Eight Immune Genes Linked to Severe Liver Scaring in Metabolic Disease
Biology

Eight Immune Genes Linked to Severe Liver Scaring in Metabolic Disease

An integrative analysis of gut metabolite pathways and single-nucleus data reveals how immune and stellate cells interact during liver fibrosis progression.

By Priya Sharma
Published:
Email this Article
Eight Immune Genes Linked to Severe Liver Scaring in Metabolic Disease
Photo by julien Tromeur on Unsplash

Human liver cells show distinct shifts in immune-related genes as metabolic liver injury advances toward severe fibrosis, according to an integrative transcriptomic analysis published in PLOS One.

Connecting Gut Metabolites to Liver Fibrosis

Metabolic dysfunction-associated steatotic liver disease, or MASLD, affects millions of people globally and ranges from simple fat accumulation to severe scarring known as cirrhosis. Because routine blood tests often fail to catch advanced scarring, researchers have searched for molecular clues that better reflect the underlying damage. In the new study, investigators set out to discover whether host genes tied to gut microbial metabolites and immune responses could help distinguish early liver fibrosis from advanced disease.

The research team began by combining two major biological databases to build a candidate list of 72 genes linked to both gut metabolites and immune functions. They then analyzed liver tissue RNA sequencing data from 206 patients with non-alcoholic fatty liver disease, comparing 138 individuals with early fibrosis against 68 patients with advanced fibrosis. Out of the 72 candidate genes, 23 showed significant differences in expression between the two groups. Using advanced machine-learning algorithms, the team narrowed this down to an eight-gene signature featuring CCL2, MYC, CASP1, CAT, DEFB1, NLRP6, HMOX1, and SMURF2.

Fig 1: Overall workflow of the integrative analysis for identifying microbiota–metabolite-associated immune genes related to MASLD fibrosis severity. Discovery compared 138 non-advanced and 68 advanced NAFLD cases. The gutMGene/ImmPort overlap yielded a 72-gene biological prior; 57 genes were testable, and 23 passed transcriptome-wide FDR screening. LASSO and balanced Random Forest selected eight features. Full-pipeline internal assessment and fixed-model external assessment were kept distinct. GSE244832 supplied cellular localization and descriptive communication context; its disease groups were not converted to the bulk fibrosis endpoint.
Overall workflow of the integrative analysis for identifying microbiota–metabolite-associated immune genes related to MASLD fibrosis severity. Discovery compared 138 non-advanced and 68 advanced NAFLD cases. The gutMGene/ImmPort overlap yielded a 72-gene biological prior; 57 genes were testable, and 23 passed transcriptome-wide FDR screening. LASSO and balanced Random Forest selected eight features. Full-pipeline internal assessment and fixed-model external assessment were kept distinct. GSE244832 supplied cellular localization and descriptive communication context; its disease groups were not converted to the bulk fibrosis endpoint. Credit: Zhi-yun Liu et al. / PLOS One

Cellular Whereabouts of the Eight Signature Genes

To understand which liver cells produce these eight signature genes, the authors turned to single-nucleus RNA sequencing data encompassing 128,223 nuclei from 18 donors. This high-resolution mapping revealed that the genes are far from uniformly distributed. For instance, CCL2 showed its highest average expression in fibroblast and stellate cells, which are the primary drivers of scar tissue production in the liver. Meanwhile, CASP1 and HMOX1 were most prominent in macrophage and Kupffer cells, which form the liver’s frontline immune defense.

Further analysis showed that CAT was primarily expressed in hepatocytes, the main functional cells of the liver, while MYC and DEFB1 peaked in cholangiocytes, the cells lining the bile ducts. These distinct cellular locations suggest that bulk liver tissue measurements capture a complex chorus of signals coming from multiple interacting cell types rather than a single source.

Fig 2: Transcriptomic landscape of MASLD fibrosis severity in GSE135251. (A) PCA based on variance-stabilized expression in all 206 primary NAFLD cases. (B) Global differential expression; thresholds are transcriptome-wide adjusted P < 0.05 and |log2 fold change| > 0.5. (C) The top 25 upregulated and top 25 downregulated global DEGs, ranked within each direction by adjusted P value and shown across all 206 patients using gene-wise standardized VST values; genes are rows and patients are columns. (D) The ten GO Biological Process terms with the smallest BH-adjusted P values, using the defined gene background; GeneRatio denotes the overlap divided by 618 GO-mapped DEGs, bubble size denotes overlap count, and color denotes adjusted P value.
Transcriptomic landscape of MASLD fibrosis severity in GSE135251. (A) PCA based on variance-stabilized expression in all 206 primary NAFLD cases. (B) Global differential expression; thresholds are transcriptome-wide adjusted P < 0.05 and |log2 fold change| > 0.5. (C) The top 25 upregulated and top 25 downregulated global DEGs, ranked within each direction by adjusted P value and shown across all 206 patients using gene-wise standardized VST values; genes are rows and patients are columns. (D) The ten GO Biological Process terms with the smallest BH-adjusted P values, using the defined gene background; GeneRatio denotes the overlap divided by 618 GO-mapped DEGs, bubble size denotes overlap count, and color denotes adjusted P value. Credit: Zhi-yun Liu et al. / PLOS One

Immune-Stromal Communication Across Disease States

The researchers also investigated how immune and stromal cells talk to each other across different stages of the disease using computational cell-to-cell communication tools. The analysis indicated that tissue samples from patients with non-alcoholic steatohepatitis featured a substantially larger and more active network of inferred molecular interactions than normal liver samples. Macrophages and stellate cells maintained two-way communication in all disease conditions, with signals flowing from stellate cells to macrophages making up a large share of the identified crosstalk.

While the eight-gene model demonstrated strong statistical performance when tested on the primary patient group, the authors noted several important limitations. When tested on an external mixed cohort of 78 liver biopsies, some genes did not replicate their original direction or significance, and one gene showed an opposite association. Furthermore, the database connections linking gut microbes to these host genes represent previously reported associations rather than proven causal pathways in the study participants. The investigators emphasize that the signature remains a research-stage tool that requires extensive prospective evaluation before it can guide clinical care.

The research was published in PLOS One on November 16, 2026.

Fact Checked

This article has been fact checked for accuracy, with information verified against reputable sources. Learn more about us and our editorial process.

Last reviewed on .

Article history

  • Latest version

Reference(s)

  1. Liu, Zhi-yun. “Integrative transcriptomic and single-nucleus analyses identify microbiota–metabolite-associated immune signatures of fibrosis severity in MASLD.” PLOS ONE, vol. 21, no. 11, pp. e0362860, doi: 10.1371/journal.pone.0362860. <https://doi.org/10.1371/journal.pone.0362860>.

Cite this page:

Sharma, Priya. “Eight Immune Genes Linked to Severe Liver Scaring in Metabolic Disease.” BioScience. BioScience ISSN 2521-5760, 16 November 2026. <https://www.bioscience.com.pk/en/subject/biology/integrative-transcriptomic-and-single-nucleus-analyses-identify-microbiota-metabolite-associated-immune-signatures-of-fibrosis-severity-in-masld>. Sharma, P. (2026, November 16). “Eight Immune Genes Linked to Severe Liver Scaring in Metabolic Disease.” BioScience. ISSN 2521-5760. Retrieved November 16, 2026 from https://www.bioscience.com.pk/en/subject/biology/integrative-transcriptomic-and-single-nucleus-analyses-identify-microbiota-metabolite-associated-immune-signatures-of-fibrosis-severity-in-masld Sharma, Priya. “Eight Immune Genes Linked to Severe Liver Scaring in Metabolic Disease.” BioScience. ISSN 2521-5760. https://www.bioscience.com.pk/en/subject/biology/integrative-transcriptomic-and-single-nucleus-analyses-identify-microbiota-metabolite-associated-immune-signatures-of-fibrosis-severity-in-masld (accessed November 16, 2026).
End of the article