Fig 2: HGFAC overexpression enhances glucose homeostasis.(A) Immunoblot and quantification by densitometry of plasma HGFAC collected 3 days after 8-week-old male mice were transduced with adenovirus expressing GFP (ADV-GFP) or HGFAC (ADV-HGFAC). (B) Weights and lean and fat mass of ADV-GFP and ADV-HGFAC mice after 9 days of transduction (n = 10/group). (C) IP glucose tolerance test and corresponding iAUC performed 5 days after viral transduction (n = 8–9/group). (D) Overnight-fasted and 3-hour refed glycemia and peripheral insulin levels of GFP- and HGFAC-transduced mice (n = 10). (E) Hepatic mRNA levels of Hgfac, Pparg and -a, and PPARγ targets measured by qPCR 14 days after viral transduction. (F) Hepatic PPARγ, phospho-S293 PDHA, total PDHA, and PCNA immunoblots of liver from ADV-HGFAC– and ADV-GFP–transduced mice and quantification of PPARγ normalized to P85, phosphorylated PDHA normalized to total PDHA, and PCNA normalized to the total protein content (n = 4–5/group). (G) Hepatic and circulating triglyceride levels 14 days after viral transduction in ad libitum–fed mice. (H) Pparg mRNA levels in AML12 cells after overnight treatment with 50 ng/mL HGF or BSA. (I) c-MET phosphorylation by HGF in AML12 cells is inhibited by the c-MET inhibitor PHA-665752 (2.5 μM) preventing induction of Pparg mRNA. Data represent means ± SEM. Statistics assessed by 2-tailed unpaired t test, *P < 0.05; or by 2-way ANOVA with Holm-Šídák multiple comparisons between individual groups, ^P < 0.05 for comparison of effects of inhibitor within HGF treatment condition, $P < 0.05 for effect of HGF within inhibitor or control treatment.

Fig 3: The phenotype in HGFAC-KO mice recapitulates the phenotype of a putative loss-of-function variant in human HGFAC.(A) Schematic depiction of Hgfac gene and the deleted region in red; FWD and REV indicate the positions of forward and reverse primers, respectively, used in genomic PCR shown in (B) confirming the deletion of an 857 bp region in the Hgfac gene. (C) Representative immunoblot of circulating HGFAC in control (wild-type, littermate control) and KO (HGFAC KO) plasma. (D) Hepatic Hgfac mRNA levels measured by qPCR in control and HGFAC-KO mice (n = 7–9/group). (E) Immunoblot and quantification of phosphorylated c-MET in HepG2 cells treated with activated sera of control and HGFAC-KO mice (n = 3/condition). (F) Forest plot of phenotypes associated with the rs3748034 putative loss-of-function coding variant in human HGFAC. (G) Quantification of plasma triglyceride and cholesterol levels in ad libitum chow-fed male control and HGFAC-KO mice (n = 8–13/group), (H) plasma albumin concentrations in male control and HGFAC-KO mice (n = 11–17/group), and plasma platelet levels in male control and HGFAC-KO mice (n = 9–17/group). Data represent means ± SEM. Statistics were assessed by 2-tailed unpaired t test, *P < 0.05; or 1-way ANOVA with Holm-Šídák multiple comparisons test between groups, &P < 0.05.

Fig 4: ChREBP links nutritional status to circulating HGFAC.(A) ChIP was performed from livers of control and ChREBP-LKO mice with anti-ChREBP or control IgG. qPCR was performed on immunoprecipitated chromatin with primers spanning the E-box in the Pklr promoter and the putative ChREBP binding site in proximity to HGFAC and in nonspecific regions (neg) in proximity to both ChREBP response elements (n = 3/group). (B) Hepatic Chrebpβ and Hgfac mRNA expression of overnight-fasted and 4-hour chow- or HFrD-fed Wistar rats (n = 7/group). (C) Liver mRNA expression and (D) circulating levels of HGFAC in control and ChREBP-LKO mice after 8 weeks on chow versus HFrD with densitometric quantification (n = 4–5/group). (E) Correlation between HGFAC mRNA expression and a composite vector comprising canonical ChREBP targets in human livers from the GTEx project (Pearson’s correlation R2 = 0.44, P < 0.0001, n = 226). (F) Factors ranked by odds ratio for enrichment of the 300 genes most highly coexpressed with the factor in the ARCHS4 project that are also present in the top 5% of genes that correlate with HGFAC expression in the GTEx project. Combined score = log(P) × z, where P is calculated by Fisher’s exact test and z score is calculated by assessing the deviation from the expected rank. The size and color of the circles correspond to the enrichment score and adjusted P value, respectively. (G) Expression of HGFAC mRNA in livers of healthy controls, obese nondiabetic participants, and obese participantswith well-controlled diabetes and poorly controlled diabetes, (n = 4–5/group). Data represent means ± SEM. Statistics were assessed by 2-way ANOVA with Holm-Šídák multiple comparisons between individual groups, #P < 0.05, for main effects, ^P < 0.05 for comparison across genotypes within diets; or 1-way ANOVA with Holm-Šídák multiple comparisons test between control and other groups, &P < 0.05.

Fig 5: ChREBP-mediated activation of an HGFAC/HGF/PPARγ signaling axis mediates an adaptive response to preserve glucose tolerance in the setting of diets high in sugar.Glucose and fructose from high-sugar diets enhance production of sugar metabolites (hexose-phosphates) in the liver that activate hepatic ChREBP and lead to increased Hgfac transcription and translation. HGFAC is secreted into the circulation, where, once activated, it can act in a paracrine or endocrine fashion to proteolytically cleave and activate HGF. HGF binds and activates the c-MET tyrosine kinase receptor on hepatocytes and other cell types. In liver, this leads to upregulation of PPARγ expression that in turn activates transcriptional programs to promote hepatic triglyceride storage and to decrease circulating triglycerides. Additionally, hepatic PPARγ activity decreases activation of the pyruvate dehydrogenase complex, and this contributes to enhance systemic glucose tolerance.