Fetuin A for Treatment of Renal Disorders

Abstract

The present invention relates to Fetuin A (AHSG) for use in a method for treating a renal disorder, wherein an amount of Fetuin A effective for treating the renal disorder is administered to a subject in need thereof. The present invention further relates to a pharmaceutical composition for use in treating a renal disorder comprising Fetuin A and optionally at least one pharmaceutical acceptable carrier.

Claims

1. Fetuin A (AHSG) for use in a method for treating a renal disorder, wherein an amount of Fetuin A effective for treating the renal disorder is administered to a subject in need thereof.

2. Fetuin A for use in a method according to claim 1, wherein the renal disorder is selected from the group consisting of acute renal disorders, chronic renal disorders, kidney fibrosis, chronic kidney disease, renal insufficiency, renal inflammation, acute kidney injuries, ischemic renal disorders, disorders related to kidney hypoxia, renal ischemia-reperfusion injury, disorders related to kidney transplantation, disorders related to cardiovascular surgery, kidney tissue damage and combinations thereof.

3. Fetuin A for use in a method according to claim 2, wherein the renal disorder is a disorder related to kidney transplantation, in particular wherein the disorder related to kidney transplantation is selected from the group consisting of delayed graft function, organ rejection of kidney transplants, kidney tissue damage resulting from kidney transplantation, inflammation resulting from kidney transplantation, ischemia-reperfusion injury resulting from kidney transplantation, and combinations thereof.

4. Fetuin A for use in a method according to any one of the preceding claims, wherein the renal disorder is caused by hypoxia during surgery, preferably wherein the hypoxia is caused by ischemia during kidney transplantation or during cardiovascular surgery.

5. Fetuin A for use in a method according to any one of the preceding claims, wherein the Fetuin A is human Fetuin A, preferably wherein the Fetuin A is derived from human blood plasma.

6. Fetuin A for use in a method according to any one of the preceding claims, wherein the Fetuin A is at least 70%, at least 80%, at least 90%, at least 95% or at least 99% identical and/or homologous to SEQ ID NO: 1.

7. Fetuin A for use in a method according to any one of the preceding claims, wherein the Fetuin A comprises or consists of SEQ ID NO: 1 or fragments thereof.

8. Fetuin A for use in a method according to any one of the preceding claims, wherein the Fetuin A is recombinantly expressed, preferably in a host cell system providing posttranslational modifications, more preferably glycosylation.

9. Fetuin A for use in a method according to any one of the preceding claims, wherein the Fetuin A is administered to the subject by way of injection or inhalation.

10. Fetuin A for use in a method according to claim 9, wherein the Fetuin A is injected intravenously to the subject.

11. Fetuin A for use in a method according to any one of the preceding claims, wherein the Fetuin A is administered to the subject in a concentration of from 1 to 200 mg/kg (of body weight), from 5 to 100 mg/kg (of body weight) or from 10 to 50 mg/kg (of body weight).

12. Fetuin A for use in a method according to any one of the preceding claims, wherein the treatment comprises modulation of calcification levels in the kidney tissue, in particular hypoxia-related calcification levels.

13. Fetuin A for use in a method according to any one of the preceding claims, wherein the treatment comprises removal of calcium mineral depositions in kidney tissue, in particular the removal of calcium mineral depositions in transplanted kidney tissue.

14. Fetuin A for use in a method according to any one of the preceding claims, wherein at least two or at least three doses of the Fetuin A are administered.

15. Fetuin A for use in a method according to any one of the preceding claims, wherein at least one dose of Fetuin A is administered prior, during or after an ischemic incident and/or surgery, in particular wherein the ischemic incident and/or the surgery is a kidney transplantation or cardiovascular surgery.

16. Fetuin A for use in a method according to claim 15, wherein at least one dose of Fetuin A is administered within 48 hours before an ischemic incident and/or surgery or wherein at least one dose of Fetuin A is administered within 48 hours after an ischemic incident and/or surgery.

17. Fetuin A for use in a method according to any one of the preceding claims, wherein Fetuin A is administered in divided doses over a total duration of 1 to 50 days, in particular over a total duration of 1 to 25 days or over a total duration of 2 to 10 days.

18. Pharmaceutical composition for use in treating a renal disorder, comprising Fetuin A and optionally at least one pharmaceutical acceptable carrier.

19. The pharmaceutical composition for use in treating a renal disorder of claim 18, wherein the renal disorder is selected from the group consisting of acute renal disorders, chronic renal disorders, kidney fibrosis, chronic kidney disease, renal insufficiency, renal inflammation, acute kidney injuries, ischemic renal disorders, disorders related to hypoxia, renal ischemia-reperfusion injury, disorders related to kidney transplantation, disorders related to cardiovascular surgery, kidney tissue damage and combinations thereof.

20. The pharmaceutical composition for use in treating a renal disorder of claim 19, wherein the disorders related to kidney transplantation are selected from the group consisting of delayed graft function, organ rejection of kidney transplants, kidney tissue damage resulting from kidney transplantation, inflammation resulting from kidney transplantation, renal ischemia-reperfusion injury resulting from kidney transplantation and combinations thereof.

21. The pharmaceutical composition for use in treating a renal disorder of claims 18 to 20, wherein the renal disorder is caused by hypoxia during surgery, preferably wherein the hypoxia is caused by ischemia during kidney transplantation or during cardiovascular surgery.

22. The pharmaceutical composition for use in treating a renal disorder according to any one of claims 18 to 21, wherein the Fetuin A is defined as in any one of claims 5 to 17.

Description

FIGURES

[0078] FIG. 1 Chronic fetal hypoxia induces intra-uterine growth restriction in mice: FIG. 1 A. Experimental setup and time points of analysis. FIG. 1 B. Mean number of pups per litter: FIG. 1 C. Total weight distribution of E18.5 fetuses (n=49 for each experimental condition; No: normoxic, Hy: hypoxic, Cc: caloric controls). FIG. 1 D. The mean number of nephrons per E18.5 kidney was determined by staining for the glomerular marker nephrin. Unpaired t test with Welch’s correction.

[0079] FIG. 2 Hypoxia-induced gene expression in the kidney: FIG. 2 A. Hierarchical clustering of cDNA microarray data comparing the renal gene expression profiles of hypoxic, normoxic caloric control group E18.5 fetuses (n=3 per experimental condition). White regions indicate induction and vertically hatched region repression (degree of induction, repression not indicated in FIG. 2 A). The position of Ahsg is marked by an arrow. Clustering was performed for genes with at least 1.3-fold regulation of hypoxic vs. both normoxic controls; 1way ANOVA; P<0.05. FIGS. 2 B-E. Relative mRNA values of Ahsg (B), Apoa2 (C), Fgg (D) and Gys2 (E) in E18.5 kidneys (circles) or liver samples (squares). FIG. 2 F. Fetuin A plasma levels of E18.5 fetuses assessed by ELISA.

[0080] FIG. 3 Fetal hypoxia induced Fetuin A expression in the proximal tubulus: FIG. 3 A-H. Fetuin A staining on E18.5 kidney sections of normoxic (A-D) or hypoxic (E-H), wildtype (A,C,E,G) or Clcn5 KO (B,D,F,H) fetuses. Arrowheads in (D,H) indicate intraluminal Fetuin A staining that results from the impaired endocytosis of low molecular weight proteins in the PT of Clcn5 KO mice. (I-L) Immunofluorescence staining of the indicated nephron segment marker proteins and Fetuin A on E18.5 kidney sections. PT, proximal tubulus; TAL, thick ascending limb; DCT, distal convoluted tubulus; CD, collecting duct; Scale bars, 300 .Math.m (overview images) or 50 .Math.m.

[0081] FIG. 4 Hypoxia activated Fetuin A expression in vitro: FIG. 4 A. Depiction of the potential HREs of mouse Ahsg that were used to generate the luciferase reporter gene constructs (1-8). Mutated HREs are shown. FIG. 4 B. Luciferase activity in NRK cells transfected with the reporter constructs. The fold change in luciferase light emission between hypoxic and normoxic culture conditions is shown. Each condition is normalized to the empty vector control (pGL3). Dunnett’s multiple comparisons test. FIGS. 4 C-E. Expression of Fetuin A in mouse primary proximal tubular cells (pPTC) isolated from 4 different mice (C), in the rat cell line NRK (D) and in the human cell line HK-2 (E) cultured under normoxic or hypoxic conditions.

[0082] FIG. 5 Fetuin A deficiency aggravated CKD progression in hypoxic IUGR kidneys: FIGS. 5 A-B. The decline in kidney function of 9 week old mice as assessed by the protein/creatinine ratio and the glomerular filtration rate (GFR). FIG. 5 B. is most pronounced fetal hypoxic Ahsg KO animals, showing an additive effect of hypoxia and Fetuin A deficiency (n≥5 for each experimental condition). FIGS. 5 C-D. Representative image of picrosirius red staining for collagen shows a stronger, more intricate pattern on kidney sections derived from P14 fetal hypoxic Ahsg KO mice (D) compard to fetal hypoxic wt mice (C). FIGS. 5 E-G. The relative expression levels of the fibrosis markers Acta2 (E), Col1a1 (F) and Fn1 (G) are markedly enhanced in kidneys of fetal hypoxic Ahsg KO mice (n≥5 for each experimental condition). Fisher’s LSD test (A-B). Scale bars, 50 .Math.m.

[0083] FIG. 6 Fetuin A deficiency promoted accumulation of calcium mineral particles and macrophages in hypoxic IUGR kidneys: FIGS. 6 A-l. Detection of calcium biominerals by ATTO 488 fluorescently labelled Fetuin A (488-FA) staining on E18.5 kidney sections of normoxic wt (A,D,G), hypoxic wt (B,E,H) or hypoxic Ahsg KO embryos (C,F,l). Hypoxic Ahsg KO mice exhibit the strongest staining intensity in the PT (C) compared to normoxic and hypoxic wt mice (A,B) indicative of an increased mineralized matrix turnover. The arrowheads point towards granular staining pattern, reflecting bulk accumulation of 488-FA in the papilla (F) and cortex (l) only present in hypoxic Ahsg KO mice. FIGS. 6 J-M. Representative immunofluorescence staining for the macrophage marker F4/80 (j′, k′, l′, m′)and Fetuin A (j‴, k‴, I‴, m‴), counterstained with DAPI (j⁗, k⁗, I⁗, m⁗) and 488-FA (j″ , k″, l″, m″) on E18.5 kidney sections for the indicated genotypes and oxygen conditions. An accumulation of macrophages and diffuse as well as granular 488-FA staining is only detected in hypoxic Ahsg KO mice (M). FIG. 6 N. Quantification of total F4/80 stained area per field of view. Multiple sections of at least 3 mice were analysed for each experimental condition. Scale bars, 100 .Math.m.

[0084] FIG. 7 Fetuin A attenuated hypoxia-induced expression of fibrotic markers: FIG. 7 A. Representative Western blot showing the induction of fibronectin and α-smooth muscle actin (α-SMA) upon hypoxic culture conditions in primary proximal tubular cells (pPTCs) isolated from wt or Ahsg KO mice. FIGS. 7 B-D. Fetuin A supplementation attenuates the hypoxia-induced expression of the fibrotic markers Acta2 (B), Col1a1 (C) and Fn1 (D) in pPTCs (n≥5 for each experimental condition). FIGS. 7 E-F. Only Fetuin A, but not BSA, has a diminishing effect on the expression of fibrotic markers (n≥4 for each experimental condition). FIG. 7 G. Representative Western blot depicting that Fetuin A supplementation reduces the expression of fibronectin and collagen type I (Col1a1) protein. FIG. 7 H. TGF-β1 treatment and hypoxia have an additive effect on the phosphorylation of Smad3. A representative Western blot is shown. FIG. 7 l. Quantification of Smad3 activation shown in (H). Unpaired t test with Welch’s correction (B-D, only for comparison of normoxic wt and normoxic Ahsg KO samples), Fisher’s LSD test (l).

[0085] FIG. 8 Hypoxia, resp IRI induced deposition of calcium containing microparticles in the kidney: Detection of calcium containing microparticles with 488-FA following ischemia reperfusion (IRP). 488-FA staining on control side (l) and ischemia reperfusion (m) 7d after surgery shows the presence of calcium biominerals only in IRP kidneys.

[0086] FIG. 9 Fetuin A supplementation attenuated the expression of fibrotic markers following ischemic reperfusion injury: FIGS. 9 A-F. Fetuin A supplementation attenuates ischemia-induced expression of Col1a1 (A), Col3a1 (B), Col6a1 (C), but has no effect on the expression of the non-collagen fibrotic markers Acta2 (D), Fn1 (E) or Vim (F). One-way ANOVA test.

[0087] FIG. 10 Prophylactic administration of human Fetuin A in mice undergoing ischemia reperfusion injury (IRI) led to less damage and less fibrotic remodeling: FIGS. 10 A-D. Expression levels of early damage markers Kim1 (A), Krt18 (B), Krt20 (C) and Lcn2 (D) are reduced in IRI kidneys pretreated with Fetuin A compared to IRI kidneys pretreated with PBS. FIGS. 10 E and F. Fibrotic remodeling markers Clu (E) and Tgfb1 (F) are reduced as well compared to PBS control during IRI. FIG. 10 G. The non-collagenous fibrotic marker Vim (G) is reduced in Fetuin A pretreated IRI kidneys compared to PBS pretreated IRI kidneys. FIGS. 10 H-J. Collagen expression upon Fetuin A prophylactic treatment is shown for Col1a1 (H), Col3a1 (l) and Col6a1 (J). FIG. 10 K. Arg1 (K) expression is repressed in Fetuin A pretreated mice compared in PBS pretreated mice, indicating a reduced requirement for collagen synthesis. FIGS. 10 L-N. Pretreatment with Fetuin A also reduces the expression of chemokines Ccl2 (L), ll-6 (M) and Tnf-α (N) involved in damage signaling and repair that are usually increased under ischemic conditions. Legend for FIG. 10: open circles: PBS injected, unoperated control kidneys; open triangles: PBS injected, IRI kidneys; filled black circles: Fetuin A injected, unoperated control kidneys; filled black triangles: Fetuin A injected, IRI kidneys. P values in FIG. 10 are denoted for ordinary one-way ANOVA with Tukey correction for multiple comparisons (regular font) and paired t test (bold font).

[0088] FIG. 11 Presence of exogenous Fetuin A in the renal tissue of mice upon intravenous administration: FIG. 11 A. Western blots. Lanes A to C: Fetuin A injected, unoperated control kidneys; Lanes D to F: Fetuin A injected, IRI kidneys; Lanes G to l: PBS injected, unoperated control kidneys; Lanes J to L: PBS injected, IRI kidneys; Lane M: empty; Lane N: human Fetuin A protein used for injection. Top panel: Western blot for endogenous mouse Fetuin A using the ab187051 anti-Fetuin A antibody (specific for mouse Fetuin A) from abcam. Middle panel: Western blot for human Fetuin A using the sc-133146-HRP anti-Fetuin A antibody (specific for human Fetuin A) from Santa Cruz. Bottom panel: Rplp0 served as loading control. Kidney samples from mice injected with human Fetuin A protein showed a strong band for human Fetuin A (lanes A-F). No Fetuin A was detected in PBS-injected control samples (lanes G-L). PBS-injected IRI kidneys (lanes J-L) showed the highest abundance of mouse Fetuin A among all samples. Human Fetuin A was not detected with the mouse-specific anti-Fetuin A antibody ab187051, because human Fetuin A protein (lane N) gave no signal. FIG. 11 B. ELISA for mouse Fetuin-A using the MFTA00 ELISA kit (specific for mouse Fetuin-A) from bio-techne. Endogenous mouse Fetuin A was only elevated in PBS-injected IRI kidneys compared to all other samples. FIG. 11 C. ELISA for human Fetuin A using the DFTA00 ELISA kit (specific for human Fetuin-A) from bio-techne. Human Fetuin A could not be detected in kidney samples of PBS-injected mice. Fetuin A injected IRI kidneys have higher tissue levels of human Fetuin A than the unoperated contralateral (right) kidney. FIGS. 11 D-F. Exogenous Fetuin A was present in the renal tissue of mice upon intravenous administration. Fluorescently labeled Fetuin A (Alexa-488-labeled Fetuin A) as present in the proximal tubules (PT) 15, 30, and 60 minutes after intravenous injection. 15 minutes after injection of Alexa-488-labeled Fetuin A, a fine punctated staining was detected in the proximal tubules. The intensity of this staining increased at 30 minutes and at 60 minutes after injection. G, glomerulus. Legend for FIGS. 11 B and C: open circles: PBS injected, unoperated control kidneys; open triangles: PBS injected, IRI kidneys; filled black circles: Fetuin A injected, unoperated control kidneys; filled black triangles: Fetuin A injected, lRl kidneys. P values in FIGS. 11 B and C are denoted for ordinary one-way ANOVA with Tukey correction for multiple comparisons (regular font) and paired t test (bold font).

[0089] FIG. 12 Time course of human Fetuin A in the serum of mice upon intravenous administration: FIG. 12 A. Lanes A and B: 1 minute after injection; lanes C and D: 1 hour after injection; lanes E and F: 1 day after injection; lanes G and H: 1 week after injection; lanes I and J: control PBS injection; lane K: human Fetuin A protein used for injection. Western blot for human Fetuin A using the human-specific sc-133146-HRP anti-Fetuin A antibody from Santa Cruz. After 1 day, the detected amount of human Fetuin A in the serum was reduced compared to 1 minute or 1 hour post injection. After 1 week no human Fetuin A was detectable. Mouse Fetuin A was not detected with the sc-133146-HRP anti-Fetuin A antibody, because PBS-injected control animals give no signals. FIG. 12 B. Normalized ELISA data using the DFTA00 ELISA kit (specific for human Fetuin-A) from bio-techne, showed the decline of intravenous injected human Fetuin A in mouse serum samples. Compared to the injection time point (1 min), the amount of circulating human Fetuin A was reduced to 80 % after 1 hour, and to 8.5 % after 1 day. After 1 week, human Fetuin A could not be detected anymore, similar to PBS-injected mice (black circles). FIG. 12 C. Time course of human Fetuin A in the serum of mice upon intravenous administration. Normalized ELISA data showing the decline of intravenous injected human Fetuin A in mouse serum samples. Compared to the injection time point (1 min), the amount of circulating human Fetuin A was reduced to 80 % after 1 hour, and to 8.5 % after 1 day. After 1 week, human Fetuin A could not be detected anymore.

EXAMPLES

[0090] Chronic fetal hypoxia induces IUGR in mice. To model chronic fetal hypoxia, timed-mated pregnant mice were exposed to 10% oxygen from E14.5 to E18.5 (FIG. 1 A). As it was observed that hypoxic dams (Hy) ate less despite free food access and since caloric restriction itself is a known inducer of IUGR5, a second control group was included in the analysis to exclude the possibility that the findings might not be due to hypoxia, but rather to reduced ingested calories. In this caloric control (Cc) group, normoxic dams were fed with an amount of food matching the reduced amount of food consumed by the hypoxic mice. Litter size, placental mass or fetal viability were indistinguishable among the three groups, whereas maternal weight gain was significantly reduced in hypoxic dams (FIG. 1 B). Importantly, only fetuses from hypoxic dams showed LBW, fulfilling IUGR criteria (FIG. 1 C), whereas fetuses from the normoxic or the Cc group did not. Kidneys of hypoxic fetuses were smaller with significantly fewer nephrons compared to controls (FIG. 1 D).

[0091] Hypoxic fetal kidneys adopt a hepatic gene expression pattern. Next, a whole genome expression in fetal hypoxic kidneys was examined using gene arrays. 62 induced and 28 repressed genes were identified and compared to both control groups (FIG. 2 A). Of the induced genes, 17 are known to be regulated by hypoxia, including the bona fide HIF target genes transferrin, trefoil factor 3, neuritin, alpha-1- antitrypsin (Serpina1d) and alpha-1-antichymotrypsin (Serpina3n). Furthermore, more than 20% of the induced genes were found to be frequently purified from calciprotein particles (CPPs), comprising the major CPP components Fetuin A (Ahsg), albumin, Apo-A1 and thrombin (F2). Functional annotation clustering of the induced genes revealed in hypoxic kidneys an enrichment of secreted plasma proteins that are normally transcribed exclusively by the liver. Validation of the microarray data by RT-qPCR of select genes confirmed a more than 10- fold induction only in hypoxic fetal kidneys, but not in control group kidneys (FIGS. 2 B-E). No change in mRNA levels were found in fetal livers samples.

[0092] Fetuin A is produced locally in the proximal tubulus under hypoxic conditions. Interestingly, the gene with the highest induction (Ahsg) was found in 7 of the 10 annotation groups. Ahsg belongs to the cystatin superfamily of cysteine protease inhibitors, encoding for the negative acute phase glycoprotein Fetuin A, whose main function concerns mineralized matrix metabolism. Despite its strong induction in hypoxic kidneys, no significant rise could be detected in fetal plasma Fetuin A levels (FIG. 2 F), implying a specific, local, rather than a systemic functional relevance during hypoxic developmental challenges. To further address the functional relevance of Ahsg expression in fetal hypoxic kidneys, its precise localization was examined. FIG. 3 shows immunohistochemistry of fetal kidney proximal tubules (FIGS. 3 A, B, E, F) or tubule lumen (FIGS. 3 C, D, G, H). Strong Fetuin A staining demarcated cells of the proximal tubule (PT) regardless of the oxygen conditions (FIGS. 3 A, E). This is due to filtration and uptake of systemic Fetuin A into the PT via megalin-dependent endocytosis, masking any Fetuin A locally produced in the hypoxic fetal kidney. To selectively visualize Fetuin A protein of renal origin, a genetic approach was employed to block endocytosis into the PT, an alternative method to the 7 pharmacological inhibition of megalin-dependent endoycytosis using His-sRAP (histidinetagged soluble receptor-associated protein). Clcn5 knock-out (KO) mice show severely impaired endocytosis of low molecular weight proteins in the PT, mimicking Dent’s disease. Normoxic Clcn5 KO kidneys lacked the prominent Fetuin A staining in PT cells (FIG. 3 B). Instead, a strong intraluminal signal was detected (FIG. 3 D), which was not present in wildtype (wt) samples (FIG. 3 C), highlighting the impaired endocytic phenotype of Clcn5 KO mice. However, hypoxic Clcn5 KO kidneys showed strong Fetuin A staining in the PT in addition to the luminal signal (FIGS. 3 F, H), providing evidence that the observed cellular Fetuin A staining genuinely originated in the PT. Double immunofluorescence staining for Fetuin A and different renal segment markers (FIGS. 3 l-L) confirmed that Fetuin A was expressed only in the PT of hypoxic fetal kidneys. Whole mount in situ hybridization corroborated Fetuin A mRNA synthesis in hypoxic fetal kidneys, but not in normoxic kidneys.

[0093] Ahsg harbours putative HIF binding sites overlapping with enhancer regions. Having shown that Fetuin A is locally produced in hypoxic fetal kidneys, it was assessed whether the expression of Ahsg was directly activated by hypoxia. To check for potential HIF binding sites (hypoxia response elements - HRE) in the human AHSG locus, HlF-1α (HIF-1-alpha) and HIF-2α (HIF-2-alpha) ChlP-seq datasets were used derived from hypoxic MCF7 cells. A cluster of potential HREs near exon 4 of human AHSG that overlapped with H3K27Ac and H3K4Me1 (chromatin marks of active enhancer elements) and DNasel hypersensitivity was identified. Another putative HRE was located in intron 1. Screening Ahsg genes of 15 species for the presence of the consensus HIF binding sites (RCGTG) 10kb upand downstream of the ATG revealed a peak 1-5 kb downstream of the ATG with an average number of 2 HREs per 1 kb window. Notably, not only the annotated human ChlP-seq HIF sites localized within this peak, but also four potential mouse HREs. Alignment of the latter with enhancer marks revealed a close association with H3K27Ac, H3K4Me1 and DNasel hypersensitivity.

[0094] Hypoxia activates Fetuin A transcription in vitro. Five putative HREs of mouse Ahsg and their surrounding DNA, alongside with nonsense mutations of these sites, were cloned into luciferase reporter plasmids (FIG. 4 A). Normal rat kidney epithelial (NRK) cells transfected with reporter plasmids containing only the putative -2kb HRE did not show increased luciferase activity under hypoxic conditions (FIG. 4 B). Conversely, NRK cells carrying reporter plasmids containing the downstream HREs significantly increased luminescence in hypoxia. No increased luciferase activity was detected when these HREs were mutated. Furthermore, there was no enhanced luciferase activity when up- and downstream HREs were combined. In summary, the HREs located downstream of the ATG confer hypoxia inducibility to the mouse Ahsg gene. Lastly, Fetuin A protein in primary mouse PT cells (pPTCs), NRK cells and human HK2 cells was detected when cultured under hypoxic conditions (FIGS. 4 C-E). Taken together, these findings identified Fetuin A as a novel, evolutionary conserved HIF-dependent target gene.

[0095] Fetuin A deficiency aggravates CKD progression in hypoxic IUGR kidneys. To investigate how the induction of Fetuin A in fetal hypoxic IUGR kidneys affects renal function in the long term, urinary protein levels were measured and the glomerular filtration rate (GFR) was determined in adult mice (FIGS. 5 A,B). GFR was reduced and proteinuria was increased in 9 weeks old mice exposed to fetal hypoxia vs. normoxia. Furthermore, assessment of fibrotic tissue remodelling revealed more collagen structures extending deeper into subcortical regions and the highest expression levels of fibrotic markers in kidneys of 9 fetal hypoxic Ahsg KO mice compared to hypoxic wt and normoxic controls, respectively (FIGS. 5 C-G). Importantly, these results clearly show that renal function was systematically more affected in Ahsg KO mice, showing an additive effect of hypoxia and Fetuin A deficiency.

[0096] Fetuin A deficiency promotes accumulation of calcium mineral particles and macrophages in hypoxic IUGR kidneys. Adult Fetuin A KO mice are prone to soft tissue calcification, however overt calcification in the kidneys of hypoxic Ahsg KO mice as assessed by von-Kossa staining was not detected. To still test whether the expression of Fetuin A in fetal hypoxic kidneys affected mineralized matrix handling for the presence of calcium containing nanoparticles was probed by incubating freshly cut kidney sections of E18.5 embryos with ATTO 488 fluorescently labelled Fetuin A (488-FA). Due to the high affinity binding of Fetuin A to calcium phosphate, 488-FA staining is more sensitive to detect calcium containing matrix and cell remnants than the commonly used mineral staining protocols. Thus, positive 488-FA staining in the absence of von-Kossa or Alizarin-Red staining also highlights structures merely enriched with calcium, including amorphous calcium-phosphate aggregates that often precede overt calcifications. 488-FA staining revealed in normoxic wt kidneys intense labelling of the PT, a site of major calcium resorption and thus also of mineralized matrix handling (FIG. 6 A). PT staining intensity was reduced in hypoxic wt and increased in hypoxic Ahsg KO, suggesting increased and reduced mineralized matrix turnover, respectively (FIGS. 6 B, C). Only hypoxic Ahsg KO kidneys also showed a granular staining pattern in the papillary region and less frequently in the nephrogenic region of the outer cortex (arrowheads in FIGS. 6 D-l), indicating bulk mineral deposition in the absence of endogenous Fetuin A. Excess bulk mineral or cellular debris is often found at sites of enhanced cell death. Indeed, TUNEL staining confirmed apoptosis in hypoxic Ahsg KO kidneys, but not in normoxic or hypoxic wt kidneys. Cleaved caspase-3 immunostaining in hypoxic Ahsg KO kidneys further corroborated cell death in these kidneys. Lastly, accumulation of mineralized material triggers an inflammatory response, leading to macrophage infiltration, tissue damage and fibrosis. Staining for the macrophage marker F4/80 revealed a significant accumulation of these cells only in hypoxic Ahsg KO vs. wt kidneys, suggesting a protective effect of Fetuin A (FIGS. 6 J-N). Collectively, these data illustrate for the first time the role of Fetuin A in binding and clearance of mineralized matrix in the kidney, protecting tissue integrity.

[0097] Fetuin A attenuates the hypoxia-induced expression of fibrosis markers by antagonising TGF-β signaling. Kidneys of fetal hypoxic Ahsg KO mice generally exhibited higher expression levels of fibrotic markers vs. fetal hypoxic wt animals (FIG. 5), despite similar mRNA levels of transforming growth factor beta-1 (Tgfb1), a potent inducer of fibrosis. These findings were elucidated in vitro, using freshly isolated pPTCs. The expression of fibrotic markers was generally enhanced in Ahsg KO pPTCs compared to wt cells and even further increased under hypoxic culture conditions (FIGS. 7 A-D, G). This hypoxia-induced increase was blunted by supplementation of Fetuin A to the culture medium (FIGS. 7 B-G). Stimulation of pPTCs with recombinant TGF-β1 resulted in robust phosphorylation of its intracellular signal transducer Smad3, which was more than three times stronger in Ahsg KO pPTCs vs. wt cells (FIGS. 7 H, l), showing that Ahsg KO cells responded more vividly to TGF-β1. In contrast, adding Fetuin A before TGF- β1 treatment decreased Smad3 phosphorylation. These findings are in line with previous reports, describing Fetuin A as a TGF-βtype ll receptor mimic and cytokine antagonist. Collectively, the results suggest that Fetuin A reduces hypoxia- 11 induced renal fibrosis by direct antagonization of TGF-β1 signalling, in addition to its protective role as a mineral chaperone.

[0098] Fetuin A significantly reduces in vivo the expression of fibrosis markers in the kidney after ischemia reperfusion injury. Next, the therapeutic potential of Fetuin A supplementation on the treatment of kidney injury was experimentally explored. To that purpose, a well-established mouse model of unilateral ischemia-reperfusion injury was used. This IRI model, as indicated by its name, best mimicks ischemia-reperfusion lesions in the kidney following cardiovascular surgery, kidney transplant or removal of renal tumors. It is straightforward and reproducible. Ischemia was induced in the left kidney by clamping the renal vessels for 30 min, the right kidney served as a control. It was first shown by accumulation of calcium mineral particles and macrophages in hypoxic lUGR kidneys that ischemia-reperfusion indeed induces deposition of calcium mineral nanoparticles in the operated kidney but not in the contralateral control kidney (FIG. 8). Additionally, treatment of IRI mice by Fetuin A vs. NaCl 0.9% showed that already at day 5, Fetuin A treatment significantly reduces specific fibrosis markers (Collagens Col1a1, Col3a1 and Col6a1) in the ischemic, injured kidney compared to animals treated with NaCl 0.9% (FIGS. 9 A, B, C). In contrast, Fetuin A supplementation has no effect on the expression of the non-collagen fibrotic markers Acta2, Fn1 or Vim (FIGS. 9 D, E, F). Therefore, Fetuin A supplementation attenuates the expression of fibrotic markers following ischemic reperfusion injury. In summary, these in vivo experiments clearly establish the role of Fetuin A as mineral chaperone safeguarding the kidney from hypoxic, resp. IR injuries.

[0099] Prophylactic administration of human Fetuin A in mice undergoing ischemia reperfusion injury (IRI) led to less tissue damage and reduced fibrotic remodeling. 900 .Math.g Fetuin A were injected intravenously 24 h and again 3 h before unilateral IRI of the left kidney (20 min ischemia time). PBS was injected in control animals. Kidneys were collected 24 h post operation, the right kidney served as control. Expression levels of early damage markers Kim1 (FIG. 10 A), Krt18 (FIG. 10 B), Krt20 (FIG. 10 C) and Lcn2 (FIG. 10 D) were shown to be reduced in IRI kidneys pretreated with Fetuin A. This already shows that pretreatment with Fetuin A alleviated the tissue damage of reperfusion injury (IRI). Additionally, the observed reduction in Vim expression (FIG. 10 G) indicated less epithelial-to-mesenchymal transition (EMT). Collagenous remodeling is a relatively late event during fibrotic remodeling, which does not play an important role 24 hours after IRI and, consequently, Col1a 1 (FIG. 10 H), Col3a1 (FIG. 10 l) and Col6a1 (FIG. 10 J) did not show an unexpected change. Arg1 (FIG. 10 K) expression was repressed in Fetuin A pretreated mice, indicating a reduced requirement for collagen synthesis. Arg1 is important for increased synthesis of collagens, since it is required for the supply of L-proline, a main component of collagens. Moreover, pretreatment with Fetuin A reduced the expression of chemokines Ccl2 (FIG. 10 L), ll-6 (FIG. 10 M) and Tnf-α (FIG. 10 N) involved in damage signaling and repair. In conclusion these experiments, in particular the early damage markers, show that prophylactic administration of human Fetuin A in mice undergoing ischemia reperfusion injury (IRI) leads to less damage and reduced fibrotic remodeling even at relatively low concentrations of Fetuin A.

[0100] Presence of exogenous Fetuin A in the renal tissue of mice after intravenous (iv) administration. No human Fetuin A was detected in PBS-injected samples (FIG. 11 A, lanes G-L), confirming the specificity of the sc-133146-HRP anti-Fetuin A antibody for human Fetuin A and not mouse Fetuin A. PBS-injected IRI kidneys (FIG. 11 A, lanes J-L) showed the highest abundance of mouse Fetuin A among all samples, indicative of a high degree of biomineralization and tissue damage in these kidneys. Kidney samples from mice injected with human Fetuin A protein showed a strong band for human Fetuin A (FIG. 11 A, lanes A-F), confirming that Fetuin A was present in the renal tissue after intravenous administration. Endogenous mouse Fetuin A was only elevated in PBS-injected IRI kidneys compared to all other samples, indicative of a high degree of biomineralization and tissue damage in these kidneys (FIG. 11 B). Human Fetuin A was not detected in PBS-injected kidney samples, confirming the specificity for human Fetuin A and not mouse Fetuin A of the DFTA00 ELISA kit from R&D systems (bio-techne) (FIG. 11 C). Fetuin A injected IRI kidneys had higher tissue levels of human Fetuin A, indicative of a higher degree of biomineralization in the ischemic kidneys compared to unoperated control kidneys (FIG. 11 C). Exogenous Fetuin A was present in the proximal tubules of mice after intravenous administration (FIGS. 11 D-F). Without being bound by any theory, it could be hypothesized that Fetuin A may be drawn from the blood to sites of tissue damage, counteracting further tissue destruction and lowering its concentration in blood. In conclusion these experiments show that that intravenous administration of Fetuin A results in an increase of Fetuin A in kidney tissue.

[0101] Time course of human Fetuin A in the serum of mice upon intravenous administration. After 1 day, the detected amount of human Fetuin A in the serum was reduced compared to 1 minute or 1 hour post injection (FIG. 12 A) and after 1 week no human Fetuin A was detectable. Mouse Fetuin A was not detected with the sc-133146-HRP anti-human-Fetuin A antibody as can be derived from the fact that PBS-injected animals gave no signals. Normalized ELISA data using the human-Fetuin-A specific DFTA00 ELISA kit showed the decline of intravenous injected human Fetuin A in mouse serum samples (FIGS. 12 B, C). Compared to the injection time point (1 min), the amount of circulating human Fetuin A was reduced to 80 % after 1 hour, and to 8.5 % after 1 day. After 1 week, human Fetuin A was not detectable anymore, similar to PBS-injected control mice (black circles). In conclusion these experiments show that to maintain the presence of human Fetuin A in circulation, the injection has to be repeated every day.

METHODS

[0102] Animals. Breeding, genotyping and all animal experiments were conducted according to the Swiss law for the welfare of animals and were approved by the local authorities (Canton of Bern BE96/11, BE105/14 and BE105/17). All mice, including Ahsg.sup.tm1Mbl mice and Clcn5.sup.tm1Gug mice were maintained on a C57BL/6 background and were housed in IVC cages with free access to chow and water and a 12 h day/night cycle.

[0103] Induction of hypoxia in pregnant mice. For timed matings, females in breeding were checked for vaginal plugs every morning and if present the time point was set to gestational day (E) 0.5. AhsgKO mice were obtained from heterozygous breeding pairs, also giving rise to heterozygous and wt littermates that were used as controls. Daily food consumption (weight difference of the food initially provided and the food remaining after 24 h) during pregnancy and the maternal weight gain from E0.5 to E18.5 was recorded. For induction of hypoxia, E13.5 pregnant mice were transferred into a hypoxic glove box (Coy Laboratory Products, Grass Lake, USA). The next day, the oxygen content was gradually lowered to 10% within 6-8 hours with intermittent pauses at 16% and 12.5% to acclimatize the animals to the increasing hypoxic conditions. An electric fan inside the chamber maintained adequate air circulation. The CO.sub.2 level was kept low by chelating excess CO.sub.2 in soda lime (Sigma, 72073) filled cartridges connected to the air circulation system. Excess humidity was absorbed by silica gel orange granulate (Sigma, 1.01969), changed every day. Mice in the caloric control group received food matching the average amount of food consumed by the mice of the hypoxia group. Pregnant hypoxic or control mice were euthanized on E18.5 and fetuses and placentas were collected, weighed and prepared for further analysis. Fetal kidneys were dissected in PBS using a Leica M80 stereoscope. For primary proximal tubular cells (pPTC) preparation, kidneys of 3-4 weeks normoxic mice were isolated. Assessment of renal functions (GFR and proteinuria) was performed in 9 weeks old fetal hypoxic or normoxic.

[0104] Induction of ischemia and administration for therapeutic treatment experiments. For experiments in FIGS. 8 and 9 mice were anesthetized with isoflurane. In deeply anesthetized animals a midline abdominal incision was conducted. Ischemia was induced in the left kidney by clamping the renal vessels for 30 min, the right kidney served as a control (FIG. 9). Immediately after surgery, mice received the first treatment of either physiological NaCl solution or bovine Fetuin A monomer (100 .Math.g/g body weight) via intraperitoneal injection. Injections were repeated daily for 4 days. Kidneys were isolated on day 5 and analyzed. NaCl solution: Natrium Chloratum «Bichsel» 0.9% (154 mM); (REF 100 0 090, Bichsel, Interlaken, CH). Fetuin A solution: 0.338 EU/ml LPS, 10 mM Na.sub.3PO.sub.4, 8 g/l NaCl (137 mM).

[0105] Induction of ischemia and administration for prophylactic treatment (prevention) experiments. For experiments in FIG. 10 mice were injected intravenously with 900 .Math.g Fetuin A 24 h and 3 h before unilateral ischemia. PBS was injected intravenously in control animals. For ischemia induction, mice were anesthetized with isoflurane. In deeply anesthetized animals a midline abdominal incision was conducted and ischemia was induced in the left kidney by clamping the renal vessels for 20 min. The right kidney served as a unoperated control. Kidneys were collected 24 h post operation. Mice for experiment in FIG. 10 had an average body weight of 22 g. PBS solution: 2.67 mM KCI, 1.47 mM KH.sub.2PO.sub.4, 137.93 mM NaCl, 8.06 mM Na.sub.2HPO.sub.4-7xH.sub.2O.

[0106] Transcutaneous assessment of glomerular filtration. The glomerular filtration rate (GFR) was determined in conscious animals as described in Schreiber, A. et al. Transcutaneous measurement of renal function in conscious mice. Am J Physiol Renal Physiol 303, F783-788 (2012). Briefly, the plasma clearance of FITC-sinistrin (Fresenius-Kabi, Ll9830076) is measured across the skin using light-emitting diodes with an emission maximum for FITC at 470 nm and a photodiode detecting the fluorescent light with a maximum sensitivity at 525 nm. The decrease in fluorescence intensity over time is then converted into GFR.

[0107] Proteinuria. Urine protein content was determined using the Bradford Assay. 3 .Math.l of urine and 150 .Math.l of 1x Bradford reagent were mixed, incubated at RT for 5 min and absorbance was measured at 595 nm. (5x Bradford reagent was prepared by dissolving 50 mg Brilliant Blue G-250 (Sigma, B-1131) in 24 ml ethanol and 50 ml 85% phosphoric acid, then adjusting the total volume to 100 ml with ultra-pure water). Urine creatinine content was determined using the Jaffe method. 10 .Math.l of 1:10 diluted urine was mixed with 100 .Math.l Creatinine reagent, incubated at RT for 10 min and absorbance was measured at 510 nm. (Creatinine reagent consisted of 10 mM picric acid and 250 mM NaOH, pH 13). Finally, the protein creatinine ratio was calculated for each sample.

[0108] Glomerular count. 100 .Math.m Z-stack images of whole-mount E18.5 kidneys stained for nephrin (R&D AF3159) were analysed with the open source image processing software Fiji (lmageJ, version 2.0.0-rc69/1.52i, https://imagej.net/Fiji). In the TrackMate v3.8.0 plugin, the Downsample LoG detector was set to 80.0 pixel for the estimated blob diameter with a 16-pixel threshold and downsampling factor 2. The number of spots per frame were added to mice, the number of glomeruli per kidney was calculated. A 100 .Math.m distance between frames was chosen to avoid double counting of identical glomeruli in consecutive images, given an average glomerular diameter of 80 .Math.m.

[0109] Microarray analysis. Total RNA from male hypoxic, normoxic and caloric control group E18.5 kidneys was isolated using the RNeasy Mini Kit (Qiagen, 74104). Only high quality RNA (RIN>8, 260/280 ratio>2, 260/230 ratio>1.8) was used for further analysis. 100 ng total RNA samples were processed with the Ambion® WT Expression Kit (4411973, life technologies) kit. 5.5 .Math.g of the cDNA was fragmented and labelled with GeneChip® WT Terminal Labeling kit (901525, Affymetrix). 2.3 .Math.g biotinylated fragments were hybridized to Affymetrix Mouse Gene 1.0 ST arrays at 45° C. for 16 h, washed and stained according to the protocol described in Affymetrix GeneChip® Expression Analysis Manual (Fluidics protocol FS450_0007). The arrays were scanned with Affymetrix GeneChip® Scanner 3000 7G and raw data was extracted from the scanned images and analysed with the Affymetrix Power Tools software package. Hybridization quality was assessed using Affymetrix Expression Console software (version 1.1.2800.28061). Normalized expression signals were calculated from Affymetrix CEL files by the Robust Multi-array Average algorithm (RMA). Differential hybridized features were identified using the R Bioconductor package “limma” that implements linear models for microarray data. P values were adjusted for multiple testing with Benjamini and Hochberg’s method to control the false discovery rate (FDR). Probe sets showing at least 1.3-fold change and a FDR<0.05 were considered significant. Differential expression values between hypoxic, normoxic and caloric control group E18.5 kidneys were mapped with Heatmapper (http://www.heatmapper.ca) using average linkage and Euclidean distance measurement. For functional annotation, GO-term analysis was performed using the DAVID platform (https://david.ncifcrf.gov).

[0110] RT-qPCR. Total RNA was isolated using TRlzol® reagent (Invitrogen 15596026) according to the manufacturer’s protocol. RNA concentration and quality was determined with a Nanodrop 1000 spectrophotometer (ThermoFisher Scientific, Switzerland) and 1000 ng were transcribed into cDNA using PrimeScript RT Reagent Kit (Takara, RR037A). cDNA was diluted to 2 ng/.Math.l and qPCR was performed with either TaqMan Gene Expression Assays (ThermoFisher) or FAM-labelled UPL probe (Roche) plus corresponding gene-specific primers and TaqMan Fast Universal PCR Master mix (Applied Biosystems, 4352042) on a 7500 Fast Real-Time PCR System (Applied Biosystems). Data analysis was performed with Microsoft Excel. The 2.sup.(-ΔCt) method was used to calculate the relative expression levels for RT-qPCR.

[0111] Identification of HIF binding sites. For the identification of putative HIF binding sites, the 20 kb genomic region overlapping with the Ahsg locus (10 kb up- and downstream of the start codon) of 15 different species (cat, chicken, chimp, cow, dog, ghost shark, horse, human, mouse, pig, rabbit, rat, sheep, xenopus, zebrafish) was analysed with the JASPAR database (http://jaspar.genereg.net). The relative profile score threshold was set to 90%. For enrichment analysis, putative sites of all species were clustered in 1kb windows. HREs with a relative score >0.9, >0.93 and >0.97 are shown. For the alignment of human HIF alpha sites identified by ChlP-seq of hypoxic MCF7 cells (cf. Schodel J, Oikonomopoulos S, Ragoussis J, Pugh CW et al. Blood 2011 Jun 9;117(23):e207-17.; Series GSE28352; https://www.ncbi.nlm.nih.gov/ geo/query/acc.cgi?acc=GSE28352) with active regulatory marks of the AHSG locus, the HIF-1-alpha and HIF-2-alpha data sets encompassing chr3: 180,000,000-190,000,000 (including the AHSG locus) were converted into BAM files using the web-based Galaxy platform (https://usegalaxy.org) and uploaded to the human assembly GRCh37/hg19 on the USCS genome browser (https://genome.ucsc.edu). To this alignment, the following data sets were added: layered chromatin marks often found near active regulatory elements of 7 cell lines (H3K27Ac and H3K4Me1, ENCODE) and open chromatin of hypoxic MCF7 cells (DNasel HS, ENCODE). For the mouse Ahsg locus, the potential HIF binding sites identified with JASPAR were aligned with data sets (DNasel HS, ENCODE/UW; H3K27Ac and H3K4Me1, ENCODE/LICR) derived from 8 weeks mouse liver, heart and kidney, showing the mean signal intensity (bar graphs, auto-scaled, log-transformed, smoothened (16 pixels)).

[0112] Molecular cloning. For luciferase assays, the 2.5 kb promoter fragment upstream of the mouse Ahsg ATG was amplified from genomic DNA (C57BL/6) using specific primers and PrimeSTAR® GXL DNA Polymerase (Takara, R050A). The 500 bp promoter fragments (wt and mutant) and the 500 bp fragment of intronic sequences (wt and mutant) were synthesized by IDTDNA (https://eu.idtdna.com/pages). The promoter fragments were inserted into the pGL3-basic-P2P-607 plasmid with Ncol and Sacl restriction enzymes (both NEB, R3193S and R3156S). Intronic fragments were inserted into the pGL3- basic vector or the pGL3-basic vector carrying the promoter fragments using BamHl and Sall restriction enzymes (both NEB, R3136S and R3138S). For in-situ hybridization, the cDNA of exons 2-5 of mouse Ahsg was obtained from IDTDNA and cloned into pBluescriptll KS— using Spel and EcoRl restriction enzymes (both NEB, R3133S and R3101S).

[0113] Luciferase assay. 24 h after transfection, cells were washed twice with PBS, lysed (250 mM KCI, 50 mM Tris/H.sub.3PO.sub.4 pH 7.8, 10% glycerol, 0.1% NP40) on ice for 20 min and centrifuged at 14000 rpm at 4° C. for 10 min. Of the supernatant 10 .Math.l were used for each reaction. Injection of reaction solutions (Luciferase: 100 .Math.l of 25 mM Tris/H.sub.3PO.sub.4 pH 7.8, 10 mM MgSO4, 2 mM ATP pH 7.5, 50 .Math.M luciferin; Renilla: 100 .Math.l of 50 mM Tris/HCI pH 7.6, 100 mM NaCl, 1 mM EDTA, 0.5 .Math.M coelenterazine) and activity measurement was performed with a Fluoroskan Ascent FL (ThermoFisher). Each sample was measured in duplicates and luciferase activity was normalized by renilla activity.

[0114] Whole-mount in situ hybridization. E18.5 kidneys were fixed in 4 % PFA, dehydrated and stored in methanol at -20° C. In situ hybridization using a digoxigenin-labelled riboprobe was performed as described in Rudloff, S. & Kemler, R. Differential requirements for beta-catenin during mouse development. Development 139, 3711-3721 (2012). Probes were generated using the DIG RNA Labeling Mix (Roche, 11175025910) and T3 or T7 RNA polymerase (both Roche, 11031163001 or 10881767001). An alkaline phosphatase-conjugated antibody was used to detect the DIG-labelled probes (Roche, 11093274910).

[0115] TUNEL staining. Fragmented DNA in apoptotic cells was detected using the Promega DeadEnd Colorimetric TUNEL System (G7360) according to the manufacturer.

[0116] Histochemistry. For immunohistochemistry, PFA-fixed, paraffin-embedded tissue sections were rehydrated and endogenous peroxidase was blocked by incubating the slides in 1.5 % H.sub.2O.sub.2 solution (0.02 M citric acid, 0.06 M Na.sub.2HPO.sub.4) at RT for 15 min in the dark. Antigen retrieval was performed by boiling in Tris-EDTA buffer pH 9 for 20 min followed by slow cool down to RT. After blocking in 2 % BSA in PBS at RT for 1 h, the sections were incubated with primary antibodies in blocking solution o/n at 4° C. Following three washing steps in PBS, the sections were incubated with HRP-conjugated secondary antibodies (mouse or rabbit: Dako EnVision+ System from Agilent (K4001 or K4003), goat: SCBT, sc-2304) for 1 h at RT. After three washing steps in PBS, the signal was developed with DAB (Agilent, K3468). The sections were counterstained with Harris haematoxylin solution (Sigma, HHS16), dehydrated and mounted using Eukitt medium (Sigma, 03989). For Picrosirius red staining of collagen, dewaxed, rehydrated tissue sections were incubated in staining solution (0.5 g Direct Red 80 in saturated aqueous solution of picric acid (both Sigma, 365548 and P6744)) for 1 h at RT. After washing twice in acidified water (0.5% glacial acetic acid), the sections were dehydrated and mounted using Eukitt.

[0117] Immunofluorescent staining. Cryosections were fixed in 4 % PFA at RT for 10 min, washed twice in PBS and permeabilised by incubation in PBST (0.1% Triton X-100 in PBS) at RT for 10 min. After blocking in 10 % FCS, 0.5 % Tween-20 in PBS at RT for 1 h, the sections were incubated with primary antibodies in blocking solution o/n at 4° C. Following three washing steps in PBS, the sections were incubated with fluorescence-conjugated secondary antibodies in blocking solution in the dark for 1 h at RT. DNA was stained with DAPI 1:5000 in PBS. Sections were mounted in MOWIOL solution (2.4 g MOWIOL 4-88 reagent (Merck, 475904) in 6 g glycerol and 18 ml 0.13 M Tris pH 8.5). For whole mount immunofluorescence staining of E18.5 kidneys, the iDisco staining protocol (https://idisco.info/idisco-protocol/) with methanol pre-treatment was applied. An incubation time n=1 day and a solution volume of 1.6 ml was used for the relevant steps. The kidneys were mounted in 8-well glass chamber slides (ThermoFisher, 154534) and imaged immediately.

[0118] Fluorescent detection of calcium. Thick cryosections (30 .Math.m) were incubated with 10 ng/ml ATTO 488 fluorescently labelled Fetuin A (in calcium-free PBS) in the dark at RT for 60 min, rinsed 3 times with PBS and mounted with MOWIOL solution. Nuclei were counterstained with DAPI.

[0119] lmaging. Fluorescence imaging was performed on a IMIC digital microscope (FEI, Type 4001) using the Polychrome V light source, an Orca-R.sup.2 camera controller from Hamamatsu (C10600) and Live Acquisition software (FEI, version 2.6.0.14). Image analysis was performed using Offline Analysis software (FEI). Bright field imaging was performed on a Nikon E600 microscope equipped with Nikon objectives (Plan Fluor ELWD 20x/0.45, Plan Apo 40x/1.0 Oil and 60x/1.40 Oil) using a Digital Sight DS-UE camera controller and DSRi1 camera (both Nikon). Image analysis was performed using Nikon software NIS Elements 4.0.

[0120] ELISA. Mouse Fetuin A levels, e.g. in serum or renal tissue, were determined using the mouse-specific Fetuin A/AHSG Quantikine ELISA Kit (R&D, MFTA00) according to the manufacturer bio-techne. Human Fetuin A levels, e.g. in serum or renal tissue, were determined using the human-specific Fetuin A/AHSG Quantikine ELISA Kit (R&D, DFTA00) according to the manufacturer bio-techne.

[0121] Cell culture. The normal rat kidney (NRK) cell line was cultured in DMEM (Gibco, 41965- 039) and 10% fetal bovine serum (FBS). The human kidney (HK-2) cell line was cultured in Keratinocyte-SFM medium (Gibco, 17005-075). For luciferase assays, NRK cells were transfected with luciferase reporter plasmids and pCMV-Renilla (10% of total transfected DNA, used for normalization) using jetPrime® reagent (Polyplus, 114-07), stimulated with 1 mM DMOG (Echelon Biosciences, F-0010) 6 h after and harvested 24 h after transfection. For hypoxia, cell culture was performed at 0.2 % oxygen for 48 h. Primary proximal tubular cells (pPTC) were isolated from 3-4 weeks old kidneys (cf. Terryn, S. et al. A primary culture of mouse proximal tubular cells, established on collagen-coated membranes. Am J Physiol Renal Physiol 293, F476-485 (2007)). Briefly, proximal tubular fragments were obtained by digesting cortical kidney tissue with collagenase and filtration through an 80 .Math.m pore size membrane. pPTCs were cultured in DMEM/F12 (Gibco, 21041-025) supplemented with 15 mM HEPES, 0.55 mM NaPyruvate, 1% NEAA and renal epithelial cell growth medium (REGM) supplements (Lonza, CC-4127). Instead of FBS, serum from Ahsg KO mice was used. Upon confluency, pPTCs were split and replated once using Accutase solution (Sigma, A6964). For 24 hypoxia, cell culture was performed at 0.2 % oxygen for 48 h. 100 .Math.g/ml Fetuin A (Sigma, F3385) or bovine serum albumin (BSA, Sigma, A3059) was added to the culture medium 48 h before the end of the experiment. Before treatment with 5 ng/ml rmTGF-β1 (R&D, 7666-MB) for 5 min, cells were starved for 24 h.

[0122] Western blot analysis. Total protein lysates were obtained using RIPA buffer (Sigma, R0278) supplemented with protease inhibitors (Roche, 11836153001). Proteins were separated by SDS-PAGE and blotted onto PVDF membranes (ThermoFisher, 88518). Upon blocking with 5 % milk in TBST, the membranes were incubated with primary antibodies at 4° C. o/n, incubation with HRP-conjugated secondary antibodies was performed at RT for 1 h. The signal was detected with ECL (GE Healthcare, RPN2106) or SuperSignal (ThermoFisher, 34076) depending on signal intensity. Densiometric analysis was performed with the open source image processing software Fiji (lmageJ, version 2.0.0-rc69/1.52i, https://imagej.net/Fiji).

[0123] Calculation of concentration for intraperitoneal injected bovine Fetuin A in the mouse. The reference blood volume of mice was 58.5 ml/kg (nc3rs.org.uk), 77-80 ml/kg (jax.org). Based on these values, 68.5 ml/kg was used for calculations. The calculated molecular weight of bovine Fetuin A (UniProtKB - P12763 (FETUA_BOVIN)) is 38.4 kDa. Thus, a concentration of 38 .Math.M in the blood would be reached, if all injected Fetuin A was in the blood or if intravenous injection of Fetuin A was performed. Furthermore, Fetuin A undergoes posttranslational modifications. Here, especially glycosylation plays a predominant role, increasing the molecular weight to 51-67 kDa. Therefore, a molecular weight of 60 kDa was used in the calculation (where 1 Da = 1 g/mol). This results in a concentration of 24.3 .Math.M in the blood of injected mice, if all injected Fetuin A was in the blood or if intravenous injection of Fetuin A was performed and based on the injected amount of 100 .Math.g/g body weight:

[00001]$100\frac{\text{mg}}{\text{kg}}\times {\left(68.5\frac{\text{ml}}{\text{kg}}\right)}^{-1}\times {\left(60\text{kDa}\right)}^{-1}=24.3\frac{\text{μ}\text{}\text{mol}}{\text{L}}$

[0124] However, intraperitoneal injections were performed, which results in lower concentrations in the blood, because Fetuin A has to be transported from the abdominal cavity into circulation. During this process, not all molecules will be absorbed, due to partial degradation in the cavity or pinocytosis and metabolization by mesothelial cells. In a study (Regoeczi et al (1989), Am J Physiol.; 256(4 Pt 1): E447-52), it was shown that intraperitoneal injection of albumin reached approximately 40 % of the dose in the blood after 2-3 hours compared to the intravenous injection of the same amount of albumin. Based on a 40 % uptake in mouse the resulting concentration after intraperitoneal injection is:

[00002]$100\frac{\text{mg}}{\text{kg}}\times {\left(68.5\frac{\text{ml}}{\text{kg}}\right)}^{-1}\times {\left(60\text{kDa}\right)}^{-1}\times 0.4=9.7\frac{\text{μ}\text{}\text{mol}}{\text{L}}$

[0125] Therefore, if 100 mg/kg body weight of Fetuin A is injected intraperitoneal, the resulting concentration in blood would be about 9.7 .Math.M. As described above, it was found that such a relatively low dose is sufficient to treat renal ischemic damages.

[0126] Calculation of concentration for intravenous injected human Fetuin A in the mouse. The reference blood volume of mice was 58.5 ml/kg (nc3rs.org.uk), 77-80 ml/kg (jax.org). Based on these values, 68.5 ml/kg was used for calculations. The calculated molecular weight of human Fetuin A (UniProtKB - P12763 (FETUA_BOVIN)) is 38.4 kDa. Furthermore, Fetuin A undergoes posttranslational modifications. Here, especially glycosylation plays a predominant role, increasing the molecular weight to 51-67 kDa. Therefore, a molecular weight of 60 kDa was used in the calculation. In the experiments for prophylactic treatment, the mice had an average body weight of 22 g. For a mouse with a body weight of 22 g the corresponding blood volume is about

[00003]$\frac{68.5\text{mL}}{1000\text{g}}\times 22\text{g = 1}\text{.51 ml}$

[0127] With an injection of 900 .Math.g of human Fetuin A this results in a concentration in blood of

[00004]$\frac{900\text{μ}\text{}\text{g}}{1.51\text{mL}}\times {\left(60\text{kDa}\right)}^{-1}=9.9\frac{\text{μ}\text{}\text{mol}}{\text{L}}$

[0128] Therefore, if 900 .Math.g of human Fetuin A is injected intravenously in mice having an average body weight of 22 g, the resulting concentration in blood would be about 9.9 .Math.M. As described above, it was found that such a relatively low dose is sufficient to prevent renal ischemic damages.

[0129] Calculation of concentration and dosing for human Fetuin A in humans. The reference blood volume for a human adult is 77 ml/kg (male) and 65 ml/kg (female), which is an average volume of 71 ml/kg. The molecular weight of human Fetuin A (UniProtKB - P02765 (FETUA_HUMAN) is generally comparable to bovine Fetuin A, i.e. also about 60 kDa (with glycosylation). For administration of 10 mg/kg (iv), 25 mg/kg (iv), 50 mg/kg (iv) of Fetuin A to a human adult the resulting concentration in human blood would be:

[00005]$10\frac{\text{mg}}{\text{kg}}\times {\left(71\frac{\text{ml}}{\text{kg}}\right)}^{-1}\times {\left(60\text{kDa}\right)}^{-1}=2.35\text{μ}\text{}\text{M}$

[00006]$25\frac{\text{mg}}{\text{kg}}\times {\left(71\frac{\text{ml}}{\text{kg}}\right)}^{-1}\times {\left(60\text{kDa}\right)}^{-1}=5.87\text{μ}\text{}\text{M}$

[00007]$50\frac{\text{mg}}{\text{kg}}\times {\left(71\frac{\text{ml}}{\text{kg}}\right)}^{-1}\times {\left(60\text{kDa}\right)}^{-1}=11.74\text{μ}\text{}\text{M}$

[0130] Similarly, it can be calculated which dose corresponds to the above mentioned concentration of 9.7 .Math.M:

[00008]$9.7\text{μ}\text{}\text{M}\times \text{71}\frac{\text{ml}}{\text{kg}}\times 60\text{kDa = 41}\text{.3}\frac{\text{mg}}{\text{kg}}$

[0131] Similarly, it can be calculated which dose corresponds to the above mentioned concentration of 9.9 .Math.M:

[00009]$9.9\text{μ}\text{}\text{M}\times \text{71}\frac{\text{ml}}{\text{kg}}\times 60\text{kDa = 42}\text{.2}\frac{\text{mg}}{\text{kg}}$

[0132] Therefore, the calculated equivalent to 100 .Math.g/g body weight of Fetuin A being injected intraperitoneal to a mouse, is about 41.3 mg/kg when injected intravenously into an adult human and the calculated equivalent to 900 .Math.g of Fetuin A being injected intravenously to a mouse with a body weight of 22 g is about 42.2 mg/kg when injected intravenously into an adult human.

[0133] Data analysis. Statistical analysis and graphs were performed with Prism 7 (https://www.graphpad.com). Two groups were compared by t-tests, multiple groups by one-way ANOVA (Tukey’s multiple comparisons test, unless specified otherwise). ****, P<0.0001; ***, P<0.001; **, P<0.01; *, P<0.05; ns, not significant; Error bars, standard deviation (SD); Whiskers, min to max.