Figure 3. IRSp53-dependent phase... | Rockefeller University Press

$IRSp53-dependent phase separation of the PSD complex. (A) Schematic diagram showing the protein interaction network of the reconstituted 5× PSD. GKAP and IRSp53 act in parallel to connect proteins in the PSD core (PSD-95) and the pallium (Shank3 and Homer3). (B) Sedimentation experiments showing the protein distribution of different PSD components in aqueous (S) or condensed phase (P). Each component was at 5 μM in 3× PSD (PSD-95, Shank3, and Homer3) or 4× PSD (PSD-95 and GKAP or IRSp53, Shank3, and Homer3). Quantifications of the PSD-95, Homer3, and Shank3 distribution are shown underneath the SDS-PAGE. The results are from three independent batches of sedimentation experiments and are presented as mean ± SD. ****, P < 0.0001 using one-way ANOVA with Dunnett’s multiple comparison test. (C) Fluorescence images showing that the mixture of 5× PSD components at 5 μM undergoes LLPS at room temperature. PSD-95, IRSp53, Shank3, and Homer3 were labeled with different fluorophores as indicated and were highly enriched into the condensate droplets. GKAP was not labeled and is thus invisible. (D) Sedimentation experiments showing the phase separation capability of PSD components when 4× PSD (PSD-95, GKAP, Shank3, and Homer3) was mixed with IRSp53 (WT and mutant) at 5 μM. To help compare the LLPS level in different assay conditions, only the protein distribution in the pellet fraction is shown, with total protein input shown in the second lane. A full SDS-PAGE, including both the pellet and supernatant fractions, is included in Fig. S3 A. Quantifications of the PSD-95, Shank3, and IRSp53 distributions are shown on the right. Quantifications of all PSD components are shown in Fig. S3 A. *, P < 0.05; **, P < 0.01; ***, P < 0.001 using one-way ANOVA with Dunnett’s multiple comparison test. (E) Sedimentation experiments showing the phase separation capability of PSD components when 4× PSD* (PSD-95, GKAP, Shank3_ΔPRM, and Homer3) was mixed with IRSp53 (WT and mutant) at 5 μM. To help compare the LLPS levels in different assay conditions, only protein distribution in the pellet fraction is shown, with total protein input shown in the first lane. A full SDS-PAGE, including both the pellet and supernatant fractions, is included in Fig. S3 B. Quantifications of the PSD-95, Shank3, and IRSp53 distributions are shown on the right. Quantifications of all PSD components are shown in Fig. S3 B. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 using one-way ANOVA with Dunnett’s multiple comparison test. (F) Fluorescence images showing the LLPS level of PSD components when mixed with IRSp53 WT or mutant as illustrated in D. PSD-95 and Shank3 were fluorescently labeled, and their enrichment into the condensates was monitored. Identical imaging settings were used for all groups. (G) Fluorescence images showing the LLPS level of PSD components when mixed with IRSp53 WT or mutant as illustrated in E. PSD-95 and Shank3 WT or ΔPRM were fluorescently labeled, and their enrichment into the condensates was monitored. Identical imaging settings were used for all groups. (H) Representative fluorescence images of cultured IRSp53 knockout cortical neurons with the re-expression of mEGFP-tagged IRSp53 WT or ΔPBM at DIV7 (i). mCherry was cotransfected as the cell fill. Endogenous IRSp53 was deleted upon expression of Cre recombinase, and mEGFP-tagged IRSp53 WT or ΔPBM cDNA was transfected for 10 d. At DIV17, live imaging was performed without fixation. For quantification of the imaging data, spine enrichment (ii) was measured using ImageJ, and spine head width (iii) was measured using IMARIS. 22–42 neurons from three independent batches of cultures were imaged for each group in double-blinded mode (n = 42 for Cre only, n = 36 for Cre co-expression with IRSp53 WT, and n = 22 for Cre coexpression with IRSp53_ΔPBM). Note that data in Fig. 6 were collected from the same set of experiments with the same control groups (i.e., Cre only and Cre co-expression with IRSp53 WT). Error bars indicate mean ± SEM. ****, P < 0.0001 using two-tailed t test (ii) or one-way ANOVA with Dunnett’s multiple comparison test (iii). Source data are available for this figure: SourceData F3.$

Figure 3.

IRSp53-dependent phase separation of the PSD complex. (A) Schematic diagram showing the protein interaction network of the reconstituted 5× PSD. GKAP and IRSp53 act in parallel to connect proteins in the PSD core (PSD-95) and the pallium (Shank3 and Homer3). (B) Sedimentation experiments showing the protein distribution of different PSD components in aqueous (S) or condensed phase (P). Each component was at 5 μM in 3× PSD (PSD-95, Shank3, and Homer3) or 4× PSD (PSD-95 and GKAP or IRSp53, Shank3, and Homer3). Quantifications of the PSD-95, Homer3, and Shank3 distribution are shown underneath the SDS-PAGE. The results are from three independent batches of sedimentation experiments and are presented as mean ± SD. ****, P < 0.0001 using one-way ANOVA with Dunnett’s multiple comparison test. (C) Fluorescence images showing that the mixture of 5× PSD components at 5 μM undergoes LLPS at room temperature. PSD-95, IRSp53, Shank3, and Homer3 were labeled with different fluorophores as indicated and were highly enriched into the condensate droplets. GKAP was not labeled and is thus invisible. (D) Sedimentation experiments showing the phase separation capability of PSD components when 4× PSD (PSD-95, GKAP, Shank3, and Homer3) was mixed with IRSp53 (WT and mutant) at 5 μM. To help compare the LLPS level in different assay conditions, only the protein distribution in the pellet fraction is shown, with total protein input shown in the second lane. A full SDS-PAGE, including both the pellet and supernatant fractions, is included in Fig. S3 A. Quantifications of the PSD-95, Shank3, and IRSp53 distributions are shown on the right. Quantifications of all PSD components are shown in Fig. S3 A. *, P < 0.05; **, P < 0.01; ***, P < 0.001 using one-way ANOVA with Dunnett’s multiple comparison test. (E) Sedimentation experiments showing the phase separation capability of PSD components when 4× PSD* (PSD-95, GKAP, Shank3_ΔPRM, and Homer3) was mixed with IRSp53 (WT and mutant) at 5 μM. To help compare the LLPS levels in different assay conditions, only protein distribution in the pellet fraction is shown, with total protein input shown in the first lane. A full SDS-PAGE, including both the pellet and supernatant fractions, is included in Fig. S3 B. Quantifications of the PSD-95, Shank3, and IRSp53 distributions are shown on the right. Quantifications of all PSD components are shown in Fig. S3 B. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 using one-way ANOVA with Dunnett’s multiple comparison test. (F) Fluorescence images showing the LLPS level of PSD components when mixed with IRSp53 WT or mutant as illustrated in D. PSD-95 and Shank3 were fluorescently labeled, and their enrichment into the condensates was monitored. Identical imaging settings were used for all groups. (G) Fluorescence images showing the LLPS level of PSD components when mixed with IRSp53 WT or mutant as illustrated in E. PSD-95 and Shank3 WT or ΔPRM were fluorescently labeled, and their enrichment into the condensates was monitored. Identical imaging settings were used for all groups. (H) Representative fluorescence images of cultured IRSp53 knockout cortical neurons with the re-expression of mEGFP-tagged IRSp53 WT or ΔPBM at DIV7 (i). mCherry was cotransfected as the cell fill. Endogenous IRSp53 was deleted upon expression of Cre recombinase, and mEGFP-tagged IRSp53 WT or ΔPBM cDNA was transfected for 10 d. At DIV17, live imaging was performed without fixation. For quantification of the imaging data, spine enrichment (ii) was measured using ImageJ, and spine head width (iii) was measured using IMARIS. 22–42 neurons from three independent batches of cultures were imaged for each group in double-blinded mode (n = 42 for Cre only, n = 36 for Cre co-expression with IRSp53 WT, and n = 22 for Cre coexpression with IRSp53_ΔPBM). Note that data in Fig. 6 were collected from the same set of experiments with the same control groups (i.e., Cre only and Cre co-expression with IRSp53 WT). Error bars indicate mean ± SEM. ****, P < 0.0001 using two-tailed t test (ii) or one-way ANOVA with Dunnett’s multiple comparison test (iii). Source data are available for this figure: SourceData F3.

This Feature Is Available To Subscribers Only