler, A. Boucherot, et al., Interstitial vascular rarefaction and reduced VEGF-A expression in human diabetic nephropathy, J. Am. Soc. Nephrol. 18 (2007) 1765–1776. [33] X. Bai, J. Geng, X. Li, F. Yang, J. Tian, VEGF-A inhibition ameliorates podocyte apoptosis via repression of activating protein 1 in diabetes, Am. J. Nephrol. 40 (2014) 523–534. [34] P.N. De Francesco, J.M. Mucci, R. Ceci, et al., Higher apoptotic state in Fabry disease peripheral blood mononuclear cells: effect of globotriaosylceramide, Mol. Genet. Metab. 104 (2011) 319–324. [35] A. Frustaci, C. Chimenti, D. Doheny, et al., Evolution of cardiac pathology in classic
[36] [37] [38] [39] [40]
10
Fabry disease: progressive cardiomyocyte enlargement leads to increased cell death and fibrosis, and correlates with severity of ventricular hypertrophy, Int. J. Cardiol. 248 (2017) 257–262. L.S. Gewin, Renal fibrosis: primacy of the proximal tubule, Matrix Biol. 68–69 (2018) 248–262. L. Shu, A. Vivekanandan-Giri, S. Pennathur, et al., Establishing 3-nitrotyrosine as a biomarker for the vasculopathy of Fabry disease, Kidney Int. 86 (2014) 58–66. J.M. Aerts, J.E. Groener, S. Kuiper, et al., Elevated globotriaosylsphingosine is a hallmark of Fabry disease, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 2812–2817. M.C. Liebau, F. Braun, K. Höpker, et al., Dysregulated autophagy contributes to podocyte damage in Fabry’s disease, PlosOne 8 (2013) e63506. N. Guérard, D. Oder, P. Nordbeck, et al., Lucerastat, an Iminosugar for substrate reduction therapy: tolerability, pharmacodynamics, and pharmacokinetics in patients with Fabry disease on enzyme replacement, Clin. Pharmacol. Ther. 103 (2018) 703–711.
Contents lists available at ScienceDirect
Molecular Genetics and Metabolism journal homepage: www.elsevier.com/locate/ymgme
Regular Article
Pathogenesis of Fabry nephropathy: The pathways leading to fibrosis ⁎
Paula Adriana Rozenfelda, , María de los Angeles Bollab, Pedro Quietoc, Antonio Pisanid, Sandro Feriozzie, Pablo Neumanf, Constanza Bondara a
IIFP, Universidad Nacional de La Plata, CONICET, Facultad de Ciencias Exactas, Departamento de Ciencias Biológicas, 47 y 115 (1900), La Plata, Argentina Servicio de Anatomía Patológica, Hospital San Martin, Calle 1 y 70 (1900), La Plata, Argentina Servicio de Nefrología, Hospital Rodolfo Rossi, Calle 37 N° 193 (1900), La Plata, Argentina d Chair of Nephrology, Department of Public Health, University Federico II of Naples, Italy e Nephrology and Dialysis Unit, Belcolle Hospital, Viterbo, Italy f Servicio de Diálisis y Nefrologia, IPENSA, Calle 59 N°434 (1900), La Plata, Argentina b c
A R T I C LE I N FO
A B S T R A C T
Keywords: Fabry Nephropathy TGF-β Kidney biopsy Fibrosis
Background: Kidney is one of the main target organs in Fabry disease, a lysosomal X-linked genetic disorder. Renal involvement is characterized by proteinuria and progressive chronic kidney disease leading to end-stage renal disease. Pathogenic mechanisms in the progression of renal damage in Fabry disease are not thoroughly known yet. The lysosomal Gb3 deposition is the first step of complex pathological pathways resulting in renal sclerosis/fibrosis. Our hypothesis is that Fabry disease associated cellular alterations in tubular cells induce the production of TGF-β1, which mediate transdifferentiation of renal cells into myofibroblasts resulting in fibrosis of renal tissue. Objectives: The aim of this work is to study the mechanisms leading to fibrosis in kidney from human Fabry patients. Methods: Fifteen renal biopsies from naïve Fabry patients were included. Histological and immunohistochemical analysis was carried out. Results: Positive staining for TGF-β1 was found in tubular epithelial cells in biopsies from Fabry patients. Apoptosis was determined by active caspase 3 staining in tubular and mesangial glomerular cells. Due to TGF-β1 is the main profibrotic cytokine and induces accumulation of myofibroblasts, we performed a study for its marker α-smooth muscle actin (α-SMA). This study revealed expression of α-SMA on pericytes surrounding peritubular capillaries, smooth muscle cells of blood vessels, mesangial cells and periglomerular zone. TGF-β1 is produced mainly by tubular cells in Fabry kidney biopsies probably inducing cellular trans-differentiation of renal cells into myofibroblasts. A positive staining for a marker of myofibroblasts was present, affirming the presence of those profibrotic cells. Conclusions: These results show for the first time that TGF-β1 is expressed in human renal tissue from Fabry patients, and that this profibrotic cytokine is mainly produced by proximal tubular cells. In addition both, peritubular interstitium and glomeruli, presented cells positive for myofibroblasts markers.
1. Introduction Fabry disease is a X-linked genetic lysosomal disorder caused by pathogenic mutations in GLA gene, resulting in deficient alfa galactosidase A activity leading to globotriaosylceramide (Gb3) deposits [1] . One of the main target organs is the kidney. Due to chronic and progressive nature of this disorder, manifestations in the male with the classical form [2], may start in infancy with microalbuminuria, and then become evident with proteinuria. Around the 3rd-4th decade of
life glomerular function starts to be affected, with development of chronic kidney disease (CKD) and ultimately end-stage renal disease (ESRD) within their fifth decade [3]. In the late onset variant of the disease and in the females the clinical features appear later and the rate of progression is slower [2,4]. Enzyme replacement therapy (ERT) is a specific targeted treatment, and it has been available for more than 15 years. The results show that ERT slows renal deterioration, and that the outcome depends on the renal function and proteinuria at baseline of the patients [5,6]. Early intervention is crucial because ERT is less
⁎ Corresponding author at: Departamento de Ciencias Biológicas, Facultad de Ciencias Exactas, Universidad Nacional de La Plata-CONICET (CCT-La Plata), 47 y 115, La Plata 1900, Argentina. E-mail address: [email protected] (P.A. Rozenfeld).
https://doi.org/10.1016/j.ymgme.2019.10.010 Received 27 April 2019; Received in revised form 25 October 2019; Accepted 26 October 2019 1096-7192/ © 2019 Elsevier Inc. All rights reserved.
Please cite this article as: Paula Adriana Rozenfeld, et al., Molecular Genetics and Metabolism, https://doi.org/10.1016/j.ymgme.2019.10.010
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
effective in more advanced disease with irreversible damage [7,8]. Recently, positive results have been also reported with chaperone therapy in patients with amenable mutations [9]. It is well known that histological damage is present in renal biopsies of Fabry patients for many years and usually decades before proteinuria and/or elevated serum creatinine is produced [10]. Light microscopy of renal biopsy specimens shows cellular vacuolization due to intracytoplasmic glycosphingolipid accumulation affecting podocytes, epithelial cells of tubules, endothelial cells of peritubular capillaries and vascular intima. On electron microscopy the deposits appear as typical osmiophilic inclusion bodies in the cytoplasm of all kinds of renal cells and show a characteristic ‘myelin figures’ and ‘zebra´ appearance. These pathological features are also evident in heterozygous females. Deposits occur before the development of renal impairment [11,12]. In overt disease renal biopsies show focal and segmental glomerulosclerosis (FSGS), vascular changes and interstitial fibrosis [13,14]. The pathogenic link between the metabolic abnormality at cell level and tissue injury is still unclear. The initial metabolic derangement may promote the production of secondary mediators of injury that lead to inflammation, parenchymal cell loss, and fibrosis. Different types of damage to renal tissue leads ultimately to CKD. Regardless of the initial insult(s), CKD is characterized by kidney injury responses seen histopathologically as glomerulosclerosis (glomerular fibrosis and associated capillary loop destruction), interstitial fibrosis, tubular atrophy, peritubular capillary rarefaction, and inflammation [15]. Renal fibrosis, characterized by excessive deposition of extracellular matrix (ECM), is recognized as a common pathological feature of CKD which associates to the development of ESRD. Transforming growth factor β (TGF-β1) is known as the most potent profibrogenic cytokine; it can be produced by all types of renal resident cells and regulates cell differentiation, proliferation, ECM production, and apoptosis [16,17]. TGF-β1 directly induces the production of ECM, including upregulation of expression of collagen I and fibronectin [18]. TGF-β1 is believed to play critical roles in the transdifferentiation toward myofibroblast of several types of cells, including epithelial cells, endothelial cells, bone marrow derived-mesenchymal stem cells and pericytes, although the origin of myofibroblast in kidney is still undefined [19,20]. Ferrari et al suggested that TGF-β1 activates fibroblast growth factor 2 (FGF-2) expression in endothelial cells, which then promotes vascular-endothelial growth factor (VEGF) production [21]. VEGF is expressed in glomerular podocytes and tubular epithelial cells. Moreover, these authors suggest apoptosis via caspase activation signaling may be induced by the increased expression of TGF-β1 and VEGF in the kidney in the Fabry disease mouse model [22]. Understanding of Fabry kidney pathology has begun with the use of animal models and in vitro assays. A study using the Fabry mouse model, GLA gene knockout, revealed higher thrombospondin 1, VEGF and TGF-β1 expression levels in homogenate of kidneys from Fabry mice when compared to wild-type littermates [22]. In vitro assays using normal podocytes, tubular and mesangial cells exposed to Gb3 or LysoGb3 showed the production of TGF-β1 by these cells, and the consequent deposition of ECM proteins [23–25]. To the best of our knowledge, the pathogenic mechanisms leading to renal damage in human specimens from Fabry patients have not yet been revealed. Our hypothesis is that Fabry disease associated cellular alterations in tubular cells induce the production of TGF-β1, which mediates transdifferentiation of renal cells into myofibroblasts. These myofibroblasts secrete extracellular matrix proteins resulting in fibrosis of renal tissue. TGF-β1 could also mediate apoptosis of renal cells through the induction of FGF-2, VEGF and caspase-3.
Table 1 Fabry patients: demographic and clinical data. Proteinuria is expressed as mg/ 24hs. GFR was determined by creatinine clearance (ml/min) (marked with &) or calculated by CKD-EPI formulae (marked with #) from serum creatinine (ml/ min/1.73 m2). #
Gender
Mutation
Classic or Variant
Age at diagnosis
Age at biopsy
Proteinuria
GFR
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
M M F F F M F F M M M F F M M
c.618_619delTT c.1079G > A c.1079G > A c.1066C > T c.1066C > T c.1066C > T c.1066C > T c.667 T > G c.1066C > T c.1066C > T c.1066C > T c.658C > T c.520 T > G c.1024C > T c.145C > G
C V V C C C C V C C C C V C C
23 53 52 43 50 33 34 58 45 28 48 26 54 35 41
23 53 54 43 51 33 34 64 46 28 48 27 54 35 41
800 500 800 200 350 200 250 200 270 3000 2500 160 190 1270 2470
105& 68& 76& 115# 80# 103# 149# 66# 96# 52# 46# 110# 197# 24# 75#
Table 2 Non-Fabry individuals: demographic and clinical data. #
Gender
Age at biopsy
Diagnosis/renal diagnosis
1
M
55
2 3 4 5
F M F M
35 32 67 68
Clear cell renal cell carcinoma - Biopsy was taken from normal tissue Focal segmental glomerulosclerosis Membranous glomerulonephritis Focal segmental glomerulosclerosis Diabetes - Focal segmental glomerulosclerosis
Table 3 Renal biopsies: histological findings. Fabry
Number of total glomeruli
Global GS
FSGS
Inflammatory infiltration
Interstitial fibrosis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
16 20 8 16 17 6 2 5 25 13 20 22 36 10 25
0 0 0 0 0 0 0 1 1 5 1 0 1 1 3
− − − − − − − − − − − − − + +
mild no no no no no no no no moderate mild mild mild moderate moderate
moderate no no mild mild mild no moderate moderate moderate mild mild mild moderate moderate
0 17 2 1 2
− + − + +
no moderate moderate mild moderate
no moderate moderate mild moderate
Non-Fabry 1 70 2 31 3 7 4 9 5 12
The table displays the number of total glomeruli, the number of global glomerular sclerosis (global GS), presence (+) or absence (−) of focal and segmentary glomerular sclerosis (FSGS), the score of inflammatory infiltration (no, mild or moderate) and the score of interstitial inflammation (no, mild or moderate) from Fabry and non-Fabry biopsies.
2. Materials and methods 2.1. Patients and samples A total of 15 renal biopsies from treatment naïve Fabry patients (7 2
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
Fig. 1. Histological analysis of kidney biopsies. 1A and 1B) Representative images of H&E staining of kidney biopsies from Fabry patients showing vacuolization of mesangial cells and podocytes. Picture B is a higher magnification, where the black arrows point to mesangial cells and the blue arrow point to podocyte cell. C) Mallory trichromic reaction in a Fabry biopsy showing mild interstitial fibrosis. D) H&E staining of a Fabry biopsy showing global glomerular sclerosis and interstitial inflammatory lymphocyte infiltration. E) Representative image of H&E staining of kidney biopsies from a Fabry patient showing thickening of interstitial sector. F) Representative image of H&E staining of kidney biopsies from a Fabry patient showing thickening of the arteriolar walls (yellow arrow). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
experienced nephropathologist; in addition, interstitial fibrosis (using Mallory's trichrome reaction) and inflammatory infiltration (using hematoxylin-eosin staining) were scored by light microscopy in mild or moderate when present.
female and 8 male) were included in this study. Patients' age ranged from 23 to 64 years (median age 43). Patient demographic and clinical data are shown in Table 1. Renal biopsies from five non Fabry patients were included as controls (Table 2). Ages from these individuals ranged from 32 to 68 (median age 55). The procedures followed were in accordance with the ethical standards of the Ethical Committee of Framingham (La Plata, Argentina) and with the Helsinki Declaration of 1975, as revised in 2013. All patients provided written informed consent to participate in this study.
2.3. Immunohistochemistry for Gb3 (CD77) Paraffin-embedded kidney biopsies from Fabry patients and nonFabry controls were used. Sections of 5 μm were rehydrated and heatinduced antigen retrieval was performed. Non-specific binding was blocked with 5% bovine serum albumin (Sigma) in PBS for 1 h at room temperature. The specimens were then incubated with a monoclonal antiGb3/CD77 1:500 (Becton Dickinson, Franklin Lakes, NJ, USA) for at least 1 h at room temperature. For detection, HRP-conjugated goat antihuman Ig antibody was used.
2.2. Renal biopsies for histological investigations We received twenty renal biopsies as paraffin blocks for histological study. Serial sections (3 μm) were stained with hematoxylin-eosin and Mallory's trichrome reaction. Glomerular sclerosis was analyzed by an 3
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
Fig. 2. Gb3 immunohistochemistry of kidney biopsies. Representative images of kidney biopsies from a Fabry patient showing Gb3 deposits in epithelial tubular cells, epithelial Bowman cells, endothelial glomerular cells, podocytes and mesangial cells (C) and from a non-Fabry individual with FSGS showing mild Gb3 expression at tubular epithelial cells (D).
sclerosis was present in 7 of 15 Fabry patients (46,7%) (Fig. 1D). The presence of interstitial fibrosis was identified in 12 of 15 (80%) kidney biopsies assayed from Fabry individuals. Interstitial fibrosis was scored as mild in 6 patients, and as moderate in the other 6 patients (Fig. 1C and E). At interstitial level, inflammatory infiltration was found in 7 of 15 patients (46,7%) (Fig. 1D). This infiltration was scored as mild in 26,7% and moderate in 20% of them (Table 3). Sixty seven percent of patients had hyaline thickening of arteriolar wall (Fig. 1F).
2.4. Immunofluorescence confocal microscopy Paraffin-embedded kidney biopsies from Fabry patients and nonFabry controls were used. Sections of 5 μm were rehydrated and heatinduced antigen retrieval was performed. Sections were blocked with goat serum and overnight incubations were performed using the following primary antibodies, all from Abcam: mouse anti-human FGF2 (cat. ab181), rabbit anti-human active caspase 3 (cat. ab49822), rabbit anti-human TGF-β1 (cat. ab66043), rabbit anti-human VEGFA (cat. ab52917) or rabbit anti-human SMA (cat. ab5694) and mouse antihuman CD31 (cat. ab9498). For detection, secondary antibodies were incubated for one hour. FGF2 staining was performed with Alexa Fluor 488-conjugated goat anti-mouse IgG antibody (Invitrogen, cat A11020). For active caspase 3, TGF-β1 and VEGFA staining, Alexa Fluor 488conjugated goat anti-rabbit IgG (Invitrogen, cat A11008) antibody was used. In the double staining protocol, Alexa Fluor 647-conjugated goat anti-rabbit IgG (Invitrogen, cat A21246) and Alexa Fluor 488-conjugated goat anti-mouse IgG antibodies were used for the detection of SMA and CD31, respectively. The nuclei were stained with propidium iodide (1 mg/ml) for 15 min (Sigma, cat P4170). The samples were mounted using fluorescent mounting medium (DakoCytomation, cat S3023) and visualized in a TCS SP5 Leica confocal microscope. The images were obtained using the Leica LAS AF software.
3.3. Gb3 Immunohistochemical analysis Immunohistochemical staining of Gb3 was performed in kidney biopsies using a monoclonal antibody specific for Gb3. This analysis showed Gb3 expression in epithelial tubular cells, epithelial Bowman cells, endothelial glomerular cells, podocytes and mesangial cells from Fabry samples (Fig. 2A). This study confirmed the Gb3 nature of the deposits causing vacuolization of cells. To compare, we also performed the staining in non-Fabry samples, where mild Gb3 expression could be observed only at the epithelial tubular cells (Fig. 2B), as it has been shown by previous reports [26]. 3.4. Immunofluorescence analysis 3.4.1. -TGF-β1 immunofluorescence Previous reports have shown an increment of deposition of extracellular matrix in mesangium and interstitium in kidney biopsies from Fabry patients. It is known that TGF-β1 is the most potent profibrogenic cytokine and it can be potentially produced by all types of renal resident cells [17,18]. In order to unravel the fibrotic mechanism in Fabry nephropathy, we decided to carry out an immunofluorescence analysis. Positive staining for TGF-β1 was found in proximal tubular epithelial cells in all the biopsies from Fabry patients. Occasionally, there was positive staining for TGF-β1 in the vascular smooth muscle of blood vessels. (3/15). The glomeruli did not show any positivity for TGF-β1 staining. As a control of this assay we stained a biopsy from a diabetic non-Fabry patient, which shows TGF-β1 staining both in tubular epithelium and mesangium. We also analyzed a kidney biopsy taken from a section of undamaged tissue of a patient with renal cancer. This tissue presented low or no level of staining for TGF-β1 (Fig. 3). Similarly, the others non-Fabry controls included in this work showed expression at tubular cells and no expression in glomeruli (not shown).
3. Results 3.1. Demographic and clinical data of Fabry patients Demographic and clinical data of naïve Fabry patients included in the study are shown in Table 1. We included 8 males and 7 females with Fabry disease. Eleven of the patients have classical pathogenic mutations (age range: 23–48 years; median age: 35) while the rest are late onset ones (age range:33–64 years; median age: 52). All Fabry patients have pathological proteinuria (Table 1). Renal function expressed as estimated glomerular filtration rate (CKD-EPI formula) is highly variable, with CKD stages from 1 to 4. Non Fabry controls were also included for comparison (3 Males and 2 females). Non Fabry controls were biopsied because of kidney affection other than Fabry disease (Described in Table 2). 3.2. Histological examinations
3.4.2. -FGF-2 and VEGF immunofluorescence It has been suggested that TGF-β1 activates FGF-2 expression in endothelial cells, which then promotes VEGF production [22]. We detected FGF-2 expression in 75% of the stained biopsies from Fabry patients, in tubular cells, mesangial glomerulus cells and blood vessels.
Histological findings from renal biopsies from Fabry and non-Fabry individuals are detailed in Table 3. Histopathological examination revealed enlarged and vacuolated podocytes and mesangial cells in all samples from Fabry patients (Fig. 1A and B). Global glomerular 4
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
Fig. 3. TGF-β1 expression in renal tissue. TGF-β1 analysis was performed by immunofluorescence staining and confocal microscopy in kidney biopsies. Nuclei are shown in red while TGF- β is shown in green. Representative images of kidney biopsies from Fabry patients (A-C) and non-Fabry individuals (D-F) are shown. A) Fabry patients showed positive TGF-β1 staining in proximal tubular cells but not in glomeruli. B) In addition to renal tubules, Fabry patients occasionally presented TGF-β1 expression in the vascular smooth muscle of blood vessels. C) Higher magnification showing lack of expression of TGF- β in Fabry glomerulus. D) Undamaged tissue from an individual with renal cancer showed no expression for TGF- β at the glomerulus and low level of expression in tubular cells. E and F) TGF- β staining was observed in the glomeruli and tubules from a diabetic patient presenting FSGS. Staining was stronger in tubular cells. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
5
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
Fig. 4. FGF2 and VEGFA expression in renal tissue. The analysis was performed by immunofluorescence staining and confocal microscopy in kidney biopsies from Fabry and control individuals. Nuclei are shown in red while FGF2 or VEGFA are shown in green. Representative images are shown. A) FGF2 was found in tubular, mesangial and endothelial cells from most of kidney biopsies taken from Fabry patients. B) Similarly, the expression of FGF2 was found in the whole tissue section obtained from a non-Fabry patient with FSGS. C) VEGFA was expressed at the glomerular level in Fabry patients. D) A non-Fabry individual showing VEGFA expression in the glomeruli and blood vessels. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
control individuals tested in this work, showing a similar pattern of expression (tubules and glomeruli) (Fig. 5).
We also evaluated FGF-2 expression in a non-Fabry patient with FSGS and we found a similar pattern of expression, where the glomerulus, the tubular cells and blood vessels showed to express this protein (Fig. 4A and B). By contrast, we only found FGF-2expression in tubule from undamaged tissue from an individual with renal cancer (data not shown). Positive staining for VEGF was detected in glomerulus and blood vessel cells from Fabry kidney biopsies, as well as from a non-Fabry control with FSGS (Fig. 4C and D).
3.6. Detection of myofibroblasts TGF-β1 is the main profibrotic cytokine, it induces accumulation of myofibroblasts which are the primary cells to synthesize and deposit pathological components of fibrillar matrix characteristic of fibrosis. For this reason, we decided to analyze the staining for a marker of myofibroblasts, the αSMA. We also included CD31 staining to evidence endothelial cells. In Fabry kidneys we could detect positive staining for αSMA on pericytes surrounding peritubular capillaries, mesangial cells and periglomerular zone. On the other hand, non-Fabry controls showed different patterns of expression for αSMA. The undamaged tissue taken from a patient with renal cancer showed positive staining only on peritubular capillaries. Samples from non-Fabry FSGS and membranous glomerulonephritis tissues showed αSMA expression on peritubular capillaries, periglomerular zone and, in the last case, also in peritubular space. (Fig. 6).
3.5. Apoptosis determination Overexpression of TGF-β1, FGF2 and VEGF has been associated to increased apoptosis in kidneys of a Fabry mouse model [22]. For this reason, we aimed to analyze a marker of apoptosis in these kidney specimens. Active caspase 3 staining in biopsies included in this study revealed induction of apoptosis in tubules from all the Fabry biopsies and in the mesangial glomerular cells in 80% of them. Besides, a small percentage (18%) of Fabry specimens displayed this marker in periglomerular zone. Active caspase-3 staining was also present in the 6
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
Fig. 5. Active caspase 3 staining. Immunofluorescence assays revealed the expression of active caspase 3 in kidney biopsies. Nuclei are shown in red while active Caspase 3 is shown in green. Representative images are shown. A) All Fabry patients showed the expression of this proapoptotic protein at tubular cells; part of them also presented active caspase 3 positive cells in the glomeruli. B) A similar pattern of expression was observed in a non-Fabry individual presenting FSGS. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
4. Discussion
endothelial cells, which then promotes VEGF production [22]. VEGF is physiologically expressed by podocytes and tubular cells [30]. Overexpression of VEGF, as happens in diabetes, can cause thickening of the glomerular basement membrane, glomerular enlargement and mesangial proliferation in addition to foot process effacement [31]. The expression of FGF-2 and VEGF in Fabry kidney specimens was proved in this work. We found VEGF expression in glomerulus and blood vessel cells, but not in tubules. This expression pattern was also observed in diabetic nephropathy [32]. It was reported that VEGF in conjunction with TGF-β1 induces podocyte apoptosis [33]. Indeed, casapase-3 positive staining was detected in Fabry kidneys from our work. A proapoptotic role for Gb3 and a higher apoptotic state in Fabry PBMCs has been previously reported [34]. Moreover, apoptosis of cardiomyocytes from hearts of Fabry patients was recently reported [35]. In agreement with these observations, our results strongly suggest that renal cells in Fabry patients undergo apoptosis. Fibrosis is a prominent feature in CKD, including Fabry disease. We and others have shown fibrosis in Fabry biopsies. Fibrosis implies pathological deposition of extracellular matrix proteins, by myofibroblasts. Formation and proliferation of myofibroblasts is induced by TGF-β1, the main profibrotic cytokine. In the present work we could detect positive staining for a marker of myofibroblasts, affirming the presence of those profibrotic cells in human Fabry kidneys. There is still some controversy about the origin of myofibroblasts in injured kidneys. At the interstitium, the pericytes attached to peritubular capillaries could transdifferentiate into myofibroblasts. In the glomerulus, mesangial cells, which are a glomerular pericyte, would be the major source of glomerular fibrosis [12]. However, recent fate-tracing studies have implicated resident fibroblasts and/or pericytes as the likely precursor cells to the myofibroblast [36]. In this work, we detected αSMA, a marker of myofibroblasts, in Fabry kidneys. Particularly, positive staining for αSMA was found in pericytes attached to peritubular capillaries of interstitium, mesangial cells and periglomerular zone. This location may explain the presence of fibrosis both in glomerulus (“glomerulosclerosis”) and insterstitium (“tubulointerstitial fibrosis”) in Fabry kidneys. The results of this work may provide additional evidence to explain the renal damage progression in Fabry nephropathy, beside the vision of the podocyte as the main cell affected leading to kidney problems in Fabry. The proximal tubular kidney cell is the one that produces TGF-
Various insults to the kidney are known to lead to chronic kidney disease. Regardless of the etiology, chronic disease is characterized by kidney damage seen histopathologically as glomerulosclerosis, interstitial fibrosis and inflammation [15]. Kidney is one of the target organs affected in Fabry disease and kidney biopsies from patients with overt nephropathy may display the same features of chronic kidney disease of different etiologies. The primary insult in Fabry disease is the glycolipid deposits in almost every glomerular cell type, with massive deposits in podocytes, but also in mesangial, endothelial cells and parietal epithelial cells. Not only those deposits appear in glomerular cells, they are also detected in tubular cells and peritubular capillary endothelium [11,27]. In this work, the classical histological analysis revealed similar results, with all kidney cells affected by glycolipid deposits. Another interesting feature revealed by histological analysis was the prominent interstitial fibrosis, shown in 80% of Fabry kidney biopsies assayed. Glomerular sclerosis was also found although with a minor frequency. Podocytes are reported to be the cell type with the highest score of Gb3 inclusions and vacuolization. In a small study, patients treated with ERT showed clearance of the podocytes and concomitant reduction of podocyte effacement [27]. The massive deposits in podocytes have identified podocytes as the target cell type whose affection could correlate to pathophysiology of chronic kidney disease. Moreover, an in vitro study has shown the production of TGF-β1 and extracellular matrix proteins when normal podocytes are exposed to lyso-Gb3 [23]. Extensive evidence in experimental animal models of both, type 1 and type 2 diabetes, implicates TGF-β1 as an important mediator of diabetic kidney disease, playing a pathophysiological role in promoting glomerulosclerosis, interstitial fibrosis, and the decline in GFR [28]. TGFβ1 mRNA and protein levels are increased in both the glomerular and tubular compartments of various models of experimental diabetes in rats and mice [29]. Considering the above data, we evaluated the production of TGF-β1 in kidney biopsies from Fabry patients in order to identify the pathological mechanism associated with nephropathy and fibrosis in Fabry disease and the main cell types involved. The immunofluorescence analysis carried out in this work revealed that TGF-β1 is produced mainly by proximal tubular cells in all Fabry kidney biopsies studied, but not by glomerular cells, including podocytes. It has been suggested that TGF-β1 activates FGF-2 expression in 7
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
Fig. 6. Myofibroblasts markers in kidney biopsies. Immunofluorescence staining of αSMA (green) and CD31 (red), and confocal microscopy in kidney biopsies was performed. Nuclei are shown in blue. Representative images are shown. Kidney biopsies from Fabry patients (A and B), a non-Fabry patient with renal cancer (D), and a non-Fabry patient with FSGS (E) are shown. C and F shows glomeruli staining with higher magnification for a Fabry patient (C) and a non-Fabry control with membranous glomerulonephritis (F). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
8
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
β1, the main profibrotic cytokine. Therefore, these could be the cells in which the pathological profibrotic changes are initiated. Moreover, the presence of cells with myofibroblast markers in the glomeruli may play a role in producing renal fibrosis. Of course, it does not rule out other pathways stimulated by Gb3 deposition. Podocytes also may release pro-inflamamtory and/or profibrotic cytokines as many experimental papers have demonstrated. In addition it has been reported that glycolipids could induce other cellular mechanisms that could mediate fibrosis. It was shown Gb3 deposition is associated with a reduction in the activity of endothelial nitric oxide synthase resulting in a decreased nitric oxide bioavailability or enzyme uncoupling determining the production of reactive species and an increase of adhesion molecule [37]. Another work showed that lyso-Gb3 deposits stimulate the proliferation of smooth muscle cells with a re-modeling of the intima-media layer of the blood vessels [38]. Both mechanisms could coexist and can cause the release of cytokines with tissue damage in all organs. Moreover, the vascular damage can impair the blood blow to the organs with the following histological damage. A dysregulation of autophagy could also contribute to tissue damage. Gb3 deposition is associated with inhibition of mTOR kinase activity, a negative regulator of the autophagic machinery [39]. Our data suggest that Fabry disease associated cellular alterations, likely Gb3 deposition, is followed by the activation of biological processes resulting in tissue fibrosis. A recent article by Braun et al. has demonstrated in a model of podocyte cell culture, that Gb3 deposition triggers autophagy alterations and stimulate pro-fibrotic processes. Moreover, the addition of recombinant alpha-galactosidase determined a complete clearance of Gb3, however the pathologic pathways continued to be active. This evidence demonstrates that the deposition of Gb3 activates pathogenic biological pathways that can become independent from the Gb3 deposition itself. Recently, new molecules (pegunigalsidasealfa, moss-made agalsidasealfa, lucerastat, venglustat) have been proposed for treatment of Fabry disease and some of them might have a role in modulating fibrosis processes; particularly lucerastat and venglustat that, inhibiting the glucosylceramidesynthetase (GCS), are able to reduce the synthesis of precursors and the levels of the metabolites that accumulate. The reduction of the substrate deposition might contribute to slow down the processes involved in fibrosis pathways [40]. This study has a limitation, associated to the relatively small number of patients included, and the less representation of variant patients as compared to higher number of classical ones; and the inclusion of both, males and females. Therefore, more studies including more biopsies should be carried out in order to validate these results. In conclusion, our results show for the first time that TGF- β1 is expressed in human renal tissue from Fabry patients, and that this profibrotic cytokine is mainly produced by proximal tubular cells. In addition both, peritubular interstitium and glomeruli, presented cells positive for myofibroblasts markers such as αSMA, suggesting a role for pro-fibrotic cells in the pathogenic pathways leading to Fabry nephropathy. On the other hand, active caspase 3 expression at the tubular epithelium suggest that apoptosis of the tubular cell is also involved in tissue damage. Through these mechanisms the pathological process of fibrosis characteristic of Fabry nephropathy would be initiated and propagated.
Research program under IIR-ARG-001540 (Japan) and by Universidad Nacional de La Plata [X679 to PR] (Argentina). None of the funders was involved in the study design, data collection, data analysis, data interpretation, writing of the report, or decision to submit for publication. References [1] R.O. Brady, A.E. Gal, R.M. Bradley, et al., Enzymatic defect in Fabry’s disease Ceramidetrihexosidase deficiency, N. Engl. J. Med. 276 (1967) 1163–1167. [2] M. Arends, C. Wanner, D. Hughes, et al., Characterization of classical and nonclassical Fabry disease: a multicenter study, J. Am. Soc. Nephrol. 28 (2017) 1631–1641. [3] B. Najafian, M. Mauer, R.J. Hopkin, et al., Renal complications of Fabry disease in children, Pediatr. Nephrol. 28 (2013) 679–687. [4] M. Arends, M. Biegstraaten, D.A. Hughes, et al., Retrospective study of long-term outcomes of enzyme replacement therapy in Fabry disease: analysis of prognostic factors, PLoS One 12 (8) (2017 Aug 1) e0182379. [5] M. Beck, D. Hughes, C. Kampmann, et al., Long term outcomes with agalsidase alfa enzyme replacement therapy: analysis using deconstructed composite events, Mol. Genet. Metab. Rep 14 (2018) 31–35. [6] D.P. Germain, J. Charrow, R.J. Desnick, et al., Ten-year outcome of enzyme replacement therapy with agalsidase beta in patients with Fabry disease, J. Med. Genet. 52 (2015) 353–358. [7] M. Beck, D. Hughes, C. Kampmann, et al., Long-term outcomes with agalsidase alfa enzyme replacement therapy: analysis using deconstructed composite events, Mol. Genet. Metab. Rep. 14 (2017 Nov 9) 31–35. [8] S. Waldek, S. Feriozzi, Fabry nephropathy: a review - how can we optimize the management of Fabry nephropathy? BMC Nephrol. 15 (2014) 72. [9] D.P. Germain, K. Nicholls, R. Giugliani, et al., Efficacy of the pharmacologic chaperone migalastat in a subset of male patients with the classic phenotype of Fabry disease and migalastat-amenable variants: data from the phase 3 randomized, multicenter, double-blind clinical trial and extension study, Genet Med. (2019 Feb) 6, https://doi.org/10.1038/s41436-019-0451-z. [10] B. Najafian, E. Svarstad, L. Bostad, et al., Progressive podocyte injury and globotriaosylceramide (GL-3) accumulation in young patients with Fabry disease, Kidney Int. 79 (2011) 663–670. [11] C. Tøndel, L. Bostad, A. Hirth, et al., Renal biopsy findings in children and adolescents with Fabry disease and minimal albuminuria, Am. J. Kidney Dis. 51 (5) (2008) 767–776. [12] A. Sessa, M. Meroni, G. Battini, et al., Evolution of renal pathology in Fabry disease, Acta Paediatr. Suppl. 92 (2003) 6–8 discussion 5. [13] E. Svarstad, L. Bostad, O. Kaarbøe, et al., Focal and segmental glomerular sclerosis (FSGS) in a man and a woman with Fabry’s disease, Clin. Nephrol. 63 (2005) 394–401. [14] A.B. Fogo, L. Bostad, E. Svarstad, et al., Scoring system for renal pathology in Fabry disease: report of the international study group of Fabry Nephropathy (ISGFN), Nephrol. Dial. Transplant. 25 (2010) 2168–2177. [15] G. Campanholle, G. Ligresti, S.A. Gharib, et al., Cellular mechanisms of tissue fibrosis. 3. Novel mechanisms of kidney fibrosis, Am. J. Phys. Cell Phys. 304 (2013) C591–C603. [16] X.-M. Meng, P.M.-K. Tang, J. Li, et al., TGF-β1/Smad signaling in renal fibrosis, Front. Physiol. 6 (2015) 82. [17] B. Sutariya, D. Jhonsa, M.N. Saraf, TGF-β1: the connecting link between nephropathy and fibrosis, Immunopharmacol. Immunotoxicol. 38 (2016) 39–49. [18] R. Samarakoon, J.M. Overstreet, P.J. Higgins, TGF-β1 signaling in tissue fibrosis: redox controls, target genes and therapeutic opportunities, Cell. Signal. 25 (2013) 264–268. [19] C.-F. Wu, W.-C. Chiang, C.-F. Lai, et al., Transforming growth factor β-1 stimulates profibrotic epithelial signaling to activate pericyte-myofibroblast transition in obstructive kidney fibrosis, Am. J. Pathol. 182 (2013) 118–131. [20] Y.B.Y. Sun, X. Qu, G. Caruana, J. Li, The origin of renal fibroblasts/myofibroblasts and the signals that trigger fibrosis, Differentiation. 92 (3) (2016) 102–107. [21] G. Ferrari, G. Pintucci, G. Seghezzi, et al., VEGF, a prosurvival factor, acts in concert with TGF-beta1 to induce endothelial cell apoptosis, Proc. Natl. Acad. Sci. U. S. A. 103 (2006) 17260–17265. [22] M.H. Lee, E.N. Choi, Y.J. Jeon, et al., Possible role of transforming growth factor-β1 and vascular endothelial growth factor in Fabry disease nephropathy, Int. J. Mol. Med. 30 (2012) 1275–1280. [23] M.D. Sanchez-Niño, A.B. Sanz, S. Carrasco, et al., Globotriaosylsphingosine actions on human glomerular podocytes: implications for Fabry nephropathy, Nephrol. Dial. Transplant. 26 (2011) 1797–1802. [24] Y.J. Jeon, N. Jung, J.-W. Park, et al., Epithelial-mesenchymal transition in kidney tubular epithelial cells induced by Globotriaosylsphingosine and Globotriaosylceramide, PLoS One 10 (2015) e0136442. [25] Y.-J. Shin, Y.J. Jeon, N. Jung, et al., Substrate-specific gene expression profiles in different kidney cell types are associated with Fabry disease, Mol. Med. Rep. 12 (2015) 5049–5057. [26] C. Valbuena, D. Leitão, F. Carneiro, et al., Immunohistochemical diagnosis of Fabry nephropathy and localisation of globotriaosylceramide deposits in paraffin-embedded kidney tissue sections, Virchows Arch. 460 (2012) 211–221. [27] C. Tøndel, L. Bostad, K.K. Larsen, et al., Agalsidase benefits renal histology in young patients with Fabry disease, J. Am. Soc. Nephrol. 24 (2013) 137–148. [28] A.S. Chang, C.K. Hathaway, O. Smithies, et al., Transforming growth factor-β1 and diabetic nephropathy, Am. J. Physiol. Ren. Physiol. 201 (310) (2016) F689–F696.
Acknowledgements We acknowledge the physicians who included their patients in the study: Segundo Fernández, Laura Gil, Sebastian Calabrese and Monica Calvo. Funding This project was funded by Shire Human Genetic Therapies Inc., now part of the Takeda group of companies, Investigator Initiated 9
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
P.A. Rozenfeld, et al.
[29] F.N. Ziyadeh, Mediators of diabetic renal disease: the case for TGF-Beta as the major mediator, J. Am. Soc. Nephrol. 15 (Suppl. 1) (2004) S55–S57. [30] A. Tufro, D. Veron, VEGF and podocytes in diabetic nephropathy, Semin. Nephrol. 32 (2012) 385–393. [31] D. Veron, K.J. Reidy, C. Bertuccio, et al., Overexpression of VEGF-A in podocytes of adult mice causes glomerular disease, Kidney Int. 77 (2010) 989–999. [32] M.T. Lindenmeyer, M. Kretzler, A. Boucherot, et al., Interstitial vascular rarefaction and reduced VEGF-A expression in human diabetic nephropathy, J. Am. Soc. Nephrol. 18 (2007) 1765–1776. [33] X. Bai, J. Geng, X. Li, F. Yang, J. Tian, VEGF-A inhibition ameliorates podocyte apoptosis via repression of activating protein 1 in diabetes, Am. J. Nephrol. 40 (2014) 523–534. [34] P.N. De Francesco, J.M. Mucci, R. Ceci, et al., Higher apoptotic state in Fabry disease peripheral blood mononuclear cells: effect of globotriaosylceramide, Mol. Genet. Metab. 104 (2011) 319–324. [35] A. Frustaci, C. Chimenti, D. Doheny, et al., Evolution of cardiac pathology in classic
[36] [37] [38] [39] [40]
10
Fabry disease: progressive cardiomyocyte enlargement leads to increased cell death and fibrosis, and correlates with severity of ventricular hypertrophy, Int. J. Cardiol. 248 (2017) 257–262. L.S. Gewin, Renal fibrosis: primacy of the proximal tubule, Matrix Biol. 68–69 (2018) 248–262. L. Shu, A. Vivekanandan-Giri, S. Pennathur, et al., Establishing 3-nitrotyrosine as a biomarker for the vasculopathy of Fabry disease, Kidney Int. 86 (2014) 58–66. J.M. Aerts, J.E. Groener, S. Kuiper, et al., Elevated globotriaosylsphingosine is a hallmark of Fabry disease, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 2812–2817. M.C. Liebau, F. Braun, K. Höpker, et al., Dysregulated autophagy contributes to podocyte damage in Fabry’s disease, PlosOne 8 (2013) e63506. N. Guérard, D. Oder, P. Nordbeck, et al., Lucerastat, an Iminosugar for substrate reduction therapy: tolerability, pharmacodynamics, and pharmacokinetics in patients with Fabry disease on enzyme replacement, Clin. Pharmacol. Ther. 103 (2018) 703–711.