Skip to main content
. 2020 Nov 13;9:e56058. doi: 10.7554/eLife.56058

Figure 1. ZDHHC14 interacts with and palmitoylates both palmitoylated isoforms of PSD93 but more robustly palmitoylates PSD93β.

(A) Schematic of ZDHHC14 showing predicted transmembrane domains (TM, gray boxes), DHHC cysteine rich catalytic domain (blue box) and the C-terminal region used for yeast 2-hybrid screening (Bait), including the LSSV motif. (B) HEK293T cells were transfected with the indicated constructs and lysates subjected to GST-pulldown. Eluates from pulldowns were immunoblotted to detect GST (bottom, left) and PSD93 (top, left). Total expression levels of GST-tagged proteins (bottom, right) and PSD93 (top, right) in parent lysates were also determined. Images are representative of three independent experiments. (C) HEK293T cells were transfected with the indicated constructs and palmitoyl-proteins (isolated by ABE; left panels) and total protein levels (in parent lysates; right panels) were assessed by western blotting with the indicated antibodies. Parallel samples processed in the absence of the key ABE reagent hydroxylamine (NH2OH) confirm the specificity of the ABE assay. (D) Quantified PSD93α palmitoyl:total levels from C, normalized to the empty vector condition (Welch’s 1-way ANOVA p=0.0008, W(2,6.98) = 23.80, N = 6; Dunnett’s T3 multiple comparison post hoc test **p<0.01, 95% CI vector versus wtZDHHC14 [−6.91,–1.77], vector versus ZDHHC14 LSSE [−3.10, 0.86], and wtZDHHC14 versus ZDHHC14 LSSE [0.60, 5.83]). (E) Quantified PSD93β palmitoyl:total levels from C, normalized to the empty vector condition (1-way ANOVA p<0.0001, F(2,9)=60.69, N = 4; Bonferroni post hoc test **p<0.01, ***p<0.001, ****p<0.0001, 95% CI vector versus wtZDHHC14 [−6.18,–3.56], vector versus ZDHHC14 LSSE [−3.44,–0.84], and wtZDHHC14 versus ZDHHC14 LSSE [1.44, 4.04]). Uncropped western blot images are in Figure 1—figure supplement 4.

Figure 1—source data 1. Source data for Figure 1D and Figure 1—figure supplement 3.

Figure 1.

Figure 1—figure supplement 1. The PDZ ligand of ZDHHC14 is highly conserved in vertebrates.

Figure 1—figure supplement 1.

(A) Schematics of domain organization of ZDHHC14 orthologs, including human ZDHHC14 (Homo sapiens; NP_078906.2), mouse ZDHHC14 (Mus musculus; NP_666185.3), rat ZDHHC14 (Rattus norvegicus; NP_001034432.1), western clawed frog ZDHHC14 (Xenopus tropicalis; XP_004914695.1), Zebrafish ZDHHC14 (Danio rerio; XP_005160409.1), sea squirt ZDHHC9 (Ciona intestinalis; XP_002127630.1), fruit fly Approximated (App; Drosophila melanogaster; NP_001137937.1) and roundworm DHHC-2 (Caenorhabditis elegans; NP_0493007.2). Predicted transmembrane domains are shown in gray boxes (TMD) and the DHHC cysteine-rich catalytic domain is shown in a blue box (DHHC-CRD). (B) Sequence alignment of the 10 C-terminal amino acids of ZDHHC14 orthologs from A reveals a PDZ ligand only in vertebrates and simple chordate lineages. PDZ-ligand consensus amino acids are in yellow, completely conserved amino acids are highlighted in dark gray, and functionally conserved amino acids in light gray. C. intestinalis and C. elegans orthologs were identified based on homology of their respective DHHC-CRDs with mammalian ZDHHC14. A more limited version of this alignment was shown in Thomas and Hayashi, 2013a.
Figure 1—figure supplement 2. Further Yeast Two-Hybrid analysis suggests that the ZDHHC14 C-terminus binds the third PDZ domain of PSD93.

Figure 1—figure supplement 2.

(A) Schematic of the domain organization of PSD93 showing PDZ domains (yellow boxes), SH3 domain (orange box), and guanylate kinase (GK) domain (light brown). Solid lines: regions of the three unique ‘hit’ PSD93 cDNA clones confirmed by DNA sequencing. Restriction digest analysis suggested that each ‘hit’ insert was approx.1.5kb in length and thus likely extended into the GK region of PSD93. (B) Growth of four re-spotted individual yeast colonies on media selecting for presence of ‘bait’ (pPC97) and ‘prey’ (pPC86) plasmids (Leu- Trp-, upper panel) or presence of ‘bait’ and ‘prey’ plasmids and ‘bait’-‘prey’ protein–protein interaction (Leu- Trp- His-, lower panel), after HF7C yeast were back-transformed with the indicated ‘bait’ vectors and a ‘prey’ vector containing the third PDZ domain (PDZ3) of PSD93. Wild type ZDHHC14 directly interacts with PSD93-PDZ3 but a ZDHHC14 with a mutated PDZ ligand (LSSE) does not.
Figure 1—figure supplement 3. Further analysis of palmitoylation:total levels of PSD93α and PSD93β with or without wt or LSSE ZDHHC14 from Figure 1C.

Figure 1—figure supplement 3.

Quantified PSD93 palmitoyl:total levels from Figure 1C, normalized to the PSD93α empty vector condition (2-way ANOVA: ZDHHC14 p<0.0001 [F(2)=111], isoform p<0.0001 [F(1)=281.7], interaction p<0.0001 [F(2)=48.88]; N = 4–6; Bonferroni post hoc test *p<0.05, **p<0.01, ****p<0.0001; 95% confidence intervals vector versus ZDHHC14 PSD93α [1.14, 7.54], vector versus ZDHHC14 PSD93β [17.81, 25.65], vector vs ZDHHC14 LSSE PSD93α [−2.08, 4.32], vector versus ZDHHC14 LSSE PSD93β [5.62, 13.45], ZDHHC14 versus ZDHHC14 LSSE PSD93α [−6.42,–0.020], ZDHHC14 versus ZDHHC14 LSSE PSD93β [−16.11,–8.28]). In this Figure supplement, data from all HEK cell ABE experiments for ZDHHC14/PSD93 are plotted and analyzed together, while in Figure 1D and E data for the PSD93 alpha and beta subunits are analyzed separately where the N are equal across all conditions (N = 6 and N = 4, respectively).
Figure 1—figure supplement 4. Uncropped Western blot images for Figure 1.

Figure 1—figure supplement 4.

Bold titles indicate the figure that the uncropped images correspond to, boxes indicate cropped regions, and dashed lines and scissors indicate where membranes were cut prior to immunoblotting.