MCLR-elicited hepatic fibrosis and carcinogenic gene expression changes persist in rats with diet-induced nonalcoholic steatohepatitis through a 4-week recovery period

Abstract

Nonalcoholic steatohepatitis (NASH) causes liver extracellular matrix (ECM) remodeling and is a risk factor for fibrosis and hepatocellular carcinoma (HCC). Microcystin-LR (MCLR) is a hepatotoxin produced by fresh-water cyanobacteria that causes a NASH-like phenotype, liver fibrosis, and is also a risk factor for HCC. The focus of the current study was to investigate and compare hepatic recovery after cessation of MCLR exposure in healthy versus NASH animals. Male Sprague-Dawley rats were fed either a control or a high fat/high cholesterol (HFHC) diet for eight weeks. Animals received either vehicle or 30 μg/kg MCLR (i.p: 2 weeks, alternate days). Animals were euthanized at one of three time points: at the completion of the MCLR exposure period and after 2 and 4 weeks of recovery. Histological staining suggested that after four weeks of recovery the MCLR-exposed HFHC group had less steatosis and more fibrosis compared to the vehicle-exposed HFHC group and MCLR-exposed control group. RNA-Seq analysis revealed dysregulation of ECM genes after MCLR exposure in both control and HFHC groups that persisted only in the HFHC groups during recovery. After 4 weeks of recovery, MCLR hepatotoxicity in pre-existing NASH persistently dysregulated genes related to cellular differentiation and HCC. These data demonstrate impaired hepatic recovery and persistent carcinogenic changes after MCLR toxicity in pre-existing NASH.

Keywords: Carcinogenesis; Fibrosis; Microcystin-LR; Nonalcoholic fatty liver disease.

Figure 1:. Representative H&E-stained liver samples.

Panels A-F represent control liver samples and G-L represent HFHC liver samples. Arrowheads indicate inflammation, arrows indicate necrosis and (*) represents steatosis. Animals were euthanized at specific times after the final vehicle or MCLR administration (24 hrs [0 weeks], recovery 2 weeks, or recovery 4 weeks). Original magnification was 100X.

Figure 2:. Histology severity scores.

(A) Steatosis, (B) Fibrosis, (C) Random inflammation, (D) Centrilobular inflammation, (E) Random necrosis, (F) Centrilobular necrosis, scores provided by the pathologist. Each point represents an individual animal in the group, n=6 per group. The horizontal lines represent the median for each group. CV= Control vehicle; CM= Control MCLR; HV= HFHC vehicle; HM= HFHC MCLR.

Figure 3:. Liver cytokine levels.

(A) IL-2, (B) IL-4, (C) IL-6, (D) IL-10, (E) IL-17A, (F) CCL11, (G) CXCL1, (H) CXCL2, (I) IFNγ, (J) VEGF-A, (K) IL-18, (L) CX3CL1, (M) CCL5, (N) CXCL5 and (O) IL-1β Data represent mean ± SD, n = 6 for each group. Two-way ANOVA with Sidak multiple comparison post-test: *p-value < 0.05 versus vehicle treatment group within each diet group and recovery period. One-way ANOVA with Dunnett multiple comparison post-test: #p-value < 0.05 versus 0 weeks within each diet and treatment group.

Figure 4:. Total differentially expressed genes.

Circles represent the number of differentially expressed genes (control MCLR versus control vehicle; blue) (HFHC MCLR versus HFHC vehicle; red). Overlapping portions represent the genes differentially expressed by MCLR in both control and HFHC groups.

Figure 5:. Representative liver Masson’s trichrome staining.

Panels A-F represent Control diet liver samples and G-L represent HFHC diet liver samples. The blue stain represents the collagen deposition depicting fibrosis. Animals were euthanized at specific times after the final vehicle or MCLR dose (24 hrs [0 weeks], recovery 2 weeks, or recovery 4 weeks). Original magnification was 100X.

Figure 6:. HSC activation and fibrosis markers.

(A) Actin alpha 2 (Acta2), (B) vascular endothelial growth factor D (Vegfd), (C) transforming growth factor beta-3 (Tgfb3), (D) platelet derived growth factor D (Pdgfd), (E) interferon gamma (Ifng), and (F) interleukin 10 (Il-10). Gene expression changes were analyzed at each time point the control or HFHC diet fed animals were euthanized after the final vehicle or MCLR dose (24 hrs [0 weeks], recovery 2 weeks, or recovery 4 weeks). Gene expression was determined by RNAseq and normalized to the median of the control vehicle group for each gene. *Adjusted p-value < 0.05 versus respective vehicle treated group for each gene. (G-H) Western blot for α-SMA. Data represent mean ± SD, n = 6 for each group. Two-way ANOVA with Sidak multiple comparison post-test: *p-value < 0.05 versus vehicle treatment group within each diet group and recovery period. One-way ANOVA with Dunnett multiple comparison post-test: #p-value < 0.05 versus 0 weeks within each diet and treatment group.

Figure 7:. Matrisome differential gene expression.

Control MCLR is compared to control vehicle and HFHC MCLR is compared to HFHC vehicle. The groups labelled are: C= Collagens; Gp= ECM glycoproteins; Pg= Proteoglycans; AP= ECM affiliated proteins; R= ECM regulators; SF= Secreted factors. Log fold change represents the Log₂ Fold Change (A and C) and p-adjusted is the DESeq2 adjusted (Benjamini Hochberg) p-value for differential expression (B and D).

Figure 8:. Differential gene expression of genes involved in carcinogenesis.

(A) Sushi repeat-containing protein X-linked 2 (Srpx2), (B) semaphorin 3A (Sema3), (C) protein tyrosine kinase 7 (Ptk7), (D) bradykinin receptor B2 (Bdkrb2), (E) dual specificity tyrosine phosphorylation kinase 4 (Dyrk4), (F) podoplanin (Pdpn), (G) mesothelin (Msln), and (H) gamma-synuclein (Sncg). Gene expression changes were analyzed at each time point the control or HFHC diet fed animals were euthanized after the final vehicle or MCLR dose (24 hrs [0 weeks], recovery 2 weeks, or recovery 4 weeks). Gene expression was determined by RNAseq and normalized to the median of the control vehicle group for each gene. *Adjusted p-value < 0.05 versus respective vehicle treated group for each gene.

Figure 9:. Differential gene expression of genes involved in cellular differentiation.

(A) Epithelial cell adhesion molecule (Epcam), (B) neurotensin (Nts), (C) brain-expressed X-linked 1 (Bex1), (D) fermitin family member 1 (Fermt), (E) interleukin 22 receptor subunit alpha 1 (Il22ra1), (F) carboxypeptidase A (Cpa), (G) cluster of differentiation 24 (Cd24), (H) cluster of differentiation 133 (Cd133), and (I) KIT ligand (Kitlg). Gene expression changes were analyzed at each time point the control or HFHC diet fed animals were euthanized after the final vehicle or MCLR dose (24 hrs [0 weeks], recovery 2 weeks, or recovery 4 weeks). Gene expression was determined by RNAseq and normalized to the median of the control vehicle group for each gene. *Adjusted p-value < 0.05 versus respective vehicle treated group for each gene.

Figure 10:. Representative IHC images of Gstp foci.

Panels A-F represent control diet liver samples and G-L represent HFHC diet liver samples. Animals were euthanized at specific times after the final vehicle or MCLR dose (24 hrs [0 weeks], recovery 2 weeks, or recovery 4 weeks). Original magnification was 200X.

Figure 11:. Gstp foci area.

Gstp foci area was quantified for each treatment group and time. Data represent mean ± SD, n = 6 for each group. Two-way ANOVA with Sidak multiple comparison post-test: *p-value < 0.05 versus vehicle treatment group within each diet group and recovery period.