α-Klotho expression determines nitric oxide synthesis in response to FGF-23 in human aortic endothelial cells

Abstract

Endothelial cells (ECs) express fibroblast growth factor (FGF) receptors and are metabolically active after treatment with FGF-23. It is not known if this effect is α-Klotho independent or mediated by humoral or endogenous endothelial α-Klotho. In the present study, we aimed to characterize EC α-Klotho expression within the human vascular tree and to investigate the potential role of α-Klotho in determining FGF-23 mediated EC regulation. Human tissue and ECs from various organs were used for immunohistochemistry and Western blot. Primary cultures of human aortic endothelial cells (HAECs) and human brain microvascular endothelial cells (HBMECs) were used to generate in vitro cell models. We found endogenous α-Klotho expression in ECs from various organs except in microvascular ECs from human brain. Furthermore, FGF-23 stimulated endothelial nitric oxide synthase (eNOS) expression, nitric oxide (NO) production, and cell proliferation in HAECs. Interestingly, these effects were not observed in our HBMEC model in vitro. High phosphate treatment and endothelial α-Klotho knockdown mitigated FGF-23 mediated eNOS induction, NO production, and cell proliferation in HAECs. Rescue treatment with soluble α-Klotho did not reverse endothelial FGF-23 resistance caused by reduced or absent α-Klotho expression in HAECs. These novel observations provide evidence for differential α-Klotho functional expression in the human endothelium and its presence may play a role in determining the response to FGF-23 in the vascular tree. α-Klotho was not detected in cerebral microvascular ECs and its absence may render these cells nonresponsive to FGF-23.

Fig 1. FGF-23 increases cell proliferation, NO production, and eNOS protein expression and activation in human aortic endothelial cells (HAECs) but not in human brain microvascular endothelial cells (HBMECs).

(A) FGF-23 treatment for 48 hours stimulates proliferation in HAECs at 50 and 100 ng/ml, but not in HBMECs; n = 6. (B) NO production of HAECs was increased following treatment with 10 ng/ml FGF-23 for 16, 24, and 48 hours compared with controls. There was no FGF-23-stimulated NO production in HBMECs; n = 6. (C) FGF-23 increased phospho-eNOS (Ser1177) and eNOS protein expression in a time- and dose-dependent manner in HAECs. The phospho-eNOS (Ser1177) and eNOS protein expression of HAECs following stimulation with 10 ng/ml of FGF-23 for indicated periods of time and after incubation with FGF-23 at indicated concentrations for 48 hours was detected by Western blot; n = 6. (D) Time- and dose-response of FGF-23 on HBMECs, displayed no effect on phospho-eNOS (Ser1177) and eNOS protein expression; n = 6. Quantitative analysis of Western blot by densitometry is normalized to actin. Data represent mean ± SD of at least 3 independent experiments. *Significantly different versus control by ANOVA test with Bonferroni post-hoc analysis.

Fig 2. Full-length α-Klotho expression profiles in human endothelial cells, in vivo and in vitro.

(A) With immunohistochemistry, kidney tissues were used with anti-Klotho antibody as a positive control and with isotype antibody as a negative control. (B-F) Using immunohistochemical analysis of endothelium in various human tissues showed that α-Klotho protein was expressed in renal artery endothelium (B), in lung microvessels (C), and in colon submucosa (D). α-Klotho expression was not shown in microvascular ECs from cerebellum (E) and hippocampus (F) regions. Endothelium specific marker CD31 was used to stain consecutive tissue sections. Yellow arrows point to microvascular ECs. n = 5 in tissues of each site; Scale bar = 200 μm. (G) In vitro studies showed that the α-Klotho gene expression (upper panel) was detected in human aortic endothelial cells (HAECs), but not in human brain microvascular endothelial cells (HBMECs) with kidney as a positive control; n = 6. Western blot (lower panel) confirmed full-length α-Klotho protein expression in human umbilical venous endothelial cells (HUVECs), HAECs, pulmonary microvascular endothelial cells (HPMECs), and cardiac microvascular endothelial cells (HCMECs). No α-Klotho protein expression was found in HBMECs; n = 3. (H) Fibroblast growth factor receptor 1 (FGFR1) protein was expressed widely in human endothelial cells originating from various human tissues in vitro; n = 3. (I) Fibroblast growth factor receptor 3 (FGFR3) and fibroblast growth factor receptor 4 (FGFR4) were expressed in both HAECs and HBMECs; n = 3.

Fig 3. FGF-23-induced NO production and eNOS up-regulation are mitigated under high phosphate treatment, a state of α-Klotho deficiency.

(A) Endogenous α-Klotho expression in HAECs was suppressed after high phosphate treatment (2.5 and 5 mM β-glycerophosphate) for 24 hours. Phosphate did not influence phospho-FGFR (Tyr653/654), FGFR1, FGFR3, and FGFR4 expression; n = 6. (B and C) High phosphate (5 mM β-glycerophosphate) treatment mitigated FGF-23’s effect (10 ng/ml) on increased protein expression of phospho-eNOS (Ser1177) and eNOS (B) and NO production (C) in HAECs; n = 6. Soluble α-Klotho (0.2 nM; 26 ng/ml) did not render high phosphate-treated HAECs responsive to FGF-23. Quantitative analysis of Western blot by densitometry is normalized to actin. Data represent mean ± SD of at least 3 independent experiments. *Significantly different versus control by ANOVA test with Bonferroni post-hoc analysis.

Fig 4. Endothelial FGF-23 response depends on full-length α-Klotho but not soluble α-Klotho.

(A) FGF-23 treatment mediated proliferative effect on HAECs was mitigated by α-Klotho siRNA treatment. Treating with soluble α-Klotho (0.2 nM, 26 ng/ml) did not render HBMECs responsive to FGF-23; n = 3. (B) FGF-23 mediated NO production in HAECs was mitigated by α-Klotho siRNA treatment. Treating with soluble α-Klotho (0.2 nM, 26 ng/ml) did not render HAECs with α-Klotho knockdown responsive to FGF-23; n = 3. (C) HAEC α-Klotho knockdown with α-Klotho siRNA abrogated FGF-23 mediated eNOS protein up-regulation. Scrambled siRNA served as control (Vector) and display maintained endothelial FGF-23 response. Addition of soluble α-Klotho could not rescue HAEC resistance to FGF-23; n = 3. (D) HBMECs remained resistant to FGF-23 treatment with no increased expression of eNOS protein, time- and dose-dependent studies; n = 3. Quantitative analysis of Western blot by densitometry is normalized to actin. Data represent mean ± SD of at least 3 independent experiments. *Significantly different versus control by ANOVA test with Bonferroni post-hoc analysis.