*4.5. FGF23*/*Klotho and Vascular Calcification*

Vascular calcifications are common in patients with advanced CKD, and at present, it is responsible for the high CVD-related mortality [68]. Vascular calcification is the final consequence of a process where VSMC are transdi fferentiated into osteoblast-like cells [69].

Patients with end-stages CKD have an important disbalance of mineral metabolism with high levels of serum phosphate, which have been associated in vivo, and in vitro, with the generation of vascular calcification [70]. Di fferent authors and our group have shown that high phosphate levels can activate Wnt signaling in VSMC [71,72]. Interestingly other studies have shown that Klotho supplementation may prevent VSMC calcification through inhibition of the Wnt/β-catenin pathway [73].

We have investigated the relationship between vascular calcification, inflammation, and Wnt signaling. The administration of lipopolysaccharide (LPS) to healthy rats produced inflammation and a parallel increase in serum FGF23 levels and a reduction in renal Klotho expression. Subsequently, ex vivo experiments using slices of kidney tissue showed that LPS and also high phosphate-induced nuclear translocation of β-catenin and p65-NF-kB, with a decrease in Klotho. Inhibition of both inflammation and Wnt signaling activation decreased FGF23 levels and increased renal Klotho [74] (Figure 2). These results support the close relationship between inflammation, impairment in phosphate regulation, calcification, Klotho, and Wnt signaling.

The potential direct e ffect of FGF23 on VSMC to promote or inhibit calcification remains controversial. Some authors have found that FGF23 directly increases VSMC calcifications, while other authors emphasized that FGF23 is not involved in this process [75–77]. Similarly, it is unknown if FGF23 might or might not promote changes on the VSMC phenotype, with loss of vascular function, atherosclerosis, or even arterial sti ffness [78].

## **5. FGF23-Klotho and Wnt in Bone**

Historically, the participation of the Wnt/β-catenin pathway in bone disorders has been widely documented. SOST gene produces sclerostin that modulates the Wnt activity. Without sclerostin, Wnt activity is unrestricted, producing increased bone mineral density with hyperostosis. Thus, the canonical Wnt pathway is critical in bone formation, and its modulation could be a target in the treatment of bone disorders.

In relation to CKD, two inhibitors of the canonical Wnt pathway have been investigated: Dkk1 and sclerostin [79]. Paradoxically, despite both molecules inhibiting the Wnt ligand–LRP5/6–Frizzled interaction, the downstream responses are di fferent, illustrating the complexity of this pathway. In CKD patients, the correlation of serum Dkk1 with mineral and bone parameters is nonexistent in most studies [80,81], suggesting that Dkk1 might have a weak relation with renal osteodystrophy. Nevertheless, the serum sclerostin levels increase early in CKD before renal osteodystrophy is established. Sclerostin is produced and secreted by osteocytes, suggesting an essential role in the relationship between bone, kidney, and Wnt in CKD patients [82]. Serum sclerostin levels are higher in males than females, and the levels do not correlate with age. In CKD patients, plasma sclerostin increases progressively as the glomerular filtration rate declines, and it correlates with serum phosphate [83]. The cause of increased plasma levels of sclerostin in CKD patients is unknown. Osteocytes produce sclerostin, and VSMC transdi fferentiated into osteoblast in calcified vessels. Likewise, there is limited information about the relationship between sclerostin levels and bone in CKD. Paradoxically, there

is a positive association between serum sclerostin levels and bone mineral density in hemodialysis patients [84]. Additionally, the administration of neutralizing antibodies against sclerostin in a murine model of CKD resulted in beneficial only in low PTH conditions [85].

There are many questions in relation to sclerostin and CKD that remain to be answered. It is unknown if high levels of sclerostin protect against vascular calcification where Wnt/β-catenin promotes osteogenic transdi fferentiation of VSMC; it is also unclear to what extent high sclerostin a ffects bone turnover and renal osteodystrophy. The relationship between sclerostin and other mineral metabolism parameters, such as PTH, FGF23, vitamin D, or Klotho, is also unclear. Perhaps more studies are necessary to characterize the e ffects of Wnt activity on bone metabolism in CKD.

With respect to FGF23, Carrillo et al. identified that FGF23 directly inhibits Wnt signaling through the increase of Dkk1 levels. This action occurs in bone with the participation of soluble Klotho (sKlotho) [86]. This work provides evidence of the autocrine e ffects of FGF23, which could contribute to the generation of renal osteodystrophy (Figure 2). These results would be aligned with those indicating that an increase of sclerostin would contribute to the inhibition of osteogenesis. A recent study has shown a positive correlation between FGF23 and sclerostin levels in patients with rheumatic arthritis, suggesting a link between FGF23, reduced Wnt activity, and bone demineralization in these patients [87]. Evenepoel et al. found that sclerostin but not Dkk1 participate in alterations of mineral metabolism related to CKD [79,81].

In vitro studies have shown an interaction between FGF23, Klotho, and Wnt signaling in bone cells. The presence of Klotho in osteocytes and osteoblasts [88], suggests that the bone is another target organ for FGF23. Several studies indicate that Klotho is a negative modulator of bone formation [3]. The mechanisms are not clear, but it is speculated that Klotho allows FGF23 to enhance Dkk1 expression resulting in inhibition of Wnt signaling and osteogenesis. This hypothesis is supported by previous observations showing that Wnt activity is increased in Klotho knockout mice [89]. Ma et al. observed that in UMR-106, a bone cell line, the addition of β-glycerophosphate increased the expression of Wnt target genes; the co-administration of β-glycerophosphate and sKlotho led to a decrease in FGF23 levels and a reduction in Wnt activation, suggesting that sKlotho could modulate osteogenesis and FGF23 production [90]. In this line, other authors have observed that secreted Klotho, through the inhibition of FGFR1 and ERK phosphorylation, can delay human mesenchymal stem cell di fferentiation into osteoblasts [91,92] (Figure 2).

### **6. Wnt and the Central Nervous System**

FGF23, FGF receptors (FGFR), and the co-receptor Klotho are also expressed in the central nervous system. The biological relevance of the FGF23/Klotho system in the brain is uncertain, but there is some evidence that FGF23 directly acts on hippocampal neurons reducing memory functions and learning capacity in CKD patients [93,94]. Low serum Klotho levels have been reported to be associated with cognitive impairment [95]; however, the mechanisms are unknown. Given that Klotho is an antagonist of endogenous Wnt/β-catenin activity [32], it is reasonable to speculate that if Klotho reduces Wnt activity, upregulation of Wnt could be associated with cognitive impairment. Klotho-deficient mouse models rapidly develop cognitive impairment and show some evidence of neurodegeneration. In humans, there are reports showing a correlation between Klotho deficiency with dementia and Alzheimer's [96]. Di fferent explanations may support this association. First, both FGF23 increase and Klotho deficiency are associated with a vascular disease, which may cause cognitive deterioration because based on vascular dysfunction. Second, Vitamin D deficiency is highly prevalent in CKD patients, and alterations mainly mediate in the FGF23/Klotho axis. Vitamin D deficiency has also been related to cognitive decline in older adults. Third, Klotho plays a critical role in life-extension by regulating telomere length and telomerase activity. Both Klotho and telomeres regulate the stem cell aging process through Wnt signaling [97] (Figure 2). Klotho deficiency results in continuous activation of Wnt signaling and senescence of stem cells [98]. Long-lasting activation of Wnt signaling may cause rapid exhaustion and depletion of neural stem cells. Since stem cell dysfunction limits tissue

regeneration and potentially a ffects aging processes, the ability of secreted Klotho protein to inhibit Wnt signaling may reduce aging-like phenotypes in Klotho-deficient mice. Li et al. have shown that Klotho improves memory performance but disturbs some aspects of social behavior [99]. This has been proven in in vivo experiments, in which the addition of only the secreted Klotho protein improves the learning and memory capabilities of old animals [100]. Klotho is also being considered a new therapeutic target of neurodegenerative diseases [101]. Since CKD patients have an increase of FGF23 and a reduction of vitamin D and Klotho levels, it could be hypothesized that CKD patients may also show a decrease in the production of cerebral Klotho, which would upregulate Wnt signaling and produce the cognitive dysfunction frequently observed in these patients.

### **7. FGF23**/**Klotho and Wnt in Other Organs**

In the lungs, low Klotho may contribute to the development of idiopathic pulmonary fibrosis [102]. The co-administration of Klotho with rFGF23 reduced fibrosis and inflammation through the inhibition of the TGF-β signaling and the decrease in SMAD3 phosphorylation. Klotho relevance on pulmonary disease is reinforced by recent evidence suggesting that less circulating Klotho correlates negatively with lung function parameters, such as the forced vital capacity (FVC), the forced expiratory volume in 1 s (FEV1), and the di ffusing capacity of the lung for carbon monoxide (DLCO) [102,103].

Chronic obstructive pulmonary disease (COPD) is associated with the downregulation of Klotho expression in the airways and an increase in circulating FGF23 levels [104]. Oxidative stress produced by cigarette smoking may be responsible for Klotho deficiency in such a population. Moreover, COPD patients also show elevated inflammatory parameters that may increase FGF23 production, which in turn induce the expression of locally secreted IL-1β in bronchial epithelial cells [105]. Interestingly, the instillation of sKlotho protects bronchial epithelial cells from the pro-inflammatory actions associated with cigarette smoke and FGF23 [105]. Nevertheless, the precise mechanisms whereby pulmonary Klotho expression is downregulated remain undefined. Wnt/β-catenin signaling has recently gained relevance after the demonstration of enhanced noncanonical Wnt-5a activation in human fibroblasts from COPD patients, causing enlargement and destruction of alveolar space and contributing to emphysema development [106]. The inhibition of the Wnt-5a pathways in the lung helps to recover alveolar cell functions, perhaps through the regulation of TGF-β activity by Wnt-5a [107]. Upregulation of Wnt signaling is also associated with an increment in pulmonary vascular resistance, leading to the development of pulmonary arterial hypertension [41]; upregulation of the Wnt/β-catenin pathway has been associated with the proliferation of pulmonary artery smooth muscle cells, pulmonary artery resistance, and heart failure.

Our opinion is that the evidence is limited, and further investigation to define whether deficiency of lung Klotho and Wnt/β-catenin signaling plays a role in pulmonary fibrosis and emphysema is required.

Concerning the liver, a specific e ffect of α-Klotho in the liver is only partially defined. To date, there is no evidence of a detrimental e ffect of the FGF23/Klotho complex in the liver. In fact, FGF23 promotes hepatocytes proliferation and cytokine production [94,108], despite the lack of expression of Klotho in hepatocytes [94]. Thus, FGF23 action on the liver is Klotho independent, and it is mediated by FGFR4 [46,108]. The Wnt/β-catenin, together with di fferent FGFs, are pivotal in hepatobiliary development; early in embryonal development, β-catenin warrants hepatoblast proliferation and conversion into hepatocytes [109].

The liver is tightly associated with multiple endocrine functions, such as energy homeostasis. In this line, it seems that sKlotho improves insulin sensitivity and insulin release, and it reduces lipid accumulation in the liver [110]. Reciprocally, β-Klotho likely preserves liver integrity by serving as co-receptor for endocrine FGF21, a liver-derived hormone and member of the FGF family that promotes thermogenesis and glucose uptake in adipose tissue [111]. Chronic liver injuries frequently evolve liver fibrosis with a loss of function. There is data suggesting that the Wnt/β-catenin pathway is the main regulator of liver fibrosis. Both Wnt-5a and TGF-β are related to myofibroblast proliferation, collagen deposition, and fibrosis of the liver [41,112]. Nonetheless, dysregulation of Klotho has not been identified as responsible for liver fibrosis.

In summary, a consequence of the deterioration of kidney function is the modification of regulatory systems in an attempt to restore homeostasis. However, the "price to pay" is that these adaptations may disrupt the physiology, and comorbidities may become apparent. The Wnt/β-catenin cell signaling pathway has gained attention, given the demonstrated role in the development of different CKD-associated comorbidities. Phosphate overload downregulates renal Klotho expression through the activation of the Wnt/β-catenin signaling pathway, thus contributing to the development of vascular calcification and the alteration of the regulation of mineral metabolism.

The activation of Wnt/β-catenin has other consequences; it promotes tissue fibrosis in both kidney and heart and, more importantly, the upregulation of Wnt/β-catenin may facilitate the crosstalk between the heart and kidney playing a critical role in the development of the cardiorenal syndrome. It is important to note that activation of FGF23/Klotho/Wnt signaling correlates with the severity of atherosclerotic plaques, carotid intimal media thickness, and VSMC calcification. As such, the current evidence suggests that Wnt/β-catenin activation plays an essential role in CKD progression and cardiovascular disease. Therefore, the Wnt/β-catenin pathway may deserve future evaluation as a potential therapeutic target aiming to reduce the prevalence of CKD-associated comorbidities.

**Author Contributions:** J.R.M.-C., C.R.-H., and M.V.P.-R.d.M. wrote the draft. R.S., A.M.-M., and M.R. provided expertise and feedback. All authors edited and confirmed the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by a Spanish governmen<sup>t</sup> gran<sup>t</sup> from the Programa Nacional I+D+I 2013–2016 and Instituto de Salud Carlos III (ISCIII) Grants PI18/0138, PI17/01010 cofinancing from European Funds (FEDER), Consejería de Salud (Grant PI-0136) from the Junta de Andalucía, Framework Programme 7 Syskid UE Grant FP7-241544, and EUTOX and REDinREN from the ISCIII. J.R.M.-C. is senior researcher supported by the Nicolás Monardes Programme, Consejería de Salud-Servicio Andaluz de Salud (Junta de Andalucía).

**Conflicts of Interest:** M.R. has received honorarium for lectures from Abbott, Amgen, Inc., Fresenius, and Shire. The remaining authors declare no conflicts of interest.
