University of Groningen
Klotho in vascular biology
Mencke, Rik
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Mencke, R. (2018). Klotho in vascular biology. Rijksuniversiteit Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
133
Chapter 4
Membrane-bound Klotho is not expressed
endogenously in healthy or uremic human
vascular tissue
R. Mencke K. Mirković G. Harms J.C. Struik J. van. Ark E. van Loon M. Verkaik M.H. de Borst C.J. Zeebregts JG Hoenderop M.G. Vervloet J.L. Hillebrands on behalf of the NIGRAM Consortium134
Abstract
Aims: Cardiovascular disease (CVD) is the leading cause of death in patients with chronic kidney disease (CKD), a disease state that is strongly associated with loss of renal and systemic (alpha-)Klotho. Reversely, murine Klotho deficiency causes marked medial calcification. It is therefore thought that Klotho conveys a vasculoprotective effect. Klotho expression in the vessel wall, however, is disputed.
Methods and Results: We assessed Klotho expression in healthy human renal donor arteries (N=9), CKD (renal graft recipient) arteries (N=10), carotid endarterectomy specimens (N=8), other elastic arteries (3 groups of N=3), and cultured human aortic smooth muscle cells (HASMCs) (3 primary cell lines), using immunohistochemistry (IHC), immunofluorescence, qRT-PCR, and Western blotting (WB). We have extensively validated anti-Klotho antibody KM2076 by comparing staining patterns to other anti-Klotho antibodies (SC-22220, SC-22218, and AF1819), competition assays with recombinant Klotho, IHC on Klotho-deficient kl/kl mouse kidney, and WB with recombinant Klotho. Using KM2076, we could not detect full-length Klotho in vascular tissues or HASMCs. On mRNA level, using primers against all four exon junctions, klotho expression could not be detected either. Fibroblast growth factor 23 (FGF23) injections in mice induced FGF23 signaling in kidney but not in aorta, indicating absence of Klotho-dependent FGF23 signaling in aorta.
Conclusions: Using several independent and validated methods, we conclude that full-length, membrane-bound Klotho is not expressed in healthy or uremic human vascular tissue
135
Introduction
Cardiovascular disease (CVD) is the most common cause of death in patients with chronic kidney disease (CKD) (1). The presence of CKD is an independent risk factor for the
development of CVD (2, 3). In CKD patients, CVD is characterized by atherosclerosis with especially pronounced medial thickening and medial calcification (Mönckeberg’s sclerosis) (4). Recent studies have demonstrated that the anti-aging protein Klotho plays a central and specific role in the etiology of CVD in CKD. Levels of Klotho decrease drastically and progressively in CKD, from the earliest CKD stage onward (5-7). Furthermore, Klotho-deficient mice exhibit the typical Mönckeberg’s sclerosis as seen in CKD (8) and uremic toxins are shown to silence Klotho by promoter hypermethylation in experimental conditions (9).
α-Klotho (in this article denoted Klotho) was initially identified in 1997 as a 130 kDa single-pass transmembrane protein that acts as an ageing suppressor in mice (8). The Klotho gene is mainly expressed in the distal convoluted and connecting tubule in the kidney and contains two internal repeats, termed KL1 and KL2, both at the exterior of cell membranes. Proteolytic cleavage sites are present between KL1 and KL2, and just above the membrane, suggesting that secreted forms may also exist after ectodomain shedding (10-12). In addition to these secreted products, a different transcription product exists as an alternatively spliced variant from the same gene, consisting of the KL1 internal repeat (12, 13). Secreted forms of Klotho can indeed be detected in serum, urine, and cerebrospinal fluid (6, 11), but also in media from KL-transfected cells (12).
Klotho-deficient mice (kl/kl mice) exhibit a phenotype that remarkably resembles human ageing, characterised by a short lifespan, atherosclerosis, extensive medial calcification, pulmonary emphysema, and osteoporosis, among many other ageing-related traits (8). Moreover, murine Klotho-overexpressing phenotypes are characterized by an extended lifespan (~20-30% longer) (14). In humans as well, Klotho serum levels are inversely correlated with age (15, 16).
The best recognized function of Klotho is its function in calcium and phosphate homeostasis, both directly on calcium channels and phosphate transporters (17, 18), and indirectly as a co-receptor, in conjunction with fibroblast growth factor receptor 1 (FGFR1), for the osteocyte-derived phosphatonin fibroblast growth factor 23 (FGF23) (19, 20). FGF23 down-regulates the vitamin D-converting enzyme 25-hydroxyvitamin D3 1-α-hydroxylase (CYP27B1), thereby also
affecting levels of active vitamin D (21-23). The best described function of FGF23 is induction of phosphaturia, by down-regulating the expression of phosphate transporter NaPi 2a in proximal tubular cells, thereby preventing phosphate re-absorption (24). FGF23 is a strong and independent predictor of dismal outcome in CKD and after renal transplantation (25), and is associated with total body calcification burden (26). The presence of a functional receptor, including local Klotho, for FGF23 in the vessel wall could be a link between the strong
136
epidemiological evidence that connects high levels of FGF23 in CKD, and increased all-cause and cardiovascular death (27, 28).
It is yet to be elucidated, however, how Klotho is involved in atherogenesis. Given the decrease of Klotho in CKD and the high incidence of Mönckeberg’s sclerosis in both CKD and in Klotho-deficient mice, the typical arterial calcification may result from a lack of a vasculoprotective effect of Klotho on VSMCs. Indeed, in experimental CKD models in mice, it has been demonstrated that reduced systemic Klotho increases VSMC calcification and atherosclerosis, which can be rescued by administration of exogenous Klotho (6) or by adenovirus-mediated Klotho gene delivery (29). It is possible that the same FGF23-activated Klotho/FGFR1 signalling takes place in VSMCs, akin to the renal mechanism, which would require full-length membrane-bound Klotho (30). Indeed, recent claims have been made in support of the hypothesis of endogenous vascular Klotho production (31-36). However, claims of absence of Klotho expression have also been reported (37-39). Since many of the described methods have not been validated and data on human vasculature are still inconclusive, we aimed to determine whether full-length Klotho expression could be detected in human kidney donor renal arteries and recipient iliac arteries (prone to Mönckeberg’s sclerosis), as well as in human atherosclerotic carotid endarterectomy plaques, and in cultured VSMCs.
Methods
Human tissue samples
To assess Klotho expression in vascular tissue, we used surgical leftover healthy kidney donor renal artery specimens (N=11), CKD graft recipient iliac artery specimens (N=10), carotid endarterectomy specimens from patients (N=4), and autopsy-derived elastic arteries (common carotid arteries (N=3), abdominal aorta (N=3), and common iliac arteries (N=3)). As a positive control and for antibody validation, we used human kidneys (N=6). Kidney tissue was derived from nephrectomized kidneys with Renal Cell Carcinoma (RCC). Renal, artery and endarterectomy tissue samples were collected at the University Medical Center Groningen, whereas healthy kidney donor renal artery and CKD graft recipient iliac artery specimens were collected at the VU University Medical Center (VUmc, Amsterdam) during living donor renal transplantation. At the University Medical Center Goningen, all tissue samples were obtained in accordance with Institutional Review Board guidelines. The study protocol was consistent with the Research Code of the University Medical Center Groningen
(https://www.umcg.nl/EN/Research/Researchers/General/ResearchCode/Pages/default.asp
x) and Dutch national ethical and professional guidelines (“Code of conduct; Dutch federation of biomedical scientific societies”; http://www.federa.org). At the VUmc, the use of vascular leftover tissue for research purposes was approved by the Institutional Review Board and
137
informed consent was obtained prior to inclusion. Use of human material was conform the Declaration of Helsinki. All tissue samples were snap-frozen in liquid nitrogen until cryosections were cut for immunohistochemistry, protein isolation, or RNA isolation. Additional kidney tissue samples were fixed in 10% formalin and embedded in paraffin until sections were cut for immunohistochemistry.
Cell culture
Primary human aortic smooth muscle cells (HASMCs) (ScienCell, CA, USA) from 3 different donors were cultured from passage 1 to passage 22 in Smooth Muscle Cell Medium (ScienCell), containing Smooth Muscle Cell Growth Supplement (ScienCell), 2% v/v fetal bovine serum (FBS, Lonza) and 1% v/v penicillin/streptomycin solution. Culture flasks, plates, and coverslips were coated with 0.0015% w/v poly-L-lysine solution (molecular weight 70,000-150,000, P4707, Sigma-Aldrich, USA) for 2 hours at 37 oC. HK-2 cells (a kind gift from Theo Borghuis)
were cultured in DMEM (Lonza), supplemented with 10% FBS, 1% P/S, and 1% L-glutamine (Lonza). Human embryonic kidney (HEK)293 cells were seeded onto fibronectin-coated cover slips (10 mm Ø) and were transiently transfected with lipofectamin® 2000 (Invitrogen, Breda, The Netherlands, 2 μl/μg DNA) for 2 days with or without 2 µg of the human klotho pEF1 vector (kind gift from Dr. Makoto Kuro-O, University of Texas Southwestern Medical School, USA). Cells were cultured at 5% v/v CO2/37 oC. For stainings, cells were fixed in 2% w/v
paraformaldehyde (in PBS) and permeabilized with 0.5% v/v Triton X-100.
Immunohistochemistry (IHC)
To assess Klotho expression histologically, we used human kidney as a positive control. We first assessed Klotho expression with IHC. Cryosections (4 µm thick) were mounted on APES-coated glass slides, dried, and fixed in acetone at RT. Formalin-fixed, paraffin-embedded sections (3 µm thick) were de-paraffinized in xylene, re-hydrated in a graded series of ethanol, and antigen retrieval was performed using pre-heated buffers in a microwave for 15 minutes. Four different anti-Klotho antibodies were tested: rat monoclonal KM2076 (Trans Genic Inc., Japan), goat polyclonal 22220 [E-21] (Santa Cruz Biotechnology, USA), goat polyclonal SC-22218 [T-19] (Santa Cruz Biotechnology, USA), and goat polyclonal AF1819 (R&D Systems, USA). For IHC, kidney and artery sections were incubated with KM2076 1:20 (batch TG030812) or 1:50 (batch TG200112), SC-22220 1:20, SC-22218 1:20, or AF1819 1:20 in 1% w/v BSA/PBS for 1 hour. Endogenous peroxidase was blocked in 0.075% v/v H2O2 (0.3% v/v for paraffin
sections) in PBS for 30 minutes. Sections were incubated with rabbit anti-rat-HRP (P0450, Dako, Denmark) or rabbit anti-goat-HRP (P0449, Dako) polyclonal secondary antibodies in 1% human serum/1% BSA/PBS for 30 minutes and with goat anti-rabbit-HRP or SignalStain Boost HRP, rabbit (Dako) tertiary antibodies in 1% v/v human serum/1% w/v BSA/PBS for 30 minutes when appropriate. The chromogen reaction was performed with 3-amino-9-ethyl carbazole
138
(AEC) in 0.03% v/v H2O2/0.02% w/v AEC/50 mM acetate for 15 minutes, followed by
hematoxylin counterstaining.
Competition assay
KM2076 and SC-22220 Klotho stainings were performed as described. In addition, the antibody was pre-incubated with a 40-fold molar excess of recombinant human Klotho (4.33 µg for KM2076, 4.00 µg for SC-22220) for 1 hour before incubation of tissue sections. As a negative binding control for specificity, Klotho peptide (Ab75022, Abcam) (unrelated to the KM2076 and SC-22220 epitopes) was used in a 200-fold molar excess. Serial sections were used in all competition assay stainings.
Immunofluorescence (IF)
Immunofluorescence double stainings were used to assess co-localization with distal tubule marker Epithelial Membrane Antigen (EMA; monoclonal mouse IgG2a antibody, clone E29, Dako), distal tubule marker calbindin (polyclonal rabbit anti-calbindin-D28K 1:300) and VSMC
marker α-Smooth Muscle Actin (α-SMA; monoclonal mouse IgG2a, antibody clone 1A4, Dako). Stainings were performed as with IHC, using donkey anti-rat Alexa Fluor 568 (Invitrogen, USA; kindly provided by Sander van Putten, Medical Biology, UMCG) or rabbit anti-rat-biotin (E0468, Dako) with Streptavidin Alexa Fluor 555 (Invitrogen, USA) as conjugates for KM2076. Rabbit anti-mouse-FITC (Dako) was used as a conjugate for E29 (anti-EMA), goat anti-mouse IgG2a Alexa Fluor 555 (Invitrogen) as a conjugate for 1A4 (anti-α-SMA), and goat anti-rabbit Alexa Fluor 555 (Invitrogen) as a conjugate for anti-calbindin-D28K. Nuclei were counterstained
with 4',6-diamidino-2-phenylindole (DAPI).
The staining patterns of KM2076 and SC-22220 were assessed in young (P3) and old (P22) cultured HASMCs, using rabbit anti-rat-HRP (P0450, Dako) or rabbit anti-goat-biotin (E0466) (after avidin and biotin blocking, Vector Labs, USA) and the Tyramide Signal Amplification (TSA) method (1:50 tetramethylrhodamine tyramide reagent (FP1015, Perkin Elmer, USA) in Amplification Diluent (FP1051, Perkin Elmer, USA) or conjugation with Streptavidin-Alexa Fluor 555 (Life Technologies), respectively. Nuclei were counterstained with DAPI. We visualized positive staining using a high-end fully motorized Zeiss AxioObserver Z1 microscope equipped with TissueFAXS Image Analysis Software (TissueGnostics, Austria).
139 Histochemistry for calcification
Alizarin Red staining was performed to visualise calcification in vascular tissue. Tissue sections were stained for 5 minutes in 2% w/v Alizarin Red solution (pH = 4.2), followed by 20 dips in 1:1 acetone:xylene and 20 dips in xylene.
Western blotting
To assess Klotho expression at the protein level, we lysed vascular and renal tissue and cell pellets in immunoprecipitation buffer (IPB) and determined the protein concentration of lysates using the Pierce BCA Protein measurement kit (Thermo Scintific, USA). We used human kidney and full-length recombinant human Klotho (R&D Systems, USA) as positive controls. A total of 20 μg of protein lysate or 1 ng recombinant human Klotho in PBS (R&D Systems, MN, USA) was mixed with 5x Laemmli sample buffer (containing β-mercaptoethanol). We cast 8% w/v gels and samples were heated for 5 minutes at 95 oC. After loading and separating the
proteins, using a voltage of 110 V for 90 minutes, the proteins were blotted on methanol-activated polyvinylidene difluoride (PVDF) membranes for 90 minutes at 300 mA. Aspecific binding sites were blocked in 5% non-fat dry milk in TBST for 1 hour, followed by incubation with primary antibodies (Klotho: 1:500 (KM2076), and β-actin: 1:2000 (SC-47778, Santa Cruz Biotechnology, CA, USA), as well as calponin (ab46794, Abcam), and α-SMA (1A4, Dako), as functional and structural phenotypical smooth muscle markers) in 5% w/v non-fat dry milk/TBST overnight at 4 oC. Incubation with secondary HRP-conjugated antibodies for 1 hour
in 1% w/v non-fat dry milk/TBST at RT was followed by detection of band signals with chemiluminescence, using SuperSignal West Femto Chemiluminescent Substrate (Thermo Scientific, USA) and the ChemiDoc MP Imaging System (Bio-Rad Laboratories, USA).
qRT-PCR
RNA isolation from tissue and cells was performed using the TRIzol method. RNA concentrations were measured on a Nanodrop (Thermo Scientific, USA). For samples with low yields, a SpeedVac was used to increase the RNA concentration. cDNA was synthesized using Superscript II, random primers, and 100 ng of RNA input. cDNA samples were diluted to a concentration of 2 ng/µl. Taqman mix was prepared with 20x gene expression assay mix (human klotho: Hs00183100_m1 (exon 1-2), Hs00934627_m1 (exon 2-3), Hs00935388_m1 (exon 3-4), and Hs00935389_m1 (exon 4-5)), Applied Biosystems, USA). For mouse tissue, the following primer/probe sets were used: Mm00502002_m1 (klotho, exons 4-5), Mm00656724_m1 (egr-1) (which is up-regulated by FGF23 signalling (19, 30)), Mm00438930_m1 (fgfr1), and Mm03950126_s1 (ywhaz). qRT-PCR was run by heating samples for 2 minutes at 50 oC, 10 minutes at 95 oC, followed by 40 cycles of 15 seconds at 95 oC and 60 seconds at 60 oC. qRT-PCR was performed on the ABI7900HT (Applied Biosystems,
140
USA) or on the LightCycler (Roche Applied Science, Germany). Undefined threshold values were set at 40. The housekeeping gene TATA-binding protein (TBP) was used for relative quantification in human samples and normal kidney cDNA was used as a positive control. Relative mRNA levels were expressed as 2-ΔCT.
Kl/kl mouse kidney tissue
kl/kl and WT mouse kidneys (40) were procured at the Radboud University Medical Center (Nijmegen, The Netherlands). Cryosections were stained as described (IHC/IF) for antibody validation purposes.
FGF23 treatment in a murine UUO model
This experiment was performed in adherence to the NIH Guide for the Care and Use of Laboratory Animals after local IACUC permission was given. We have injected 160 µg/kg of cleavage-resistant FGF23 or vehicle (PBS) intraperitoneally every 12 hours for 10 days in 8 to 9-weeks-old male C57BL/6JOlaHsd mice (N=10 for the FGF23 group, N=10 for the vehicle group). After 3 days of injections, 20 mice were subjected to left kidney unilateral ureteral obstruction (UUO) to induce hydronephrosis with a concomitant rise in FGF23 levels, followed by 7 additional days of FGF23 injections to constantly maintain high FGF23 levels and induce FGF23 signalling. A sham group (N=6) was included that received a sham operation and no FGF23 injections. All mice were anaesthetized using 1.5% of isoflurane by inhalation and adequacy of anaesthesia was monitored by testing rear foot reflexes. Analgesics (buprenorphine 0.05 mg/kg s.c.) were used immediately and 8 hours post-surgery. After 10 days of injections, all mice were sacrificed under anesthesia (ketamine 75 µg/g + dexmedetomidine 1 µg/g in 200 µl saline i.p.) by exsanguination and kidneys and aortas were excised.
Statistical analysis
Normally distributed data are presented as mean ± SD and non-normally distributed data are presented as median [range]. Normality of distributions was assessed using Q,Q plots. Differences between multiple groups were assessed with the Kruskal-Wallis test, followed by Dunn’s post-hoc test. A p value < 0.05 was considered significant. For statistical analysis, SPSS version 18.0.3 (Chicago, USA) was used.
141
Results
Patient characteristics are presented in Table 1. As shown in Table 1, we studied arteries from subjects with both normal renal function (living kidney donors) and end-stage renal disease (renal transplant recipients).
Immunohistochemistry for Klotho in vascular and renal tissue
Recipient artery and carotid endarterectomy specimens were characterized by various degrees of medial calcification, whereas healthy donor arteries were not calcified (Fig. 1). Assessing the donor renal arteries (N=6), recipient iliac arteries (N=6), and carotid endarterectomy plaques (N=8), staining with rat monoclonal anti-Klotho antibody KM2076 or goat polyclonal anti-Klotho antibody SC-22220 was absent (Fig. 1). We also assessed Klotho expression in the kidney (as a positive control), using a number of antibodies (KM2076, SC-22220, SC-22218, and AF1819), on both human and murine, frozen and formalin-fixed paraffin-embedded (FFPE) kidney tissue. KM2076 exhibited a clean staining pattern in both human (A) and murine (B) renal tubules, which was absent in vascular smooth muscle cells in intrarenal arteries (Fig. 2). SC-22220 and SC-22218 exhibited a similar
Figure 1. Immunohistochemistry for Klotho on donor arteries, recipient arteries, carotid endarterectomy (CEA) plaques. KM2076, SC-22220, and Alizarin Red staining on donor artery (N=6), recipient artery (N=6), and CAE plaque (N=8). Note that all arteries are negative when stained with KM2076 or SC-22220. “a” and “m” denote the adventitia and media, respectively. The magnification for all pictures is 200x; the bars represent 200 µm.
142
staining pattern, be it with more background staining. AF1819 stainings show that this antibody has a very low affinity for human Klotho, considering the virtually negative staining depicted in Figure 2A. However, AF1819 was shown to have a very high affinity for murine Klotho, again exhibiting a tubular staining pattern without any vascular staining (Fig. 2B). In serial sections, the tubular staining patterns of KM2076 and SC-22220 overlapped completely (Supplementary Figure 1). Conjugate control stainings for all anti-Klotho antibodies used were negative (data not shown).
Figure 2. Immunohistochemistry for Klotho on healthy human kidney, using four different anti-Klotho antibodies on human and mouse frozen sections and paraffin sections. (A) KM2076, SC-22220, SC22218, and AF1819 were used to stain healthy renal human tissues, to show the tubular and arterial staining patterns within a section, in both frozen sections and formalin-fixed paraffin-embedded (FFPE) sections. Note that KM2076 provides a clear tubular staining pattern, as do SC-22220, and SC-22218, be it with more aspecific background staining (open arrows). AF1819, however, shows hardly any immunoreactivity with human Klotho. None of the antibodies exhibited any positive staining in arteries (closed arrows). Magnification: 200x. After assessing the Klotho expression pattern in six different kidneys and after optimizing the different staining protocols for all antibodies and fixation methods, we used one kidney with representative Klotho expression for this series of stainings. (B) KM2076, SC-22220, SC22218, and AF1819 were used to stain healthy renal murine tissues, to show the tubular and arterial staining patterns within a section, in both frozen sections and FFPE sections. Note that KM2076 and AF1819 provide a clear tubular staining pattern, as do SC-22220, and SC-22218, be it with more aspecific background staining (open arrowheads). None of the antibodies exhibited any positive staining in arteries (closed arrowheads). After assessing the Klotho expression pattern in three different kidneys and after optimizing the different staining protocols for all antibodies and fixation methods, we used two kidneys with representative Klotho expression for this series of stainings. Magnification: 400x.
143
Figure 3. Immunofluorescence for Klotho on healthy human kidney and cultured cells. (A) Double labeling for Klotho and Epithelial Membrane Antigen (EMA), and Klotho and α-smooth muscle actin (α-SMA) with immunofluorescence on healthy human kidney sections. Some distal convoluted and connecting tubules express both Klotho and EMA, but Klotho and α-SMA never co-localize. Closed arrows denote KM2076-positive tubules and open arrow denotes arterioles. “g” indicates a glomerulus and “m” indicates the arterial media. Magnification: 400x; inset 630x. Having selected a healthy human kidney with representative Klotho expression in previous experiments, we used this kidney for the purpose of a double staining with EMA and α-SMA. (B) Immunofluorescence using KM2076 and SC-22220 on cultured HK-2 cells and vascular smooth muscle cells (VSMCs), using both young (P3) and old (P22) passages and showing no positive staining in VSMCs (magnification: 630x). Stainings were repeated three times. (C) Immunofluorescence for Klotho with KM2076 on cultured HEK cells transfected with a Klotho construct, and non-transfected controls. Only transfected cells show immunoreactivity. Magnification: 400x.
144
Immunofluorescence for Klotho in vascular and renal tissue
To characterise the staining pattern of KM2076, we performed double labelling of Klotho and cell-type specific markers, clearly showing co-localization with Epithelial Membrane Antigen (EMA), in distal convoluted and connecting tubules, but not with smooth muscle cell marker α-SMA (Fig. 3, A).
Immunofluorescence for Klotho in cultured HASMCs
We have previously shown Klotho expression with KM2076 in vitro in a renal cortical epithelial cell subset (41). In our experiments, we used HK-2 cells as a positive control, using KM2076 and SC-22220, confirming expression using PCR (Supplementary Figure 2) and confirming KM2076 staining specificity with HEK cells transfected with a Klotho construct. To confirm the VSMC-specific observation regarding the lack of Klotho positivity in human tissues with KM2076 and SC-22220, we also stained cultured HASMCs, which were negative both in young (P3) and old (P22) passages, as shown in Figure 3, B.
Western blotting for Klotho
To confirm our results obtained using antibody KM2076 for immunohistochemistry and immunofluorescence and to validate its specificity, we next performed Western blot experiments on renal and arterial protein lysates as well as on recombinant human Klotho. We were able to detect both recombinant human Klotho and renal Klotho with KM2076 on Western blot at 130 kDa, whereas no expression of Klotho in donor arteries (N=5, D1-5), recipient arteries (N=5, R1-5), or endarterectomy plaques (N=5, P1-5) was detected (Fig. 4). Also, cultured HASMCs (N=2) did not express Klotho, as assessed using antibody KM2076 (Fig. 4). All samples did express smooth muscle cell markers calponin and α-SMA. Antibody SC-22220 was also specific for Klotho on Western blot, but its sensitivity was lower as compared to KM2076 (Supplementary Figure 3). Full Western blots for Klotho are shown in Supplementary Figure 4.
qRT-PCR
Klotho mRNA was clearly and variably expressed in healthy human renal tissue (Fig. 5, A-D). This variation might be due to differences in amounts of cortical and medullary areas in the tissue fragments used for RNA analysis. Using highly sensitive primer/probe sets against all
145
four exon junctions, we detected extremely low expression of klotho mRNA in donor arteries (N=6), recipient arteries (N=10), carotid endarterectomy plaques (N=4), or other elastic arteries (N=9) (Fig. 5, A-D). The number of samples used was dependent on RNA yield and housekeeping gene expression. HASMCs repeatedly yielded undefined klotho CT values
(below the detection limit of the assay used) with all three primary cell lines (N=2/cell line). To confirm VSMC presence in the arteries under investigation, we detected abundant α-SMA expression in all samples (Fig. 5, E). The primer/probe sets showed a comparable pattern across different kidney samples (Fig. 5, F). These data indicate that klotho mRNA expression in vascular tissue is extremely low, and do not support substantial expression of full-length Klotho by VSMCs.
Antibody validation
In addition to comparing different antibody staining patterns, showing specificity for recombinant human Klotho, and showing positive staining Klotho only in transfected HEK cells and not in non-transfected control cells (Figure 3, C), we performed a competition assay by pre-incubating antibodies with recombinant human Klotho or an unrelated Klotho peptide (Ab75022) not used as the immunogen for KM2076 or SC-22220. KM2076 staining was Figure 4. Western blot analysis using KM2076 on recombinant human Klotho and on renal, vascular, and HASMC protein lysates. We used KM2076 on donor renal arteries (D), recipient arteries (R), carotid endarterectomy (CEA) plaques (P), and human aortic smooth muscle cells (HASMCs). Normal human kidney lysate and recombinant human Klotho (rec huKL) served as positive controls, revealing 130 kDa full-length Klotho. β-Actin (42 kDa, housekeeping protein) was used for as a loading control and calponin and α-SMA were used as a functional and as a structural smooth muscle cell marker, respectively. Numbers indicate individual vascular samples.
146
blocked by pre-incubation with recombinant human Klotho, but not by pre-incubation with Ab75022 or without pre-incubation (Fig. 6, A). Similarly, the SC-22220 staining pattern was blocked by recombinant human Klotho, but was otherwise present (Fig. 6, A). Furthermore, staining of kl/kl and wild-type (WT) kidney tissue showed positive staining in the distal convoluted tubules in WT kidney, but not in kl/kl kidney (Fig. 6, B). To show that the distal convoluted tubules in kl/kl kidney do not express detectable amounts of Klotho, we have performed double labelling with distal convoluted tubule marker calbindin-D28K, showing
co-localization in WT kidney, but single calbindin-D28K-positive tubules in kl/kl kidney (Fig. 6, C).
Taken together, these data indicate that both KM2076 and SC-22220 are specific for full-length membrane-bound Klotho.
Figure 5. qRT-PCR for klotho on renal and vascular tissue and HASMCs using primer/probe sets for all exon junctions. (A) Expression of klotho mRNA containing the junction of exons 1 and 2 compared to TATA-binding protein (TBP) mRNA expression, (B) expression of klotho mRNA containing the junction of exons 2 and 3 compared to TBP mRNA expression, (C) expression of klotho mRNA containing the junction of exons 3 and 4 compared to TBP mRNA expression, (D) expression of klotho mRNA containing the junction of exons 4 and 5 compared to TBP mRNA expression, (E) α-smooth muscle actin (α-SMA) mRNA expression relative to TBP mRNA expression, (F) correlation between the different klotho primer sets across kidneys, normalised to kidney 1 expression values. Standard deviations are derived from variance in triplicate measurements. Klotho mRNA is expressed abundantly and variably in human kidney, whereas it is virtually absent in human vascular structures. All arteries express the smooth muscle cell marker α-SMA. Data are presented as 2-ΔCT (A-E) or as fold change (F)
and error bars denote the standard deviation.
147
Figure 6. Validation of KM2076 and SC-22220 anti-Klotho antibodies. (A) Competition assay with KM2076 and SC-22220 using recombinant human Klotho (40-fold molar excess) and unrelated Klotho peptide Ab75022 (200-fold molar excess) on serially cut human kidney sections. Note that both KM2076 and SC-22220 staining patterns are specific for Klotho, since no antibody binding is observed after pre-incubation with recombinant human Klotho, but not after pre-incubation with a peptide not recognised by KM2076 or SC-22220. Arrowheads indicate an area of positive tubules. Bars represent 200 µm. Having selected a healthy human kidney with representative Klotho expression in previous experiments, we used this kidney in these experiments. Competition assays with recombinant Klotho were performed repeated three times. (B) Validation of KM2076 and SC-22220 antibody staining patterns using kl/kl and wild-type (WT) mouse kidney tissue. Note that the positive tubules are negative in the kl/kl mouse kidneys. Arrows indicate positive distal convoluted tubules in IHC stainings. Bars represent 100 µm. Murine kidney stainings with SC-22220 and KM2076 were performed two or three times independently. (C) Double labeling for KM2076 and calbindin-D28K on WT mouse kidney, indicating that calbindin-D28K-positive
tubules are negative in kl/kl mouse kidney. Magnification: 400x. The double staining for Klotho and calbindin-D28K
was performed on WT and kl/kl kidneys previously stained for Klotho and after previously asserting calbindin-D28K protein expression on IHC.
148
FGF23 signaling is present in kidney and absent in aorta
In a murine unilateral ureteral obstruction (UUO) model with additional FGF23 injections (N=10), significant up-regulation of Early Growth Response protein (EGR) 1 mRNA was found in the contralateral (non-obstructed) kidney as compared to the vehicle group (N=10), Kruskal-Wallis: p = 0.002 (Fig. 7, A). This up-regulation was absent in FGF23-treated aortas (Fig. 7, B), Kruskal-Wallis: p = 0.427. This indicates that FGF23 signalling is present in kidney, but absent in aorta. Fgfr1 mRNA was significantly up-regulated in both UUO groups in kidney (Kruskal-Wallis: p = 0.003), but not in aorta (Kruskal-(Kruskal-Wallis: p = 0.205) (Fig. 7, C, D). Klotho was expressed similarly in all contralateral kidneys (Kruskal-Wallis: p = 0.106) (Fig. 7, E). Klotho expression was barely detectable in aorta, at levels several thousand-fold lower than in kidneys (Fig. 7, F) and given the lack of egr-1 response in aorta, not functional as such.
Discussion
Based on elaborate attempts to demonstrate the local production or presence of Klotho in the human arterial wall, we conclude that there is no full-length Klotho expression in all human arterial tissue studied. The absence of the Klotho protein was demonstrated by negative immunohistochemistry and immunofluorescence, Western blotting, and was supported by extremely low klotho mRNA expression by qRT-PCR, using several independent and multi-validated methods. Our findings also render unlikely the possibility that the human vasculature possesses Klotho as a component of the classical receptor for FGF23 (19). However, it is still possible that Klotho is expressed in segmented fashion in arteries that we have not investigated.
Nonetheless, our results hold implications with regard to the hypothesized the canonical route of FGF23 signal transduction in the vessel wall, through membrane-bound Klotho in conjunction with FGFR1c, in FGF23 vasculotoxicity or vasculoprotection. Moreover, in the absence of functional membrane-bound Klotho, the potential protective effects of Klotho on endothelial cells or on the process of vascular calcification may not be the result of autocrine or paracrine effects of locally cleaved membrane-bound Klotho.
(Figure 6. Cont’d.) indicating that calbindin-D28K-positive tubules are negative in kl/kl mouse kidney.
Magnification: 400x. The double staining for Klotho and calbindin-D28K was performed on WT and kl/kl kidneys
149
Figure 7. Fibroblast Growth Factor (FGF) 23 signaling is present in the contralateral kidney but not in aorta in a unilateral ureteral obstruction (UUO) model. (A) Egr-1/ywhaz mRNA ratio in mouse kidney, showing significant up-regulation of Egr-1 mRNA after FGF23 treatment. (B) Egr-1/ywhaz mRNA ratio in mouse aorta, showing no differences in expression. (C) FGF receptor 1 (fgfr1)/ywhaz mRNA ratio in mouse kidney, showing significant up-regulation in the UUO groups. (D) Fgfr1/ywhaz mRNA ratio in mouse aorta, showing no differences. (E) Klotho/ywhaz mRNA ratio in mouse kidney, showing presence of Klotho in kidney, but no differences among groups. (F) Klotho/ywhaz mRNA ratio in mouse aorta. Expression was barely detectable at levels several thousand-fold lower than in kidney. ** p < 0.01. N=6 for the sham group, N=10 for the UUO control group, and N=10 for the UUO + FGF23 group.
150
If membrane-bound Klotho is not produced in the vasculature itself, one alternative hypothesis may be that the supposed protective properties of Klotho are conveyed to vessels by circulating soluble Klotho. This is supported by the findings of Hu et al., who showed experimentally that by administering exogenous Klotho, vascular calcification could be attenuated in vitro (6). Chen et al. have also shown that intraperitoneal injection of Klotho protein ameliorates vascular calcification in vivo in Klotho-deficient mice (42). The disproval of the hypothesis of endogenous vascular full-length, membrane-bound Klotho production holds important implications with regard to the manner in which the vasculature can be targeted to ameliorate or prevent the development of atherosclerosis or medial calcification (Mönckeberg’s sclerosis) in CKD patients. The possibility that systemic Klotho may convey vascular protection implies dependence on a distant Klotho source, for which the kidney is the most likely candidate, although this is also debated (38, 43). As CKD progresses, the renal Klotho source is exhausted, possibly exacerbating vascular disease. Methods to up-regulate Klotho expression in the kidney could therefore be promising interventions (44).
Even though we show that there is no FGF23-induced Klotho/FGFR1 signaling in the aorta, our results cannot completely rule out the possibility that FGF23 still has pathological effects on the vessel wall. It was shown recently that FGF23 effects in the heart and in the parathyroid gland can be mediated by a Klotho-independent pathway, using a different intracellular cascade (45, 46).
Many authors have previously published data addressing the existence of vascular Klotho expression, yielding very different results. Our data strongly refute the presence of functional membrane-bound Klotho in the vasculature, which is in line with seven studies, in which possible Klotho expression in human and mouse arteries was thoroughly investigated (37-39, 47-50). However, thirteen studies have been published opposing our conclusion, showing data indicative of vascular Klotho expression (8, 31-36, 51-56). With these disparities in mind, to comprehensively address this research question about which so many reports detail different findings, we have invested abundant effort in validating the detection methods for full-length Klotho mRNA and protein.
First of all, it is important to note that published findings on vascular Klotho mRNA are, in fact, not at all wholly discordant. At the mRNA level, both in human and in mouse, we find an extremely low amount of Klotho mRNA. All of the studies referenced above that describe Klotho mRNA expression, do so at similarly low levels or converted to arbitrary units, except for a few studies that show data that are negative even at the mRNA level, although one of these studies employed Northern blotting, as opposed to RT-PCR as performed in other studies (47-50). Akin to our data, Lindberg et al. (37) also describe murine Klotho mRNA levels in the vasculature around several thousand-fold lower than in renal tissue. It is likely that there is some Klotho expression at the mRNA level in the vasculature, although the expression level is around the detection limit of most assays.
151
At the protein level, however, the available data are less easily reconciled. In four studies, Klotho expression in vivo and in vitro could not be detected in vascular smooth muscle (37, 38, 49, 50). To that end, they used KM2076, KM2119, and an unspecified R&D antibody. Both KM2076 and KM2119 have been shown numerous times to be specific for renal 130 kDa Klotho, which is corroborated by our data. On the other hand, eight studies contain data on vascular Klotho at the protein level (32-35, 53-56). To that end, they used Ab75023, an unspecified Abcam antibody, AF1819, Immutopics and Genprice anti-Klotho antibodies, another unspecified anti-Klotho antibody, and ABIN502138, in addition to KM2076 and SC-22220, which were also used in our study. A number of observations is to be made with regard to these data. First of all, if specified, immunoreactivity between alleged vascular Klotho and anti-Klotho antibodies has always yielded a band of around 110-116 kDa (Ab75023, an unspecified Abcam antibody, and an unspecified antibody) (33, 34, 55), which is unlikely to be the same protein as the 130 kDa Klotho protein found in the kidney. Furthermore, three studies, using AF1819 on murine aorta and rat IgG2a KM2076 on rat aorta, show immunoreactivity with proteins of unspecified weight (32, 35, 53). It is interesting to note that AF1819 apparently exhibits immunoreactivity with renal Klotho (as seen in our data and in supplemental data from Fang et al. (32, 53)), however, this renal staining appears to be mutually exclusive with the vascular staining pattern that they show, rendering it unlikely that the vascular and tubular staining patterns pertain to the same protein. Differences in methods may account for our inability to detect a vascular staining pattern with AF1819. In one study, the authors describe Klotho mRNA expression in VSMCs and in aorta and show immunoreactivity in rat aorta, but fail to identify Klotho protein in VSMCs (35). This may be due to specific culture conditions or a de-differentiation effect of cell culture, but it is also possible that VSMCs are negative for Klotho protein. Lastly, the positive staining patterns described by various authors are vastly different, ranging from a very distinct medial staining pattern (32, 53) to a more vague medial staining pattern (33, 35, 56) to exclusive staining in locations of calcification (34, 54), without any normal medial staining. Vascular immunoreactivity with anti-Klotho antibodies as reported by others may concern a protein related to or unrelated to Klotho, but it is not membrane-bound Klotho as found in the kidney, nor is it functional as such as an FGF23 co-receptor.
In conclusion, having validated the specificity of our antibodies and having applied several independent methods, our elaborate attempts to demonstrate full-length Klotho gene and protein expression in the human arterial wall have yielded negative results in healthy, atherosclerotic, and uremic arterial tissue, and in HASMCs. Although FGF23 signaling in the arterial wall does not occur through Klotho, our observations do not exclude a protective effect of circulating, soluble Klotho on the arterial wall.
152
Funding
This work was supported by a consortium grant from the Dutch Kidney Foundation [NIGRAM, CP10.11] and the UMCG GSMS MD/PhD program. Fluorescence imaging was performed at the UMCG Imaging Center, supported by the Netherlands Organisation for Health Research and Development [ZonMW grant 40-00506-98-9021].
Acknowledgments
We would like to thank K. Yeung and W. Wisselink (VU University Medical Center, Department of Vascular Surgery) for tissue collection. We would like to thank G. Krenning (Medical Biology, UMCG) for calponin antibody. We are also grateful to Pieter Klok, Marian Bulthuis, and Sippie Huitema for their technical assistance.
153
References
1. A. S. Go, G. M. Chertow, D. Fan, C. E. McCulloch, C. Y. Hsu, Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N. Engl. J. Med. 351, 1296-1305 (2004).
2. G. Manjunath, H. Tighiouart, H. Ibrahim, B. MacLeod, D. N. Salem, J. L. Griffith, J. Coresh, A. S. Levey, M. J. Sarnak, Level of kidney function as a risk factor for atherosclerotic cardiovascular outcomes in the community. J. Am. Coll. Cardiol. 41, 47-55 (2003).
3. N. I. Parikh, S. J. Hwang, M. G. Larson, D. Levy, C. S. Fox, Chronic kidney disease as a predictor of cardiovascular disease (from the Framingham Heart Study). Am. J. Cardiol. 102, 47-53 (2008).
4. U. Schwarz, M. Buzello, E. Ritz, G. Stein, G. Raabe, G. Wiest, G. Mall, K. Amann, Morphology of coronary atherosclerotic lesions in patients with end-stage renal failure. Nephrol. Dial. Transplant. 15, 218-223 (2000). 5. N. Koh, T. Fujimori, S. Nishiguchi, A. Tamori, S. Shiomi, T. Nakatani, K. Sugimura, T. Kishimoto, S. Kinoshita, T. Kuroki, Y. Nabeshima, Severely reduced production of klotho in human chronic renal failure kidney. Biochem.
Biophys. Res. Commun. 280, 1015-1020 (2001).
6. M. C. Hu, M. Shi, J. Zhang, H. Quinones, C. Griffith, M. Kuro-o, O. W. Moe, Klotho deficiency causes vascular calcification in chronic kidney disease. J. Am. Soc. Nephrol. 22, 124-136 (2011).
7. Y. Shimamura, K. Hamada, K. Inoue, K. Ogata, M. Ishihara, T. Kagawa, M. Inoue, S. Fujimoto, M. Ikebe, K. Yuasa, S. Yamanaka, T. Sugiura, Y. Terada, Serum levels of soluble secreted alpha-Klotho are decreased in the early stages of chronic kidney disease, making it a probable novel biomarker for early diagnosis. Clin. Exp. Nephrol. 16, 722-729 (2012).
8. M. Kuro-o, Y. Matsumura, H. Aizawa, H. Kawaguchi, T. Suga, T. Utsugi, Y. Ohyama, M. Kurabayashi, T. Kaname, E. Kume, H. Iwasaki, A. Iida, T. Shiraki-Iida, S. Nishikawa, R. Nagai, Y. I. Nabeshima, Mutation of the mouse klotho gene leads to a syndrome resembling ageing. Nature. 390, 45-51 (1997).
9. C. Y. Sun, S. C. Chang, M. S. Wu, Suppression of Klotho expression by protein-bound uremic toxins is associated with increased DNA methyltransferase expression and DNA hypermethylation. Kidney Int. 81, 640-650 (2012). 10. L. Bloch, O. Sineshchekova, D. Reichenbach, K. Reiss, P. Saftig, M. Kuro-o, C. Kaether, Klotho is a substrate for alpha-, beta- and gamma-secretase. FEBS Lett. 583, 3221-3224 (2009).
11. A. Imura, A. Iwano, O. Tohyama, Y. Tsuji, K. Nozaki, N. Hashimoto, T. Fujimori, Y. Nabeshima, Secreted Klotho protein in sera and CSF: implication for post-translational cleavage in release of Klotho protein from cell membrane. FEBS Lett. 565, 143-147 (2004).
12. C. D. Chen, S. Podvin, E. Gillespie, S. E. Leeman, C. R. Abraham, Insulin stimulates the cleavage and release of the extracellular domain of Klotho by ADAM10 and ADAM17. Proc. Natl. Acad. Sci. U. S. A. 104, 19796-19801 (2007).
154
13. Y. Matsumura, H. Aizawa, T. Shiraki-Iida, R. Nagai, M. Kuro-o, Y. Nabeshima, Identification of the human klotho gene and its two transcripts encoding membrane and secreted klotho protein. Biochem. Biophys. Res.
Commun. 242, 626-630 (1998).
14. H. Kurosu, M. Yamamoto, J. D. Clark, J. V. Pastor, A. Nandi, P. Gurnani, O. P. McGuinness, H. Chikuda, M. Yamaguchi, H. Kawaguchi, I. Shimomura, Y. Takayama, J. Herz, C. R. Kahn, K. P. Rosenblatt, M. Kuro-o, Suppression of aging in mice by the hormone Klotho. Science. 309, 1829-1833 (2005).
15. Y. Yamazaki, A. Imura, I. Urakawa, T. Shimada, J. Murakami, Y. Aono, H. Hasegawa, T. Yamashita, K. Nakatani, Y. Saito, N. Okamoto, N. Kurumatani, N. Namba, T. Kitaoka, K. Ozono, T. Sakai, H. Hataya, S. Ichikawa, E. A. Imel, M. J. Econs, Y. Nabeshima, Establishment of sandwich ELISA for soluble alpha-Klotho measurement: Age-dependent change of soluble alpha-Klotho levels in healthy subjects. Biochem. Biophys. Res. Commun. 398, 513-518 (2010).
16. R. D. Semba, A. R. Cappola, K. Sun, S. Bandinelli, M. Dalal, C. Crasto, J. M. Guralnik, L. Ferrucci, Plasma klotho and mortality risk in older community-dwelling adults. J. Gerontol. A Biol. Sci. Med. Sci. 66, 794-800 (2011). 17. Q. Chang, S. Hoefs, A. W. van der Kemp, C. N. Topala, R. J. Bindels, J. G. Hoenderop, The beta-glucuronidase klotho hydrolyzes and activates the TRPV5 channel. Science. 310, 490-493 (2005).
18. M. C. Hu, M. Shi, J. Zhang, J. Pastor, T. Nakatani, B. Lanske, M. S. Razzaque, K. P. Rosenblatt, M. G. Baum, M. Kuro-o, O. W. Moe, Klotho: a novel phosphaturic substance acting as an autocrine enzyme in the renal proximal tubule. FASEB J. 24, 3438-3450 (2010).
19. I. Urakawa, Y. Yamazaki, T. Shimada, K. Iijima, H. Hasegawa, K. Okawa, T. Fujita, S. Fukumoto, T. Yamashita, Klotho converts canonical FGF receptor into a specific receptor for FGF23. Nature. 444, 770-774 (2006). 20. H. Kurosu, Y. Ogawa, M. Miyoshi, M. Yamamoto, A. Nandi, K. P. Rosenblatt, M. G. Baum, S. Schiavi, M. C. Hu, O. W. Moe, M. Kuro-o, Regulation of fibroblast growth factor-23 signaling by klotho. J. Biol. Chem. 281, 6120-6123 (2006).
21. H. Tsujikawa, Y. Kurotaki, T. Fujimori, K. Fukuda, Y. Nabeshima, Klotho, a gene related to a syndrome resembling human premature aging, functions in a negative regulatory circuit of vitamin D endocrine system.
Mol. Endocrinol. 17, 2393-2403 (2003).
22. T. Shimada, M. Kakitani, Y. Yamazaki, H. Hasegawa, Y. Takeuchi, T. Fujita, S. Fukumoto, K. Tomizuka, T. Yamashita, Targeted ablation of Fgf23 demonstrates an essential physiological role of FGF23 in phosphate and vitamin D metabolism. J. Clin. Invest. 113, 561-568 (2004).
23. T. Nakatani, B. Sarraj, M. Ohnishi, M. J. Densmore, T. Taguchi, R. Goetz, M. Mohammadi, B. Lanske, M. S. Razzaque, In vivo genetic evidence for klotho-dependent, fibroblast growth factor 23 (Fgf23) -mediated regulation of systemic phosphate homeostasis. FASEB J. 23, 433-441 (2009).
24. Vervloet MG, Larsson TE, Fibroblast growth factor-23 and Klotho in chronic kidney disease. Kidney
international.Supplement. 1, 130-135 (2011).
25. L. C. Baia, J. K. Humalda, M. G. Vervloet, G. Navis, S. J. Bakker, M. H. de Borst, NIGRAM Consortium, Fibroblast growth factor 23 and cardiovascular mortality after kidney transplantation. Clin. J. Am. Soc. Nephrol. 8, 1968-1978 (2013).
155
26. M. A. Mirza, T. Hansen, L. Johansson, H. Ahlstrom, A. Larsson, L. Lind, T. E. Larsson, Relationship between circulating FGF23 and total body atherosclerosis in the community. Nephrol. Dial. Transplant. 24, 3125-3131 (2009).
27. J. Kendrick, A. K. Cheung, J. S. Kaufman, T. Greene, W. L. Roberts, G. Smits, M. Chonchol, HOST Investigators, FGF-23 associates with death, cardiovascular events, and initiation of chronic dialysis. J. Am. Soc. Nephrol. 22, 1913-1922 (2011).
28. T. Isakova, H. Xie, W. Yang, D. Xie, A. H. Anderson, J. Scialla, P. Wahl, O. M. Gutierrez, S. Steigerwalt, J. He, S. Schwartz, J. Lo, A. Ojo, J. Sondheimer, C. Y. Hsu, J. Lash, M. Leonard, J. W. Kusek, H. I. Feldman, M. Wolf, Chronic Renal Insufficiency Cohort (CRIC) Study Group, Fibroblast growth factor 23 and risks of mortality and end-stage renal disease in patients with chronic kidney disease. JAMA. 305, 2432-2439 (2011).
29. T. Shiraki-Iida, A. Iida, Y. Nabeshima, H. Anazawa, S. Nishikawa, M. Noda, M. Kuro-o, Y. Nabeshima, Improvement of multiple pathophysiological phenotypes of klotho (kl/kl) mice by adenovirus-mediated expression of the klotho gene. J. Gene Med. 2, 233-242 (2000).
30. E. G. Farrow, S. I. Davis, L. J. Summers, K. E. White, Initial FGF23-mediated signaling occurs in the distal convoluted tubule. J. Am. Soc. Nephrol. 20, 955-960 (2009).
31. J. Donate-Correa, C. Mora-Fernandez, R. Martinez-Sanz, M. Muros-de-Fuentes, H. Perez, B. Meneses-Perez, V. Cazana-Perez, J. F. Navarro-Gonzalez, Expression of FGF23/KLOTHO system in human vascular tissue. Int. J. Cardiol. 165, 179-183 (2013).
32. Y. Fang, C. Ginsberg, T. Sugatani, M. C. Monier-Faugere, H. Malluche, K. A. Hruska, Early chronic kidney disease-mineral bone disorder stimulates vascular calcification. Kidney Int. 85, 142-150 (2014).
33. K. Lim, T. S. Lu, G. Molostvov, C. Lee, F. Lam, D. Zehnder, L. L. Hsiao, Vascular Klotho Deficiency Potentiates the Development of Human Artery Calcification and Mediates Resistance to FGF-23. Circulation. 125, 2243-2255 (2012).
34. D. Zhu, N. C. Mackenzie, J. L. Millan, C. Farquharson, V. E. MacRae, A protective role for FGF-23 in local defence against disrupted arterial wall integrity? Mol. Cell. Endocrinol. 372, 1-11 (2013).
35. R. Jimbo, F. Kawakami-Mori, S. Mu, D. Hirohama, B. Majtan, Y. Shimizu, Y. Yatomi, S. Fukumoto, T. Fujita, T. Shimosawa, Fibroblast growth factor 23 accelerates phosphate-induced vascular calcification in the absence of Klotho deficiency. Kidney Int. (2013).
36. J. F. Navarro-Gonzalez, J. Donate-Correa, M. Muros de Fuentes, H. Perez-Hernandez, R. Martinez-Sanz, C. Mora-Fernandez, Reduced Klotho is associated with the presence and severity of coronary artery disease. Heart. 85, 1103-1111 (2014).
37. K. Lindberg, H. Olauson, R. Amin, A. Ponnusamy, R. Goetz, R. F. Taylor, M. Mohammadi, A. Canfield, K. Kublickiene, T. E. Larsson, Arterial klotho expression and FGF23 effects on vascular calcification and function.
PLoS One. 8, e60658 (2013).
38. W. L. Lau, E. M. Leaf, M. C. Hu, M. M. Takeno, M. Kuro-O, O. W. Moe, C. M. Giachelli, Vitamin D receptor agonists increase klotho and osteopontin while decreasing aortic calcification in mice with chronic kidney disease fed a high phosphate diet. Kidney Int. 82, 1261-1270 (2012).
156
39. J. J. Scialla, W. L. Lau, M. P. Reilly, T. Isakova, H. Y. Yang, M. H. Crouthamel, N. W. Chavkin, M. Rahman, P. Wahl, A. P. Amaral, T. Hamano, S. R. Master, L. Nessel, B. Chai, D. Xie, R. R. Kallem, J. Chen, J. P. Lash, J. W. Kusek, M. J. Budoff, C. M. Giachelli, M. Wolf, Chronic Renal Insufficiency Cohort Study Investigators, Fibroblast growth factor 23 is not associated with and does not induce arterial calcification. Kidney Int. 83, 1159-1168 (2013). 40. T. E. Woudenberg-Vrenken, B. C. van der Eerden, A. W. van der Kemp, J. P. van Leeuwen, R. J. Bindels, J. G. Hoenderop, Characterization of vitamin D-deficient klotho(-/-) mice: do increased levels of serum 1,25(OH)2D3 cause disturbed calcium and phosphate homeostasis in klotho(-/-) mice? Nephrol. Dial. Transplant. 27, 4061-4068 (2012).
41. J. van Ark, H. P. Hammes, M. C. van Dijk, M. G. Vervloet, B. H. Wolffenbuttel, H. van Goor, J. L. Hillebrands, Circulating alpha-klotho levels are not disturbed in patients with type 2 diabetes with and without macrovascular disease in the absence of nephropathy. Cardiovasc. Diabetol. 12, 116 (2013).
42. T. H. Chen, M. Kuro-O, C. H. Chen, Y. M. Sue, Y. C. Chen, H. H. Wu, C. Y. Cheng, The secreted Klotho protein restores phosphate retention and suppresses accelerated aging in Klotho mutant mice. Eur. J. Pharmacol. 698, 67-73 (2013).
43. T. Akimoto, T. Kimura, Y. Watanabe, N. Ishikawa, Y. Iwazu, O. Saito, S. Muto, T. Yagisawa, E. Kusano, The impact of nephrectomy and renal transplantation on serum levels of soluble Klotho protein. Transplant. Proc. 45, 134-136 (2013).
44. M. H. de Borst, M. G. Vervloet, P. M. ter Wee, G. Navis, Cross talk between the renin-angiotensin-aldosterone system and vitamin D-FGF-23-klotho in chronic kidney disease. J. Am. Soc. Nephrol. 22, 1603-1609 (2011). 45. C. Faul, A. P. Amaral, B. Oskouei, M. C. Hu, A. Sloan, T. Isakova, O. M. Gutierrez, R. Aguillon-Prada, J. Lincoln, J. M. Hare, P. Mundel, A. Morales, J. Scialla, M. Fischer, E. Z. Soliman, J. Chen, A. S. Go, S. E. Rosas, L. Nessel, R. R. Townsend, H. I. Feldman, M. St John Sutton, A. Ojo, C. Gadegbeku, G. S. Di Marco, S. Reuter, D. Kentrup, K. Tiemann, M. Brand, J. A. Hill, O. W. Moe, M. Kuro-O, J. W. Kusek, M. G. Keane, M. Wolf, FGF23 induces left ventricular hypertrophy. J. Clin. Invest. 121, 4393-4408 (2011).
46. H. Olauson, K. Lindberg, R. Amin, T. Sato, T. Jia, R. Goetz, M. Mohammadi, G. Andersson, B. Lanske, T. E. Larsson, Parathyroid-specific deletion of Klotho unravels a novel calcineurin-dependent FGF23 signaling pathway that regulates PTH secretion. PLoS Genet. 9, e1003975 (2013).
47. T. Shiraki-Iida, H. Aizawa, Y. Matsumura, S. Sekine, A. Iida, H. Anazawa, R. Nagai, M. Kuro-o, Y. Nabeshima, Structure of the mouse klotho gene and its two transcripts encoding membrane and secreted protein. FEBS Lett. 424, 6-10 (1998).
48. H. Mitani, N. Ishizaka, T. Aizawa, M. Ohno, S. Usui, T. Suzuki, T. Amaki, I. Mori, Y. Nakamura, M. Sato, M. Nangaku, Y. Hirata, R. Nagai, In vivo klotho gene transfer ameliorates angiotensin II-induced renal damage.
Hypertension. 39, 838-843 (2002).
49. Y. Wang, Z. Sun, Klotho gene delivery prevents the progression of spontaneous hypertension and renal damage. Hypertension. 54, 810-817 (2009).
50. Y. Wang, M. Kuro-o, Z. Sun, Klotho gene delivery suppresses Nox2 expression and attenuates oxidative stress in rat aortic smooth muscle cells via the cAMP-PKA pathway. Aging Cell. 11, 410-417 (2012).
157
51. J. Voelkl, I. Alesutan, C. B. Leibrock, L. Quintanilla-Martinez, V. Kuhn, M. Feger, S. Mia, M. S. Ahmed, K. P. Rosenblatt, M. Kuro-O, F. Lang, Spironolactone ameliorates PIT1-dependent vascular osteoinduction in klotho-hypomorphic mice. J. Clin. Invest. 123, 812-822 (2013).
52. R. Nakano-Kurimoto, K. Ikeda, M. Uraoka, Y. Nakagawa, K. Yutaka, M. Koide, T. Takahashi, S. Matoba, H. Yamada, M. Okigaki, H. Matsubara, Replicative senescence of vascular smooth muscle cells enhances the calcification through initiating the osteoblastic transition. Am. J. Physiol. Heart Circ. Physiol. 297, H1673-84 (2009).
53. Y. Fang, C. Ginsberg, M. Seifert, O. Agapova, T. Sugatani, T. C. Register, B. I. Freedman, M. C. Monier-Faugere, H. Malluche, K. A. Hruska, CKD-induced wingless/integration1 inhibitors and phosphorus cause the CKD-mineral and bone disorder. J. Am. Soc. Nephrol. 25, 1760-1773 (2014).
54. N. A. van Venrooij, R. C. Pereira, Y. Tintut, M. C. Fishbein, N. Tumber, L. L. Demer, I. B. Salusky, K. Wesseling-Perry, FGF23 protein expression in coronary arteries is associated with impaired kidney function. Nephrol. Dial.
Transplant. 29, 1525-1532 (2014).
55. J. Zhang, J. R. Chang, X. H. Duan, Y. R. Yu, B. H. Zhang, Thyroid hormone attenuates vascular calcification induced by vitamin D3 plus nicotine in rats. Calcif. Tissue Int. 96, 80-87 (2015).
56. C. S. Ritter, S. Zhang, J. Delmez, J. L. Finch, E. Slatopolsky, Differential expression and regulation of Klotho by paricalcitol in the kidney, parathyroid, and aorta of uremic rats. Kidney Int. 87, 1141-1152 (2015).
158
Table 1. Patient characteristics
*CEA: carotid endarterectomy; † eGFR: estimated glomerular filtration rate. eGFR is expressed
as 4 points MDRD for all patients, except for kidney donors. For these patients, 24 hour urine collection measurement was used, because of high values obtained.
Kidney donors (N=9) Kidney recipients (N=10) CEA* patients (N=8) Age (years) 61 [32-68] 44 [18-63] 70 [57-80] Sex (M/F) 6 / 3 5 / 5 5 / 3 Diabetes 0 0 1
eGFR† before surgery
159
Supplementary methods
Staining on serial sections for Klotho
Serially cut cryosections were stained as described in the Methods section with the following modifications. The tertiary antibody used was an alkaline-phosphatase-conjugated goat anti-rabbit polyclonal (Dako, D0487), followed by a quaternary antibody (alkaline phosphatase-conjugated anti-alkaline phosphatase, Dako, D0651) 1:100 in 1% AB serum/1% BSA/PBS for 30 minutes, followed by the chromogen reaction for 30 minutes in a solution of 20 mg Naphtol AS-Mx (Sigma), 400 µl 10% MgSO4, 100 mg Fast Blue BB (Sigma), and 24 mg levamisole (Sigma)
in 100 ml 0.1 M Tris/HCl (pH 8.2).
RNA interference for Klotho
HK-2 cells were transfected in a 12-wells plate using 2 µl lipofectamine RNAiMAX (Life Technologies) per well and a final concentration of 10 nM siRNA (against Klotho: s17914, s225119, and 15204; scrambled, or GAPDH siRNAs, Life Technologies) in 200 µl Opti-MEM reduced serum medium supplied with 1 ml of 10% FCS/DMEM, for 48 h.
160
Supplementary Figure 1. Comparison of staining patterns of anti-Klotho antibodies on serially cut human renal cryosections. KM2076 and SC-22220 were used for immunohistochemistry, displaying positive staining in the same cells in the same tubules. Note the negative central arteriole. Magnification: 200x.
161
Supplementary Figure 2. Klotho expression in HK-2 cells. A PCR was performed for Klotho in HK-2 cells. Its expression can be reduced after RNA interference with siRNAs specific for the Klotho gene but not with siRNAs specific for GAPDH, validating the PCR specificity.
162
Supplementary Figure 3. Western blot for renal and recombinant Klotho using different anti-Klotho antibodies. KM2076 and SC-22220 were used to detect renal and recombinant Klotho, which was detecting by both antibodies at around 130 kDa. SC-22220 was less sensitive and deemed unsuitable for work on tissues.
163
