TACE inhibition: a promising therapeutic intervention against AATF-mediated steatohepatitis to hepatocarcinogenesis

Abstract

Metabolic dysfunction-associated steatohepatitis-driven hepatocellular carcinoma (MASH-HCC) is a global clinical challenge for which there is a limited understanding of disease pathogenesis and a subsequent lack of therapeutic interventions. We previously identified that tumor necrosis factor-alpha (TNF-α) upregulated apoptosis antagonizing transcription factor (AATF) in MASH. Here, we investigated the effect of TNF-α converting enzyme (TACE) inhibition as a promising targeted therapy against AATF-mediated steatohepatitis to hepatocarcinogenesis. A preclinical murine model that recapitulates human MASH-HCC was used in the study. C57Bl/6 mice were fed with chow diet normal water (CD) or western diet sugar water (WD) along with a low dose of carbon tetrachloride (CCl₄; 0.2 μL·g^-1, weekly) for 24 weeks. TACE activity, TNF-α levels, and AATF expression were measured. The mice were treated with the TACE inhibitor Marimastat for 12 weeks, followed by analyses of liver injury, fibrosis, inflammation, and oncogenic signaling. In vitro experiments using stable clones of AATF control and AATF knockdown were also conducted. We found that AATF expression was upregulated in WD/CCl₄ mice, which developed severe MASH at 12 weeks and advanced fibrosis with HCC at 24 weeks. WD/CCl₄ mice showed increased TACE activity with reduced hepatic expression of sirtuin 1 (Sirt1) and tissue inhibitor of metalloproteinase 3 (Timp3). The involvement of the SIRT1/TIMP3/TACE axis was confirmed by the release of TNF-α, which upregulated AATF, a key molecular driver of MASH-HCC. Interestingly, TACE inhibition by Marimastat reduced liver injury, dyslipidemia, AATF expression, and oncogenic signaling, effectively preventing hepatocarcinogenesis. Furthermore, Marimastat inhibited the activation of JNK, ERK1/2, and AKT, which are key regulators of tumorigenesis in WD/CCl₄ mice and in AATF control cells, but had no effect on AATF knockdown cells. This study shows that TACE inhibition prevents AATF-mediated inflammation, fibrosis, and oncogenesis in MASH-HCC, offering a potential target for therapeutic intervention.

Keywords: Apoptosis antagonizing transcription factor; Marimastat; hepatocellular carcinoma; metabolic dysfunction associated steatohepatitis; tumor necrosis factor (TNF)‐alpha converting enzyme.

Fig. 1

Increased hepatic AATF expression in the progressive stages of MASLD. Mice were fed with either CD/CCl₄ or WD/CCl₄ for 12 and 24 weeks. After the completion of treatment, body weight (A), serum ALT (B), and serum AST (C) were estimated. (D) HOMA‐IR was calculated using fasting insulin and fasting glucose values using the standard formula: Fasting glucose (mg·dL⁻¹) X fasting insulin (mU·L⁻¹)/405. (E) Representative liver images of CD/CCl₄ or WD/CCl₄ mice at 12 and 24 weeks with histology and tumor burden. Arrows represent the tumor nodules. (F) Representative microscopic images of H&E‐stained liver sections depicting steatosis, hepatocyte ballooning, and lobular inflammation with MASLD activity score. Hepatic AATF expression as measured by qRT‐PCR and expression normalized to β‐Actin (G), western blot (68 kDa) (n = 3 mice per group) (H), and immunohistochemistry (I). The dotted white line in the western blot (H) indicates the vertically spliced images that are juxtaposed lanes from non‐adjacent gels. All the samples were derived from the same experiment and the blots were processed in parallel. Loading control (β Actin) is run on the same blot (n = 3 independent experiments). For all images, magnification: 400× And scale bar: 50 μm. Statistical significance was analyzed by a Student's t‐test. Data expressed as mean ± SEM for n = 6 mice per group except for (H). **P < 0.001 or *P < 0.05 compared to CD/CCl4. AATF, apoptosis antagonizing transcription factor; ALT, alanine transaminase; AST, aspartate aminotransferase; CCl₄, carbon tetrachloride; CD, chow diet; HOMA‐IR, homeostatic model assessment for insulin resistance; MASLD, metabolic dysfunction associated steatotic liver disease; WD, western diet.

Fig. 2

Evaluation of inflammatory cytokines, ER stress, and fibrosis markers in CD/CCl₄ and WD/CCl₄ mice. The relative hepatic mRNA expression of (A) IL‐1β, (B) IL‐6, (C) CHOP, and (D) Grp78 was estimated by qRT‐PCR and normalized to the endogenous control, β‐Actin. (E) Representative microscopic images of picrosirius red staining with fibrosis score (F) of mice treated with CD/CCl₄ and WD/CCl₄ for 12 and 24 weeks. (G) Immunostaining of desmin done in liver sections, and the relative hepatic mRNA expression of col1A1 (H), col3a1 (I), α‐SMA (J), and TGF β (K) as estimated by qRT‐PCR and expression normalized to the endogenous control, β‐Actin. For all images, magnification: 400× And scale bar: 50 μm. Statistical significance was analyzed by a Student's t‐test. Data expressed as mean ± SEM for n = 6 mice per group, n = 3 independent experiments. **P < 0.001 or *P < 0.05 compared to CD/CCl4. CCl₄, carbon tetrachloride; CD, chow diet; CHOP, C/EBP homologous protein; col1A1, collagen 1 alpha 1; col3a1, collagen 3 alpha 1; Grp78, 78 kDa glucose‐regulated protein; IL‐1β, interleukin‐1β; IL‐6, interleukin‐6; TGF β, transforming growth factor β; WD, western diet; αSMA, alpha Smooth Muscle Actin.

Fig. 3

Activation of oncogenic signaling in experimental HCC driven by MASH. Western blots were performed on the whole liver tissue lysates of CD/CCl₄ and WD/CCl₄ mice. Representative blot images and densitometric analysis of (A) pJNK and JNK (46 kDa), (B) pERK1/2 and ERK1/2 (44 kDa), (C) pAKT and AKT (62 kDa), (D) pSTAT3 and STAT3 (86 kDa) (n = 3 mice per group). The dotted white line indicates the vertically spliced images that are juxtaposed lanes from non‐adjacent gels. All the samples were derived from the same experiment and the blots were processed in parallel. Loading controls are run on the same blot. (E) Representative microscopic images of Ki67 immunostaining with quantification of Ki67‐positive cells; magnification: 400× And scale bar: 50 μm. Relative hepatic mRNA expression of (F) CD31 and (G) AFP were estimated by qRT‐PCR and expression normalized to the endogenous control, β‐Actin. Statistical significance was analyzed by a Student's t‐test. Data expressed as mean ± SEM for n = 6 mice per group, n = 3 independent experiments **P < 0.001 compared to CD/CCl4. AFP, alpha‐fetoprotein; CCl₄, carbon tetrachloride; CD, chow diet; CD31, cluster of differentiation 31; pAKT, phospho‐Atkin kinase; p‐ERK1/2, phospho‐extracellular signal regulated protein kinase; p‐JNK, phospho‐c‐Jun N‐terminal kinase; pSTAT3, phospho‐signal transducer and activator of transcription 3; WD, western diet.

Fig. 4

Evaluation of the role of the Sirt1/TIMP3/TACE axis in the progression of MASH to HCC. Mice were fed with either CD/CCl₄ or WD/CCl₄ for 12 and 24 weeks. TNF‐α was measured by qRT‐PCR (A) and ELISA (B). Tace mRNA (C) and TACE activity (D) were measured in CD/CCl₄ or WD/CCl₄ mice. Hepatic expression of Timp3 (E) and Sirt1 (F) was measured by qRT‐PCR and expression normalized to the endogenous control, β‐Actin. Statistical significance was analyzed by a Student's t‐test. Data expressed as mean ± SEM for n = 6 mice per group, n = 3 independent experiments. **P < 0.001 or *P < 0.05 compared to CD/CCl4. CCl₄, carbon tetrachloride; CD, chow diet; Sirt1, silent information regulator 1; Tace, tumor necrosis factor‐alpha converting enzyme; Timp3, tissue inhibitor of metalloproteinase 3; TNF‐α, tumor necrosis factor‐α; WD, western diet.

Fig. 5

The TACE inhibitor Marimastat prevents the progression of MASH‐associated liver tumors in WD/CCl₄ mice. (A) Study design: Mice were fed with CD/CCl₄ or WD/CCl₄ for 12 weeks and then treated with vehicle control or Marimastat for an additional 12 weeks. (B) TACE activity, (C) serum TNF‐α, (D) serum ALT, and (E) serum AST were measured. (F) HOMA‐IR was determined based on the values of fasting glucose and fasting insulin measurements. Representative liver histology images (G) with MASLD activity score and tumor burden (H) and microscopic images of picrosirius red staining (I) of CD/CCl₄ or WD/CCl₄ mice treated with or without Marimastat. For all images, magnification: 400× And scale bar: 50 μm. Hepatic mRNA expression of col1A1 (J), col3a1 (K), α‐SMA (L), and TGF β (M) as estimated by qRT‐PCR and expression normalized to the endogenous control, β‐Actin. Statistical significance was analyzed by a Student's t‐test. Data expressed as mean ± SEM for n = 6 mice per group, n = 3 independent experiments. **P < 0.001 to CD/CCl4 vehicle control; ^## P < 0.001 or ^# P < 0.05 compared to WD/CCl4 vehicle control. ALT, alanine transaminase; AST, aspartate aminotransferase; CCl₄, carbon tetrachloride; CD, chow diet; col1A1, collagen 1 alpha 1; col3a1, collagen 3 alpha 1; HOMA‐IR, homeostatic model assessment for insulin resistance; MAR, Marimastat; MASH, metabolic dysfunction associated steatohepatitis; MASH‐HCC, metabolic dysfunction‐associated steatohepatitis‐induced hepatocellular carcinoma; MASLD, metabolic dysfunction associated steatotic liver disease; TACE, tumor necrosis factor‐alpha converting enzyme; TGF β, transforming growth factor β; TNF‐α, tumor necrosis factor‐α; VC, vehicle control; WD, western diet; αSMA, alpha Smooth Muscle Actin.

Fig. 6

Effect of the TACE inhibitor Marimastat on AATF expression. Mice were fed with CD/CCl₄ or WD/CCl₄ for 12 weeks and then treated with vehicle control or Marimastat for an additional 12 weeks. AATF expression was measured by qRT‐PCR and expression normalized to β‐Actin (A), western blot (68 kDa; n = 3 mice per group) (B), and Immunohistochemistry; magnification: 400× And scale bar: 50 μm (C). Human HCC cells, QGY‐7703 cells, were treated with Marimastat, and mRNA (D) and protein (68 kDa) (E) levels of AATF were determined. (F) TACE activity and (G) TNF‐α were also measured in QGY‐7703 cells treated with or without Marimastat. Statistical significance was analyzed by a Student's t‐test. Data expressed as mean ± SEM for n = 6 mice per group, n = 3 independent experiments. ^## P < 0.001 or ^# P < 0.05 compared to WD/CCl4 or QGY‐7703 cells vehicle control. AATF, apoptosis antagonizing transcription factor; CCl₄, carbon tetrachloride; MAR, Marimastat; TACE, tumor necrosis factor‐alpha converting enzyme; TNF‐α, tumor necrosis factor‐α; VC, vehicle control; WD, western diet.

Fig. 7

Effect of TACE inhibition on AATF‐mediated MASH‐HCC. Mice were fed with CD/CCl₄ or WD/CCl₄ for 12 weeks and then treated with vehicle control or Marimastat for an additional 12 weeks. (A) qRT‐PCR analysis was carried out to determine the hepatic expression of IL‐1β, IL‐6, CHOP, Grp78, CD31, and AFP. Expression normalized to β‐Actin (B) Representative microscopic images of Ki67 immunostaining with quantification of Ki67‐positive cells. Arrows represent the Ki67 positive cells; magnification: 400× And scale bar: 50 μm (C) Immunoblot images of pJNK and JNK (46 kDa), pERK1/2 and ERK1/2 (44 kDa), pAKT and AKT (62 kDa), pSTAT3 and STAT3 (86 kDa) with their respective densitometric analyses normalized to total proteins (n = 3 mice per group). (D) Whole cell lysates from stable clones of QGY‐7703 cells‐AATF control and AATF knockdown cells treated with or without Marimastat were subjected to immunoblot analysis of pJNK (46 kDa), pERK1/2 (44 kDa), pAKT (62 kDa), and pSTAT3 (86 kDa). Bar graphs show the densitometric values calculated after normalization to total ERK1/2, JNK, AKT, and STAT3, respectively. Statistical significance was analyzed by a Student's t‐test. Data expressed as mean ± SEM, n = 3 independent experiments. ^## P < 0.001 or ^# P < 0.05 compared to WD/CCl4 or AATF control treated with vehicle. AFP, alpha‐fetoprotein; CCl₄, carbon tetrachloride; CD31, cluster of differentiation 31; CHOP, C/EBP homologous protein; Grp78; 78 kDa glucose‐regulated protein; IL‐1β, interleukin‐1β; IL‐6, interleukin‐6; KD, knockdown; MAR, Marimastat; pAKT, phospho‐Atkin kinase; p‐ERK1/2, phospho‐extracellular signal regulated protein kinase; p‐JNK, phospho‐c‐Jun N‐terminal kinase; pSTAT3, phospho‐signal transducer and activator of transcription 3; VC, vehicle control; WD, western diet.

Fig. 8

Schematic representation depicting the effect of TACE inhibition on MASH‐associated HCC mediated by AATF. Mice were fed with CD/CCl₄ or WD/CCl₄ for 12 weeks and then treated with vehicle control or Marimastat for an additional 12 weeks. TACE inhibition by Marimastat reduces proinflammatory cytokines and prevents the progression of MASH to HCC mediated by AATF. AATF, apoptosis antagonizing transcription factor; CCl₄, carbon tetrachloride; CD, chow diet; HCC, hepatocellular carcinoma; MASH, metabolic dysfunction associated steatohepatitis; Sirt1, silent information regulator 1; SREBP1, sterol regulatory element‐binding protein 1; TACE, tumor necrosis factor‐alpha converting enzyme; Timp3, tissue inhibitor of metalloproteinase 3; TNFR1, tumor necrosis factor‐ α receptor 1; TNF‐α, tumor necrosis factor‐α; WD, western diet.