Figure 5. Murine Piezo2, but not Piezo1, harbors a PIP₂ binding motif.

(a) Schematic view of the PI(3,5)P₂ binding region of TRPML1 identified elsewhere (Dong et al., 2010) and the region in murine Piezo2 that exhibits pronounced sequence similarity to the PI(3,5)P₂ binding region of TRPML1. The indicated sequences were used to generate peptides for Piezo2, the Piezo2 3Q mutant and Piezo1. All peptides were tagged with a FLAG-epitope to allow for detection with anti-Flag antibodies on immunoblots. (b–d) Representative peptide-lipid binding assays followed by immunoblotting after incubation with indicated peptides (b, c) and densitometric quantification (d). The arrangement of lipids on the lipid-strip is indicated; blank, no lipid was spotted. The Piezo2 peptide strongly binds to PI(3,5)P₂ and PI(4,5)P₂, and weakly to PI(3,4)P₂. Neither the Piezo2 3Q mutant peptide nor the Piezo1 peptide exhibited significant binding to any of the lipids tested. One-way ANOVA followed by Dunnett’s multiple comparisons test was used to compare spot signal densities for each peptide to the respective blank, p-values are indicated by # in the graph (d). In addition, one-way ANOVA followed by Holm-Sidak’s multiple comparisons test was used to compare spot signal densities across the three peptides, p-values are indicated by * in the graph (d). The graph only presents data for those lipids, to which the Piezo2 peptide exhibited significant binding. Please see Figure 5—figure supplement 1 for summarized data on lipids tested with the Piezo2 peptide. Experiments using the Piezo2 peptides were independently repeated 6 times, of which four times were conducted in parallel with experiments using the Piezo1 peptide. AU, arbitrary units. (e) Stimulus-current curves upon co-expression of the Piezo2 P1 mutant with Mtmr2 in HEK293 cells compared to mock controls. Mtmr2 only slightly attenuated MA currents (Piezo2 P1 mutant + mock: n = 13 cells; Piezo2 P1 mutant + Mtmr2: n = 10 cells). 2-way ANOVA reported a significant (P<0.0119) overall effect of Mtmr2 overexpression on RA-MA currents of the Piezo2 P1 mutant, however a Holm-Sidak’s multiple comparisons test showed no significant difference between currents at individual stimulus magnitudes. Of note, the displacement threshold and inactivation time constant were also unaffected upon overexpression of Mtmr2 compared to mock (Supplementary file 1).

Figure 5—figure supplement 1. Quantification of peptide-lipid binding assays using the Piezo2 peptide.

Piezo2 peptide densitometry data were analyzed using one-way ANOVA followed by Dunnett’s multiple comparisons test to compare spot signal densities for different lipids to the blank, p-values are indicated by # in the graph. AU, arbitrary units.