Figure 1. Mtmr2 suppresses Piezo2-mediated RA-MA currents in HEK293 cells and DRG neurons.

(a–d) Representative images (a,c) and quantification (b,d) of a proximity ligation assay (PLA) in cultured DRG neurons (a,b) of Piezo2^GFP mice (Woo et al., 2014) and HEK293 cells (c,d). As anti-Mtmr2 antibodies failed to work in neuronal cultures, DRG were transfected with Mtmr2-myc or mock-myc and PLA was performed with antibodies against Piezo2 and myc. Please note the distribution of the PLA signal in soma and neurites of DRG. HEK293 cells were co-transfected with Piezo2-GST-IRES-GFP and Mtmr2-myc or Piezo2-GST-IRES-GFP and mock-myc and PLA was performed with antibodies against GST and myc. Only cells with pronounced GFP signal (due to expression of pmaxGFPVector in DRG and Piezo2-GST-IRES-GFP in HEK293 cells) were considered for the analysis. Cell boundaries are demarcated in yellow. In both cell types, DRG and HEK293 cells, transfection of Mtmr2-myc exhibited significantly stronger PLA signal compared to controls (b,d). Scale bar: 10 µm. Quantification of the total area of PLA signal/total soma area (fraction of PLA-positive area) in DRG cultures (p<0.0001; Mann-Whitney test; + mock: n = 53 neurons; + Mtmr2-myc: n = 53 neurons) (b). The quantification of the intensity of PLA signal in neurites of cultured DRG neurons can be found in Figure 1—figure supplement 1c. Quantification of the total area of PLA signal/total cell area in HEK293 cells (fraction of PLA-positive area) (p<0.0001; Mann-Whitney test; Piezo2-GST + mock: n = 60 cells; Piezo2-GST + Mtmr2-myc: n = 54 cells) (d). Additional controls for PLA in HEK293 cells can be found in Figure 1—figure supplement 1d. (e) Representative traces of RA-MA currents in HEK293 cells upon co-expression of Piezo2 with mock or Mtmr2 and (f) stimulus-current curves. Overexpression of Mtmr2 suppressed Piezo2 current magnitudes compared to mock overexpression (Piezo2 + mock: n = 17 cells; Piezo2 + Mtmr2: n = 12 cells; 2-way ANOVA suggested a significant effect (P<0.0001) of Mtmr2 overexpression on Piezo2 currents; Holm-Sidak’s multiple comparisons test was used to compare both conditions at individual stimulus magnitudes, p-values are indicated by * in the graph). The displacement threshold was increased upon co-expression of Mtmr2 (p=0.0098; Mann-Whitney test; Supplementary file 1). The inactivation time constant of RA-MA currents remained unchanged (Supplementary file 1). (g) Representative traces of RA-MA currents in primary cultures of DRG neurons and (h) stimulus-current curves showed a significant decrease in RA-MA current magnitude upon overexpression of Mtmr2 compared to mock ( + mock: n = 28 neurons; + Mtmr2: n = 30 neurons; 2-way ANOVA suggested a significant effect (P<0.0022) of Mtmr2 overexpression on RA-MA currents; Holm-Sidak’s multiple comparison test was performed to compare both conditions at individual stimulus magnitudes, p-values are indicated by * in the graph). The displacement threshold and inactivation time constant of RA-MA currents were not changed upon overexpression of Mtmr2 in DRG neurons (Supplementary file 1).

Figure 1—figure supplement 1. Mtmr2 is expressed in mouse DRG and also in close vicinity to Piezo2.

(a) Representative immunohistochemistry and (b) quantification of Mtmr2-positive neurons in cryosections of DRGs obtained from Piezo2^GFP mice (Woo et al., 2014). 20.37 ± 2.01% of DRG neurons exhibit positive Mtmr2 immunolabel and 24.53 ± 1.21% of DRG neurons were positive for GFP indicating the presence of Piezo2. Among Piezo2-positive cells, 51.61 ± 3.71% cells were found to be Mtmr2-positive. n > 2000 neurons, N = 3 independent animals. (c) Quantification of the intensity of PLA signal in neurites of cultured DRG neurons (p<0.0001 compared to mock, Mann Whitney test; + mock: n = 94 regions from six coverslips; + Mtmr2-myc: n = 115 regions from six coverslips. (d) Quantification of PLA signal upon co-expression of Piezo2 with mock, Mtmr2, Vti1b (Vesicle transport through interaction with t-SNAREs homolog 1B) or TRPA1 in HEK293 cells. PLA was performed using antibodies against GST and myc to detect Piezo2-GST-IRES-GFP and myc-tagged proteins, respectively. PLA signal (fraction of PLA-positive area) upon co-transfection of mock, Vti1b and TRPA1 served as negative controls and was indeed much smaller than the PLA signal upon co-transfection of Piezo2 and Mtmr2 (p<0.01, in case of TRPA1 p=0.052 compared to Piezo2 + Mtmr2, Kruskal-Wallis test followed by Dunn´s multiple comparison test; Piezo2-GST + mock: n = 60 cells; Piezo2-GST + Mtmr2-myc: n = 35 cells; Piezo2-GST + Vti1b-myc: n = 54 cells; Piezo2-GST + TRPA1-myc: n = 83 cells). Only cells with pronounced GFP signal (due to expression of Piezo2-GST-IRES-GFP) were considered for the analysis. All experiments were performed in parallel.

Figure 1—figure supplement 2. Mtmr2 overexpression does not influence Kv1.1- or Piezo1-mediated currents.

(a) Voltage-current curves upon overexpression of Kv1.1 with Mtmr2 in HEK293 cells showed no significant difference compared to overexpression with mock (Kv1.1 + mock: n = 24 cells; Kv1.1+ Mtmr2: n = 25 cells; ns; 2-way ANOVA). (b) Stimulus-current curves of Piezo1-mediated MA currents upon co-overexpression of Piezo1 (Piezo1-753-myc-IRES-GFP [Coste et al., 2015]) with Mtmr2 compared to mock. No significant difference was observed among conditions (Piezo1 + mock: n = 25 cells; Piezo1 + Mtmr2: n = 20 cells; ns; 2-way ANOVA). The displacement threshold and inactivation time constant of Piezo1-MA currents remained unchanged upon overexpression with Mtmr2 (Supplementary file 1).