Calcium-sensing receptor- and ADAM10-mediated klotho shedding is regulated by tetraspanin 5

Abstract

Soluble, circulating Klotho (sKlotho) is essential for normal health and renal function. sKlotho is shed from the renal distal convoluted tubule (DCT), its primary source, via enzymatic cleavage. However, the physiologic mechanisms that control sKlotho production, trafficking, and shedding are not fully defined. We previously found that the G protein-coupled calcium-sensing receptor (CaSR) co-localizes with membrane-bound αKlotho and the disintegrin/metalloprotease ADAM10 in the DCT and controls sKlotho in response to CaSR ligands and pHo by activating ADAM10. Here, we advance understanding of this process by showing that tetraspanin 5 (Tspan5), a scaffolding and chaperone protein, contributes to the cell surface expression and specificity of a protein complex that includes Tspan5, ADAM10, Klotho, and CaSR. These results support a model of multiprotein complexes that confer signaling specificity beyond CaSR on G protein-coupled processes. Impact statement Systemic circulating sKlotho is a determinant for normal physiology. Studies of knockout animals established its role as an anti-aging protein. The regulatory mechanisms for Klotho production and secretion are largely unknown. We report that Tspan 5 contributes to CaSR- and ADAM10-dependent Klotho shedding from the kidney, its primary source.

Keywords: ADAM10; Klotho; Tspan; calcim‐sensing receptor; distal convoluted tubule.

Fig. 1

Comparison of mouse renal cortical protein expression in distal convoluted tubules. Frozen sections of mouse kidney cortex were incubated with antibodies recognizing Klotho (red) and (A) Tspan5 (green, upper) or (B) Tspan15 (green, lower) and imaged with confocal microscopy. Merged‐images (third column) show overlap of the two colors, with selected boxed areas that were magnified and shown with rendered 3‐D stack images (last column). Rendered 3‐D Z‐stacks are shown to highlight the overlap of Tspan5 and Klotho staining (yellow) and the significantly reduced overlap of Tspan15 and Klotho. Scale bar = 5 μm. Representative images were selected from at least 3 technical replicates.

Fig. 2

Tspan5 depletion decreases cell surface expression of Klotho and the CaSR. HEK‐293 cells that stably express the CaSR were transiently transfected with Klotho in the presence of endogenous Tspan5, 15, and ADAM10 proteins. Cells were transfected with non‐specific control siRNA (si‐NC), siRNA directed against Tspan5 (si‐T5), or Tspan15 (si‐T15), grown on coverslips, and studied after 2 days. Cells were fixed under non‐permeabilizing conditions and stained with antibodies to Tspan5 or Tspan15 (green), Klotho (red) and the CaSR (magenta). (A) Enlarged merged‐images of cells transfected with si‐NC (left set of panels) or si‐T5 (right set of panels), show marked reduction in Tspan5, Klotho, and CaSR immunofluorescence. (B) Enlarged merged‐images of cells transfected with si‐NC (left set of panels) or si‐T15 (right set of panels), show reduced green Tspan15 immunofluorescence, without significant reduction in Klotho or CaSR immunofluorescence. Scale bar is 5 μm. Split images are shown below respective enlarged merged‐images. Images are representative of at least 3 independent experiments.

Fig. 3

Cell surface depletion of Tspan5 but not Tspan 15 reduces CaSR‐stimulated Klotho shedding. HEK‐293 cells that stably express the CaSR, endogenous Tspan5, 15, and ADAM10 and transiently express Klotho were transfected with non‐specific control siRNA (si‐NC), siRNA directed against Tspan5 (si‐T5), or Tspan 15 (si‐T15), and studied after 2 days. (A) Representative immunoblots of cell extracts showing expression levels of Klotho, Tspan5, or Tspan15 compared to Tubulin. Klotho expression after transfection were comparable (upper panel), and total Tspan5 and Tspan 15 proteins were partly reduced by respective siRNA. (B) Representative immunoblot of Klotho shed in 3 h of incubation in R‐568 (CaSR‐activator) supplemented tissue culture medium. Quantification of shed Klotho from at least 3 independent experiments (open circles) performed in different weeks, is shown below; **P < 0.001 by ANOVA followed by Sidak's multiple comparisons post‐test using graphpad prism. (C) Quantification of mRNA for Tspan5, Tspan15, the CaSR, and ADAM10 by qRT‐PCR after transfection of siRNA reagents. The GAPDH‐normalized levels of indicated mRNA are shown, relative to si‐NC (dotted line), after transfection with si‐T5 (light gray bars) or si‐T15 (dark gray bars); N = 3, *P < 0.05, **P < 0.001, by 2‐way ANOVA followed by Dunnett's multiple comparisons test against si‐NC control using graphpad prism.

Fig. 4

Co‐immunoprecipitation and co‐fractionation studies support transient interactions among Klotho, the CaSR, ADAM10, and Tspan5. (A) Glycerol density gradient fractionation of mouse kidney extracts. Fraction numbers are shown at the top of the panels. Representative immunoblot images for the CaSR, Klotho, ADAM10, and Tspn5 are shown for each protein and each fraction. The CaSR co‐fractionates with Klotho, ADAM10, and Tspan5. (B) Co‐immunoprecipitation of Klotho, ADAM10, and the CaSR from kidney cortex extracts. The lanes from left to right are: total input used, IP with control IgG and Protein G beads (IP‐IgG(G)), IP with Klotho antibody and Protein G beads (IP‐KL(G)), unbound fraction from IP‐IgG (UnB‐IgG) and unbound fraction from IP‐KL (UnB‐KL). Representative results from N = 4 experiments are shown.