Fig 2: CRISPR-directed editing of AKAP12’s activation-responsive phospho-sites enhances AKAP12’s scaffolding activity and inhibits HSC activation.(A) Cell extracts from human HSCs cultured for 0, 3, or 6 days (see Materials and methods) were processed for co-immunoprecipitation of AKAP12 and HSP47 or for a-SMA western blotting. Data represented as GAPDH normalized densitometry is mean ± SE from three experiments. Source data are presented in Figure 2—source data 1. P values are calculated in Figure 2—source data 4. (B) Activated human HSCs were transfected with CRISPR reagents and GFAP-SaCas9 vector to cause CRISPR-directed HDR as in Materials and methods. Untransfected (WT) or cells with SaCas9 alone were used as controls. CRISPR editing at the AKAP12 locus (left panel) was confirmed by performing PCR (right panel) using primers that specifically detected the edited region as listed in Key resource table. Four independent experiments are shown. (C) CRISPR-edited (HDR) or WT cells as in (B) above were assessed for AKAP12-HSP47 co-immunoprecipitation, HSP47, AKAP12, and a-SMA (HSC activation marker) western blotting. Data represented as GAPDH normalized densitometry is mean ± SE from three experiments. Source data are presented in Figure 2—source data 2. P values are calculated in Figure 2—source data 4. (D) Day 0 attached HSCs were culture activated till Day 3 and then transfected with CRISPR vectors till Day 5. The autofluorescence of vitamin A as a marker of HSC quiescence was visualized by fluorescence microscopy and compared to brightfield images of cells as in Materials and methods. Three independent experiments are shown. Scale bar=80 µm. Source data is presented in Figure 2—source data 5. (E) AKAP12-HSP47 interaction (left panel) and HSP47-collagen interaction (right panel) in the ER was compared between WT and HDR cells by PLA staining and co-staining with the ER marker, calreticulin as in methods. Magnification at 200×, scale bar=60 µm. Data represented as proximity ligation/fluorescence count are mean ± SE from four experiments. Source data are presented in Figure 2—source data 3. P values are calculated in Figure 2—source data 4. ER, endoplasmic reticulum; HDR, homology-directed repair; HSC, hepatic stellate cell; PLA, proximity ligation assay; WT, wild-type. Figure 2—source data 1.Source blots for Figure 2A. Figure 2—source data 2.Source blots for Figure 2C. Figure 2—source data 3.Source data for Figure 2E. Figure 2—source data 4.Post-hoc analysis for Figure 2. Figure 2—source data 5.Source data for Figure 2D.

Fig 3: PKCa phosphorylates AKAP12 and inhibits its interaction with HSP47.(A) AKAP12 is phosphorylated by PKCa at its activation-responsive phospho-sites. Recombinant WT or AKAP12 phospho-mutants were in vitro translated from their vectors and subjected to in vitro kinase assay in the presence of active PKCa enzyme as in Materials and methods. The reactions were run on a phostag gel to detect phosphorylated AKAP12 or its mutants. Representative phostag gels from three experiments are shown. Source data are presented in Figure 3—source data 1. P values are calculated in Figure 3—source data 4. (B) Direct Interaction between AKAP12 and HSP47 in recombinant system in the absence or presence of active PKCa enzyme. In vitro translated biotinylated AKAP12 was incubated with recombinant HSP47 antibody column containing bound HSP47 (left) or recombinant HSP47 was incubated with Biotin antibody column containing bound AKAP12-Biotin (right) in the presence or absence of active PKCa as in Materials and methods. Two representative data out of four experiments are shown. Source data are presented in Figure 3—source data 2. (C) Silencing Prkca in activated human HSCs enhances AKAP12-HSP47 interaction. Culture-activated human HSCs were transfected with a universal negative control siRNA (Neg) or two Prkca siRNAs (A or B) as in Materials and methods. Total protein was assessed for AKAP12-HSP47 co-immunoprecipitation or PKCa, HSP47, and GAPDH immunoblotting. Data represented as GAPDH normalized densitometry are mean ± SE from five experiments. Source data are presented in Figure 3—source data 3. P values are calculated in Figure 3—source data 4. HSC, hepatic stellate cell; WT, wild-type. Figure 3—source data 1.Source blots for Figure 3A. Figure 3—source data 2.Source blots for Figure 3B. Figure 3—source data 3.Source blots for Figure 3C. Figure 3—source data 4.Post hoc analysis for Figure 3C.

Fig 4: Phospho-editing of AKAP12 regulates AKAP12’s HSP47-scaffolding activity, HSC activation, and HSP47’s collagen-chaperoning activity in the CCl4 mouse model.(A) AKAP12-HSP47 co-immunoprecipitation, AKAP12, HSP47, and a-SMA western blotting from liver protein of CR-PDEL experiment. Data represented as GAPDH normalized densitometry are mean±SE from six experiments. Three representatives are shown. Source data are presented in Figure 6—source data 1. P values are calculated in Figure 6—source data 4. (B) AKAP12-HSP47 co-immunoprecipitation, AKAP12, HSP47, and a-SMA western blotting from liver protein of CR-PMUT experiment. Data represented as GAPDH normalized densitometry are mean±SE from six experiments. Three representatives are shown. Source data are presented in Figure 6—source data 1. P values are calculated in Figure 6—source data 4. (C) Co-immunoprecipitation of collagen with HSP47 in CRISPR-edited livers are mean±SE from five PDEL and three PMUT experiments. Source data are presented in Figure 6—source data 3. P values are calculated in Figure 6—source data 4. (D) Col1A1 or Acta2 mRNA levels by real-time PCR from PDEL (left) or PMUT (right) mouse livers. Data are mean±SE from six experimental groups. P values are calculated in Figure 6—source data 4. (E) Interaction between AKAP12 and HSP47 in desmin-positive HSCs of CRISPR model by PLA staining. Data representative of the AKAP12-HSP47 PLA count per desmin area are mean±SE from four PDEL or PMUT experiments. 200× magnification, scale bar=100 µm. P values are calculated in Figure 6—source data 4. HSC, hepatic stellate cell; PLA, proximity ligation assay. Figure 6—source data 1.Source blots for Figure 6A. Figure 6—source data 2.Source blots for Figure 6B. Figure 6—source data 3.Source blots for Figure 6C. Figure 6—source data 4.Post hoc analysis for Figure 6.

Fig 5: HSC-specific phospho-editing of AKAP12 regulates the ER stress response.(A) Heat map of total liver and HSCs ER stress/UPR signaling components in four groups, oil+EV, oil+CR, CCl4+EV and CCL4+CR. Proteomics data utilized to prepare the heatmap are presented in Supplementary file 4. Alterations in ER stress responsive elements in HSCs from AKAP12 PDEL CRISPR model (B) and total liver (C). GAPDH normalized densitometry is mean±SE from three experiments. Raw densitometry of each experiment is presented in Figure 7—figure supplement 1. Source data are presented in Figure 7—source data 1 and Figure 7—source data 2. P values are calculated in Figure 7—source data 5. (D) Inflammatory cytokine expression in HSCs from oil+EV, oil+CR, CCl4+EV and CCL4+CR groups. GAPDH normalized densitometry is mean±SE from four experiments. Raw densitometry of each experiment is presented in Figure 7—figure supplement 1. Source data are presented in Figure 7—source data 3. P values are calculated in Figure 7—source data 5. (E) ER stress response in hepatocytes co-cultured with quiescent or activated WT HSCs with or without AKAP12 PDEL CRISPR editing (CR). GAPDH normalized densitometry represented as fold over hepatocytes +WT quiescent HSCs is mean±SE from three experiments. Raw densitometry of each experiment is presented in Figure 7—figure supplement 1. Source data are presented in Figure 7—source data 4. P values are calculated in Figure 7—source data 5. (F) Summary of findings. AKAP12 interacts with HSP47 in the ER of normal HSCs and negatively regulates HSP47’s collagen-chaperoning activity and its ability to promote ER stress-directed IRE1a branch of UPR signaling. AKAP12 negatively regulates HSC activation. Pro-fibrogenic stimuli that cause HSC activation allow AKAP12’s PKCa-dependent site-specific phosphorylation. By AKAP12 CRISPR phospho-editing, we show that AKAP12 phosphorylation inhibits AKAP12’s HSP47 scaffolding activity, increases HSP47-collagen chaperoning activity and induces HSP47’s interaction with UPR signals (IRE1a and P-IRE1a). AKAP12 phosphorylation may lead to increased downstream events associated with the UPR signaling such as BIP-collagen chaperoning, phosphorylation of SMADs/P38MAPK and UPR-regulated inflammatory signaling. Blocking AKAP12 phosphorylation in activated HSCs prevents the ER stress response in hepatocytes co-cultured with activated HSCs. ER, endoplasmic reticulum; HSC, hepatic stellate cell; UPR, unfolded protein response; WT, wild-type. Figure 7—source data 1.Source blots for Figure 7B. Figure 7—source data 2.Source blots for Figure 7C. Figure 7—source data 3.Source blots for Figure 7D. Figure 7—source data 4.Source blots for Figure 7E. Figure 7—source data 5.Post hoc analysis for Figure 7.