Klotho

Klotho is a protein that has received considerable attention, not only from the biomedical research community but also from the general population. While klotho plays an important role in the regulation of phosphate metabolism, its potential role as an antiaging hormone has captured the interest of a wider audience. As CKD resembles accelerated aging, patients with CKD may benefit from klotho-focused aging research. However, based on the lack of a mechanistic understanding, it is currently challenging to develop klotho-based drugs, either for CKD or antiaging therapies.

Klotho is a single-pass transmembrane protein on the surface of proximal tubular epithelial cells where, combined with fibroblasts growth factor receptor (FGFR) isoform 1c, it serves as a receptor for the hormone fibroblast growth factor 23 (FGF23) and promotes phosphate excretion.¹ In CKD and aging, klotho expression progressively declines, resulting in the renal resistance to FGF23 and the elevation in serum phosphate levels. At high concentrations, phosphate can contribute to various pathologies, such as vascular calcification. Therefore, nephrologists aim to lower serum phosphate levels in their patients with CKD using various approaches, such as a reduction of dietary phosphate intake or the pharmacological blockade of phosphate uptake in the gut. However, to date, these interventions have shown only modest effects on reducing cardiovascular damage and mortality in patients with CKD. Because it seems that klotho expression is reduced before the loss of functional kidney mass and not accompanied by a reduction in FGFR1c expression, increasing klotho levels in the kidney might restore FGF23 sensitivity and renal phosphate excretion and thereby serve as a novel therapeutic approach to lower systemic phosphate levels.^2–4

Patients with CKD are also exposed to high serum concentrations of FGF23, which can directly target cells in a klotho-independent manner and cause tissue damage.⁵ These pathologic actions of FGF23 have been best described for the FGFR4-mediated induction of cardiac hypertrophy.⁶ Interestingly, the forced overexpression of klotho in the heart protects mice from the prohypertrophic actions of FGF23, suggesting cardioprotective effects of klotho.⁷ It seems that as expected from its physiologic role as a coreceptor for FGF23 and FGFRs in the kidney, klotho also increases the binding of FGF23 to FGFRs on heart muscle cells, thereby transforming a pathologic into a protective signal. It is possible that in the presence of klotho, FGF23 activates different FGFR isoforms and/or different downstream signaling pathways, which results in a switch in the cellular response.⁵

The entire ectodomain of klotho can be released as a soluble protein from the cell surface through proteolytic cleavage and enter the circulation.¹ Whether soluble klotho can be further processed into smaller fragments is currently unclear. The kidney is the main source of soluble klotho, and serum levels of soluble klotho seem to decline with reductions in renal klotho expression in CKD and aging. Like membrane-bound klotho, circulating soluble klotho acts as a coreceptor for FGFRs. Soluble klotho can take over the role of membrane-bound klotho and mediate FGF23/FGFR1c-induced renal phosphate excretion.⁴ Moreover, soluble klotho can protect cardiomyocytes from FGF23/FGFR4-induced hypertrophy.⁸ Overall, it seems that the actions of soluble klotho are FGF23/FGFR-dependent. However, other molecular targets and mechanisms of actions have also been proposed for soluble klotho, further contributing to its wide spectrum of beneficial effects and making it a promising drug target (Figure 1).⁹

Figure 1.

Molecular targets and therapeutic effects of soluble klotho administration in CKD. Soluble klotho has various modes of action. First, soluble klotho acts as an FGFR coreceptor. In the kidney, soluble klotho mediates the physiologic effects of FGF23 to increase phosphate (P) excretion, thereby lowering serum levels of phosphate and FGF23 and protecting tissues, such as the heart, from their pathologic actions. Soluble klotho also acts as an FGFR coreceptor on cells that do not per se express klotho to protect them from the pathologic actions of FGF23 and of paracrine FGFs. Second, soluble klotho might have more widespread effects that are independent of FGFRs. Soluble klotho can directly interfere with various ligand-receptor interactions, such as insulin/IGF1/IGF1R, TGFβ1/TGFβR, Wnt/Frizzled, and AngII/AT1R, and modify their downstream signaling pathways. Soluble klotho can also bind glycosylates lipids and proteins, like lectins, and modify their glycosylation pattern through an intrinsic enzymatic glycosidase activity. FGF23, fibroblast growth factor 23; FGFR, fibroblasts growth factor receptor; sKL, soluble klotho. The figure was generated with BioRender.com.

To date, several classic pharmacologic approaches have been used to test membranous and soluble klotho in preclinical models.¹ This includes the increase of endogenous klotho expression in the kidney by using pharmacological agents that mimic some of the known biological activators of klotho expression as well as small molecules that activate the klotho promoter or counterbalance the epigenetic silencing of the klotho gene. However, these approaches rely on the kidney as the major source of endogenous soluble klotho, which happens to be the tissue that is the origin of the clinical problem and that undergoes deterioration in CKD and to a certain extent also during the normal aging process. Therefore, systemic or local delivery of exogenous soluble klotho seems to be the more promising approach. Previous studies administered the cDNA encoding soluble klotho using plasmid-based or adenoviral-based delivery approaches. However, issues with the safety of the vehicle and with the precise dosing of the drug might limit the clinical utility of these approaches. Therefore, for klotho-based therapies, the direct delivery of the recombinant protein, similar to antibody-based therapies, might be the best path forward. However, because soluble klotho is a large protein of about 130 kDa with a short t_1/2 of <15 minutes, the protein in its native state is not suitable for pharmacological applications.⁸

About 5 years ago, we had the ambitious goal to produce the soluble klotho protein in large amounts and with high bioactivity and stability to test its therapeutic potential in preclinical models (Figure 2). To achieve this goal, three major steps were required: first, a procedure for the synthesis and purification of recombinant soluble klotho protein; second, an assay to detect and quantify its bioactivity; and third, the introduction of mutations and other modifications into the protein to increase its bioactivity and stability. Three published studies had a major impact on our design of the recombinant protein, its purification procedure, and activity assessment. First, the crystal structure analysis of the FGF23:FGFR1c:klotho protein complex revealed the presence of a Zn²⁺ ion within klotho, which is crucial for stabilizing its elongated structure and for forming the FGF23 binding pocket.⁴ Based on this finding, we have developed a procedure to synthesize and purify the recombinant mouse and human proteins from stably transfected cell lines in the absence of chelating agents, such as EDTA.⁸ We produce FLAG- and Strep-tagged soluble klotho in Expi-HEK293 liquid cultures, and the secreted soluble klotho protein is purified from the media using a three-step affinity procedure. Second, we previously wanted to determine whether soluble klotho can mediate FGF23 binding to specific FGFR isoforms. To do so, we developed a sandwich-like binding assay, similar to an ELISA, where 96-well plates are coated with FGF23 protein followed by the addition of soluble klotho and the ectodomain of different FGFR isoforms that are coupled to a Fc-tag.⁸ The formation of the protein complex is visualized by adding a horseradish peroxidase–coupled antibody that recognizes the Fc-tag of the FGFRs and by detecting enzymatic horseradish peroxidase activity. We found that in the presence of soluble klotho, FGF23 binds FGFR1c, FGFR3c, and FGFR4 with high affinity. Since then, we have used this assay to determine the bioactivity of our purified soluble klotho protein. Third, the mass spectrometry analysis of the purified soluble klotho protein identified several sites of post-translational modifications, including O- and N-glycosylations.¹⁰ Interestingly, eight asparagine (Asn) residues of soluble klotho can be covalently attached to N-glycan structures that contain rare disaccharides and have a low degree of sialylation, which is the terminal addition of sialic acid units to the oligosaccharides. Expression studies in different cell lines revealed that this unusual N-glycosylation reduces the t_1/2 of soluble klotho from about 24 hours to 30 minutes.¹⁰ Therefore, we aim to mutate and inactivate the acceptor sites for N-glycosylation within soluble klotho to increase its stability.

Figure 2.

Workflow for the production, purification, and testing of recombinant soluble klotho protein variants. The soluble klotho protein consists of two domains, KL1 and KL2, and is mutated (red flash) at specific sites to block N-glycosylation, which should increase its stability as well as binding affinity for FGF23. Mutations in the RBA should increase soluble klotho's affinity for FGFR1c. We also consider other approaches to improve protein t_1/2 that are used in other biologics, such as Fc-addition, lipidation, or PEG conjugation (red star). After expression and purification, the bioactivity of the soluble klotho variants is measured on the basis of their effects on increasing the binding affinity between FGF23 and FGFR1c. Afterward, soluble klotho variants are injected into rats, and serum soluble klotho levels over time are determined using the plate-based FGF23/FGFR1c binding assay. The bioactivity of the most stable soluble klotho variants is then further evaluated in vitro by studying their effects on signal transduction events and on biological readouts. Finally, soluble klotho variants with highest bioactivity and stability are injected into healthy mice and into mouse models of CKD to study their effects on phosphate metabolism and on CKD-associated tissue damage. RBA, receptor binding arm. The figure was generated with BioRender.com.

Our previous work indicates that soluble klotho can bind FGF23 and FGFR1c separately with high affinity.⁸ Soluble klotho increases FGFR binding affinity of FGF23 by both, first binding to FGFR and serving as a soluble FGFR coreceptor or first binding to FGF23 serving as a circulating FGF23 binding partner. Recently, we found that by mutating and thereby inactivating a specific Asn residue that serves as a N-glycosylation site in soluble klotho, its binding affinity for FGF23 is increased by about three-fold. Now, we aim to further increase FGF23 binding by mutating additional N-glycosylation sites, separately and in combination. Usually, protein glycosylation interferes with the protein's ability to interact with other molecules, while increasing the protein's t_1/2. Therefore, the finding that N-glycosylation of soluble klotho reduces its activity and its stability at the same time is somewhat unusual, and it provides us with the opportunity to improve both, protein activity and stability, by inactivating N-glycosylation sites through point mutations, which should allow for decreased dosing of soluble klotho in future animal studies. Native soluble klotho binds FGFR1c with relatively low affinity, which is not optimal for a pharmacological administration. Unlike for FGF23 binding, we have found that N-glycosylation of soluble klotho does not affect its binding affinity for FGFR1c. The crystal structure analysis revealed that soluble klotho contains a domain of 50 amino acids (also called receptor binding arm) that binds FGFR1c.⁴ Now, we will use phage display technology combined with our soluble klotho binding assay to generate soluble klotho variants carrying mutations in the receptor binding arm domain with increased binding affinity for FGFR1c.

To move from studying biology to developing drugs, we founded a startup company, called Alpha Young LLC, with the support of an Small Business Innovation Research Fast Track grant from the National Institute of Diabetes and Digestive and Kidney Diseases. Our early-stage mutant protein was the blueprint and proof of concept for the project that we now further develop into a variant with even higher bioactivity and stability. Because the “special sauce” used to produce the recombinant protein is key but not patentable, we started the company without having intellectual property. However, the specific soluble klotho protein variants generated from here on will be patented and will serve as the foundation for the company to move forward and to secure additional funding. The most promising soluble klotho mutant variants will be tested for their bioactivity in cell culture and animal models, which are relevant for CKD.

Supplementary Material

jasn-35-1270-s001.pdf ^{(1.5MB, pdf)}

Disclosures

Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/JSN/E672.

Funding

C. Yanucil: National Institute of Diabetes and Digestive and Kidney Diseases (R44DK132996 and F31DK115074). C. Faul: National Heart, Lung, and Blood Institute (R01HL145528).

Author Contributions

Conceptualization: Christian Faul, Christopher Yanucil.

Funding acquisition: Christian Faul.

Investigation: Christian Faul, Christopher Yanucil.

Methodology: Christian Faul, Christopher Yanucil.

Visualization: Christian Faul.

Writing – original draft: Christian Faul.

Writing – review & editing: Christian Faul, Christopher Yanucil.