Fig 2: Alteration in protein levels of DN-forming proteins in APPNL-G-F mouse brains and in vitro Aß42 treated cells/neurons. A Protein lysates from the cortex of 3-month (3 m) and 9-month (9 m) old WT and APPNL-G-F mice were prepared for Western blot analysis with antibodies to LAMP1, PSAP, SAP-C, SAPs, ATG9A and calnexin. B Quantification for each protein band was conducted using Fiji software and the average of each age group was normalized to the calnexin level. C Lysates prepared from N2a cells treated with vehicle only (0) or Aß42 (0.5 to 4 µM) for 16 h (h) were subjected to Western blot analysis. D Band intensity for each protein was measured and normalized to the intensity of ß-actin. E, F Primary neurons, cultured form the cortical and hippocampal regions, were also treated with vehicle or 1 µM Aß42 for 16 h (h), and protein lysates were examined by Western blot analysis (E) and similar quantification (F). N = 3 independent experiments in all western analyses (* P < 0.05, ** P < 0.01, student t-test)

Fig 3: PSAP and PGRN levels are decreased in cingulate cortex tissue of schizophrenia patients. (A) Representative immunoblots of PSAP, PGRN, and ß-actin in cingulate cortex tissue from subjects of healthy control (Ctrl), schizophrenia (SCZ), bipolar disorder (BP), and major depressive disorder (MDD). Upper panel, illustration of cingulate cortex in human brain. (B,C) Bar graphs presenting quantification of the ratio of protein levels of PSAP/ß-actin (B) and PGRN/ß-actin (C). (D,E) Scatter plots showing the Pearson’s correlation analysis of PSAP and lifetime quantity of fluphenazine or equivalent (FLU) in schizophrenia (D) and bipolar disorder group (E), respectively. Each dot represents one patient of SCZ (blue) or BP (magenta). (F,G) Scatter plots showing the Pearson’s correlation analysis of PGRN and FLU in schizophrenia (F) and bipolar disorder group (G), respectively. One-way ANOVA with Sidak’s post hoc test. N = 11–12 samples in each group. Data are presented as mean ± S.E.M. * p ? 0.05 compared to Ctrl.

Fig 4: Conditional deletion of neuronal PSAP in the cingulate cortex causes downregulation of neuronal PGRN. (A) Representative fluorescent in situ hybridization images of PSAP mRNA in the cingulate cortex of mice injected with AAV-GFP and AAV-GFP-Cre. Upper row, schematic depiction of stereotaxic injection sites in one hemisphere (the same in the other hemisphere). Scale bar, 50 µm. (B) Representative immunofluorescent staining images of GFP (cyan), Cre (orange), PSAP (magenta), and PGRN (yellow) in the cingulate cortex of mice injected with AAV-GFP and AAV-GFP-Cre, respectively. Scale bar, 20 µm. (C,D) Bar graphs showing quantification of mean fluorescence intensity (MFI) of PSAP (C) and PGRN (D) staining in GFP- or Cre-positive cells. A.U, arbitrary unit. Student’s t-test, N = 34 cells in each group. Data are presented as mean ± S.E.M. **** p ? 0.0001.

Fig 5: The catalytic activities of STT3A or STT3B are not required for DENV replication. (A) Huh7.5.1 cells were transduced with a lentivirus encoding a puromycin resistance marker, Cas9, and a single guide RNA (sgRNA) targeting the indicated genes. The sgRNAs targeting GFP act as a negative control. Two independent sgRNAs (indicated by 1 or 2 below the bars in the bar graph) were used per target gene to generate knockout cell pools. Cells were infected with luc-DENV for 3 days, and luciferase activity was measured in relative light units (RLU). Western blots for the indicated proteins are shown below the bar graph. Arrows indicate nonspecific bands. (B and C) Knockout cell pools were transduced with lentiviral vectors to rescue expression of the indicated protein. “STT3A dead” is a catalytically inactive W526A D527A double mutant. “STT3B dead” is a catalytically inactive W605A D606A double mutant with a C-terminal V5 tag. Transient transfection of constructs to express pSAP and SHBG were used to assess STT3A and STT3B catalytic activity. Western blots below the bar graph show the expression of the indicated proteins. wt and WT, wild type. (D) Knockout cell pools were transiently transfected with a luc-DENV replicon, and luciferase activity was measured at 4 h, 48 h, and 72 h posttransfection (hpt). Data are plotted as a ratio of RLU to the RLU at 4 hpt to control for transfection efficiency. GDD indicates a polymerase dead replicon with replacement of the GNN active site sequence in the NS5 RNA polymerase by GDD. (E) Cell viability for knockout cell pools was measured by Cell-Titer Glo over the course of 4 days for three independent wells. Data are normalized to day 1 values. The values for the parental Huh7.5.1 cells (P) are shown. In panels A to E, data are expressed as means plus standard deviations (SD) (error bars) for three independent biological replicates. In panels A to D, statistical significance was determined by Student’s t test and indicated by asterisks as follows: *, P < 0.05; **, P < 0.005; ***, P < 0.0005. In panel A, the means were compared to the mean for GFP sgRNA 2. In panels B and C, the means were compared to the means of their respective knockouts. In panel D, the means were compared to the mean for the GFP control.