Press releases

Basic Research Drives Innovations in Hepatology

08. May 2025

The high morbidity of chronic liver diseases that may develop into cancer calls for new options for diagnosis and treatment. At the Experimental Hepatology Days in Lyon (France), experts presented the latest findings from basic research that could be relevant for hepatological clinical practice. For example, antiretroviral treatment could delay alcohol-associated liver damage or make gene therapies for hereditary hepatopathies more effective. Artificial intelligence is optimizing diagnostics and, increasingly, prognosis assessment.

Research by Prof. Bernd Schnabl, La Jolla, CA (USA), and his team on the gut-liver axis in alcohol-associated liver disease (ALD) suggests that detectable endogenous retroviral elements (ERVs) in patients’ intestinal virome may play a key role in disease progression [1]. Schnabl explained that these “residual elements of retroviral infections in the human genome acquired during evolution” most likely stem from shed intestinal epithelial cells. While ERVs are not normally transcribed, they were activated in the intestinal epithelial cells of alcoholic patients, with ethanol alone being sufficient in the experimental setting. 

Additional model experiments revealed interesting pathomechanisms: The activated ERVs trigger the transcription factor Z-DNA binding protein 1 (ZBP1) in the intestinal epithelium. ZBP1 induces programmed cell death through necroptosis via the mixed lineage kinase domain-like protein (MLKL). According to Schnabl, the death of the intestinal epithelial cells promotes the translocation of proliferating gram-negative bacteria, particularly Escherichia coli, via a still unidentified tight junction-independent mechanism. This mechanism contributes to disease progression in the liver. 

In the absence of ZBP1, or if ERV activation in the intestinal epithelium is experimentally blocked by antiretroviral therapies such as emtricitabine, tenofovir, and nevirapine, this counteracts intestinal epithelial necroptosis and bacterial translocation. “When we use the antiretroviral substances, we see less liver damage, less liver steatosis, and less liver inflammation,” Schnabl said. Stabilizing the gut barrier in this way could also help prevent alcohol-related liver damage.

Refined gene therapies for Wilson’s disease and liver fibrosis 

The research team led by Dr. Pasquale Piccolo, Pozzuoli (Italy), has developed new strategies for gene therapy in patients with Wilson’s disease, a hereditary disorder of copper metabolism. The full-length ATP7B gene exceeds the cargo capacity of adeno-associated viruses used as vectors. For this reason, a reduced “mini-ATP7B” is usually transferred. As an alternative, Piccolo and his team used split-intein technology to transfer the DNA for ATP7B divided into two vectors and reconstitute them in the cell [2]. “Split inteins are proteins with bacterial origin that mediate the fusion of two separate polypeptides into a single protein through a non-enzymatic route,” Piccolo explained. In preclinical models, copper metabolism normalized and reduced the liver damage.

To circumvent the vector dilution effect when transgenes integrated purely episomally in target cells are not passed on to daughter cells during cell division, the researchers also transferred a mini-ATP7B using nuclease-free genome editing [3]. Instead of an artificial promoter, they used an albumin-gene homologue. The therapeutic gene could therefore be integrated directly into the genome under the control of the natural, liver-specific, and highly active albumin promoter, enabling long-lasting and highly reliable expression. In the mouse model, this was particularly successful in young animals. “Up to 90% of the liver expressed the transgene,” Piccolo reported.

In patients with liver fibrosis, structural and functional changes impair the distribution and entry of many conventional AAV vectors such as AAV8 and AAV9 into hepatocytes. However, some synthetic capsids, such as AAV-KP1, appear to remain effective and, according to Piccolo, may be a viable alternative [4].

Glycine depletion makes fatty liver susceptible to oxidative stress

Altered one-carbon metabolism, specifically a functional reversal of mitochondrial serine-hydroxymethyltransferase (SHMT2), appears to contribute to the deterioration of liver function in metabolic dysfunction-associated fatty liver disease (MASLD) [5]. These were the findings of the working group led by Dr. Eyal Gottlieb, Houston, TX (USA), who used metabolome analyses and isotope tracing in a mouse model [5]. “Instead of the SHMT converting serine into glycine, it does exactly the opposite. It depletes glycine by forming large quantities of serine,” Gottlieb said, sharing the surprising observation. 

Glycine depletion impaired downstream metabolic pathways, including glutathione synthesis. “Everything downstream of glycine appears to be affected,” Gottlieb said. Glutathione is an important radical scavenger that contributes to the neutralization of medications, xenobiotics, and reactive lipids. Because glycine deficiency reduced the availability of glutathione and, in turn, the liver’s capacity to compensate for oxidative stress, massive lipid peroxidation occurred in the MASLD mouse model of a paracetamol overdose. This led to cell necrosis, hemorrhage, and increased transaminases in the previously damaged liver. Compared to the control group with a healthy liver, mortality was significantly increased. Glycine supplementation or inhibition of SHMT2 restored the glycine and glutathione pool and protected the mice effectively from paracetamol-induced liver damage. 

Biliary duct carcinoma or liver metastasis: Will AI see this in the future?

Rare diseases such as hepatocellular carcinoma (HCC) or cholangiocellular carcinoma (CCA) are not yet the focus of manufacturers of artificial intelligence (AI)-based systems. However, the working group led by Prof. Julien Calderaro, Créteil (France), developed AI-assisted models several years ago that are able to predict survival based on histological slides from HCC resections [6]. 

There are no established predictive biomarkers for liver cancer yet, although researchers have identified molecular signatures in patients with advanced HCC that are associated with a therapeutic response to combination therapy with the anti-programmed death ligand 1 (PD-L1) antibody atezolizumab and bevacizumab, which targets vascular endothelial growth factor (VEGF) [7]. Calderaro’s team developed an AI model that detects the expression of this atezolizumab-bevacizumab response signature (ABRS) directly on tissue sections [8]. 

Indeed, Calderaro’s combination of matched AI heatmaps and spatial transcriptomics proved suitable for predicting progression-free survival in a real-life cohort. “Unfortunately, the model was not predictive for overall survival,” Calderaro noted. He is currently studying the extent to which UNI, a foundation model for feature extraction that interprets pathology images and conventional multiple-instance learning, can distinguish intrahepatic CCA from liver metastases. Among the first 29 of 400 planned biopsies AI was mistaken in only 2 cases. 

Quelle:
Symposium 240 „Experimental Hepatology Days“. Lyon (France), April 24-26, 2025 
Organization: Falk Foundation e.V.

Literatur:

  1. Jiang L et al., Hepatology. 2020 Dec;72(6):2182-2196.
  2. Padula A et al., Mol Ther Methods Clin Dev. 2022 Aug 13;26:495-504.
  3. Padula A et al., JCI Insight. 2023 Nov 8;8(21):e171281.
  4. Ferriero R et al., Nat Commun. 2025 Mar 10;16(1):2118.
  5. Ghrayeb A et al., Cell Metab. 2024 Jan 2;36(1):116-129.e7.
  6. Saillard C et al., Hepatology. 2020 Dec;72(6):2000-2013.
  7. Zhu AX et al., Nat Med. 2022 Aug;28(8):1599-1611.
  8. Zeng Q et al., Lancet Oncol. 2023 Dec;24(12):1411-1422.

Scientific Organization:

Matías Ávila, Pamplona (Spain)
Frank Tacke, Berlin (Germany)
Robert Thimme, Freiburg (Germany)
Fabien Zoulim, Lyon (France)
Jessica Zucman-Rossi, Paris (France)


Contact:

eickhoff kommunikation
Probsteigasse 15
50670 Cologne
Germany
Contact person: Barbara Steinberg
Tel.: 0221-995951-20
Mobil: 0179 118 93 79
Fax: 0221-995951-99
E-Mail: steinberg@eickhoff-kommunikation.de