Fecal Cathepsin B and S Are Associated with Liver Disease Severity and Adiposity in MASLD

Abstract

Background & Aims:

Dysregulation of hepatic cathepsins contributes to metabolic dysfunction-associated steatotic liver disease (MASLD) by promoting inflammation, apoptosis, and fibrosis. However, the role of intestinal cathepsins in MASLD has not been investigated. Given the central role of the gut-liver axis in disease progression, this represents an important knowledge gap.

Methods:

Fecal cathepsin B and S activity were measured in 95 patients with MASLD and 18 healthy controls. Cathepsin activity was correlated with liver disease severity, metabolic parameters, and gut microbiome composition.

Results:

Fecal cathepsin B and S activity were higher in patients with MASLD than in healthy controls. Cathepsin B activity was further increased in patients with metabolic dysfunction-associated steatohepatitis. Cathepsin B and S activity correlated with serum transaminases and hepatic steatosis, while cathepsin B activity was additionally associated with liver stiffness. Cathepsin B and S activity correlated with adiposity but showed no associations with other metabolic dysfunction-related parameters. Moreover, gut microbiome composition differed between patients with low versus high fecal cathepsin B or S activity, respectively.

Conclusion:

Increased fecal cathepsin B and S activity is associated with liver disease severity and adiposity in MASLD and is linked to alterations of the gut microbiome, suggesting a potential role of intestinal cathepsins in gut-liver axis dysfunction.

Lay summary

We found that the proteases cathepsin B and S are increased in stool samples of individuals with metabolic dysfunction-associated steatotic liver disease (MASLD) compared to healthy individuals. Higher levels were linked to more severe liver damage and adiposity and were associated with differences in gut bacteria. These findings suggest that fecal cathepsins may play a role in the development and progression of MASLD as part of a dysregulated gut-liver axis.

Introduction

Metabolic dysfunction-associated steatotic liver disease (MASLD) has emerged as the most prevalent chronic liver disease worldwide. MASLD and its progressive inflammatory phenotype, metabolic dysfunction-associated steatohepatitis (MASH), can advance to liver cirrhosis and hepatocellular carcinoma, resulting in substantial morbidity, impaired quality of life, and a growing burden on health-care systems.¹

Cathepsins are lysosomal proteases that play essential roles in intracellular protein turnover, autophagy, apoptosis, and extracellular matrix remodeling in the liver. Increasing evidence suggests that dysregulation of cathepsin expression and activity contributes to the pathogenesis and progression of MASLD.² Cathepsin B and S both belong to the cysteine proteases. Cathepsin B is one of the most abundant and ubiquitously expressed cathepsins, whereas cathepsin S is predominantly expressed in immune cells. Cathepsin B has been implicated in multiple mechanisms relevant to liver disease, including hepatocyte apoptosis, autophagy regulation, and activation of hepatic stellate cells, thereby promoting fibrosis.^3–5 Both cathepsin B and S have been further linked to inflammatory pathways that contribute to liver fibrosis.^3,6,7

Beyond intrahepatic mechanisms, a dysregulated gut-liver axis characterized by an impaired intestinal barrier function is a key driver of chronic liver disease.^8,9 In intestinal disorders such as inflammatory bowel disease, cathepsins, including cathepsin B and S, have been shown to promote intestinal inflammation and barrier dysfunction.^10–12 We recently demonstrated that fecal cathepsin B correlates with short-term mortality in patients with alcohol-associated hepatitis.¹³ Macrophage-derived intestinal cathepsin B proteolytically cleaves the tight junction protein occludin, which increases intestinal permeability and contributes to the progression of ethanol-induced liver disease in mice.

Building on these findings, the present study aims to investigate fecal cathepsin B and S activity in patients with MASLD.

Materials and Methods

Study population

A total of 95 patients with MASLD and 18 healthy controls (HC) were recruited between March 2015 and December 2018 at the outpatient liver department of the Clinic for Gastroenterology and Hepatology, University Hospital of Cologne, Germany.¹⁴ The study protocol was approved by the local Ethics Committee, and written informed consent was obtained from all participants. The study was performed in accordance with the Declaration of Helsinki.

MASLD was diagnosed based on the presence of hepatic steatosis in ultrasound and/or magnetic resonance imaging, and/or >5% hepatic fat content on histological assessment of liver biopsy. Additional criteria were daily alcohol consumption <10 g in women and <20 g in men, and the absence of steatogenic medications or other causes of secondary steatosis. Healthy controls met the following criteria: no known chronic diseases, body mass index <25 kg/m², daily alcohol intake <10 g in females and <20 g in males, normal abdominal ultrasound findings, and laboratory values within reference ranges. Exclusion criteria for all participants included oral or intravenous antibiotic therapy within the previous 6 months, known malignancy, pregnancy, and age <18 years. Type 2 diabetes was defined by an HbA1c ≥6.5%, a fasting glucose level ≥126 mg/dL, and/or the use of antidiabetic medications. Metabolic syndrome was diagnosed according to the criteria established by the International Diabetes Federation.¹⁵

In 77 MASLD patients, liver biopsy was performed, and liver disease was classified into MASLD or MASH. Detailed clinical data of the two patient subgroups are described in Supplementary Table S1. In patients with no liver biopsy, cirrhosis was determined if liver imaging was consistent with cirrhosis (such as nodular hepatic contour, changes in volume distribution indicating portal hypertension in the absence of portal vein thrombosis, secondary phenomena of portal hypertension such as splenomegaly, enlarged caudate lobe and left lobe lateral segment, regenerative nodules) together with clinical and laboratory signs of portal hypertension/cirrhosis (including low platelets, albumin and prothrombin time, esophageal varices). In 15 healthy controls and 92 patients, vibration-controlled transient elastography (FibroScan, Echosens, Paris, France) was performed to assess liver stiffness. Individuals were fasting before measurement. At least 10 valid measurements were performed, and the median value of these measurements was reported in kPa. Median (minimum–maximum) liver stiffness and controlled attenuation parameter (CAP) in healthy controls were 4.45 kPa (3.10–6.20) and 195 dB/m (134–238), respectively, indicating no evidence of subclinical fibrosis or steatosis.

Fecal Activity of Cathepsins

Enzymatic activity of cathepsin B and S in fecal samples was measured, respectively, using the Cathepsin B Activity Assay Kit (ab65300, Abcam) and Cathepsin S Activity Assay Kit (ab65307, Abcam), as previously described.¹³ In brief, 100 mg of feces were diluted in 300 μl of PBS, insoluble material was spun down at 10.000 x g for 1 minute, the supernatant was filtered (0.22 μm). Cathepsin activity was measured according to the manufacturer’s protocol and is reported as relative fluorescence units.

Ileal Cathepsin gene expression by real-time quantitative PCR

Biopsies from the ileum were used to assess intestinal gene expression of cathepsins. Biopsies were obtained from subjects during a clinically indicated endoscopy at the VA San Diego Healthcare System (n = 16, 94% male, median age 63 years (IQR 55,68)). Written informed consent was obtained from all subjects before the procedure. The study was approved by an institutional review board at the VA San Diego Healthcare System and at the University of California, San Diego. RNA was extracted from samples using Trizol (15596018, Invitrogen) following the manufacturer’s instructions. Total RNA (500 ng) was used for each reverse transcription reaction, and cDNA synthesis was performed using the cDNA reverse transcription kit (4368813, Applied Biosystems). Gene expression was determined by real-time quantitative PCR using iTaq Universal SYBR Green reagent (1725124, Bio-Rad) on a QuantStudio 5 Real-Time PCR Instrument (A34322, Applied Biosystems). Cycling protocol: initial denaturation (95 °C for 3 minutes (min)); 40 amplification cycles (95 °C for 5 seconds (s), 60 °C for 32 s); followed by a melt-curve (holding) stage (95 °C for 15 s, 60 °C for 1 min), and a final step (95 °C for 15 s). All reactions were performed in duplicate. Expression levels were normalized to GAPDH gene for each sample. Primers used in this study: GAPDH (FW: gtctcctctgacttcaacagcg; RV: accaccctgttgctgtagccaa), CTSB (FW: gcctgcaagcttcgatgcac; RV: ctattggagacgctgtagga), and CTSS (FW: aaacggctggtttgtgtgc; RV: cagtggtgatccagggtagg) (Integrated DNA Technologies).

Bacterial DNA extraction and 16S rRNA sequencing

DNA from human stool samples was extracted, and 16S rRNA sequencing was performed as published.¹⁶ Raw 16S sequence reads are available in the NCBI Sequence Read Archive (SRA) associated with Bioproject PRJNA540738.

Statistics and data analysis

Statistical analyses were performed using GraphPad Prism version 10.6.1 and R version 4.3.0. Continuous variables were assessed for normality and compared using non-parametric tests, as appropriate. For comparisons of two independent groups, the Mann-Whitney U test was applied. Comparisons involving three or more groups were performed using the Kruskal-Wallis test, followed by Dunn’s post hoc test for multiple comparisons. Categorical variables were compared using the χ² test or Fisher’s exact test, depending on expected cell counts. Correlations between continuous variables were assessed using Spearman’s rank correlation. Statistical significance was defined as p < 0.05, with levels indicated as follows: p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), and p < 0.0001 (****).

Beta diversity was assessed using Bray-Curtis dissimilarity, calculated from genus-level relative abundances. Principal coordinates analysis (PCoA) was performed with the phyloseq package in R, and group differences in microbiome composition were tested using permutational multivariate analysis of variance (PERMANOVA). To identify bivariate associations between individual genera and cathepsin B or S activity, we calculated Spearman rank correlations across all genera, and p-values were adjusted for multiple testing using the Benjamini-Hochberg false discovery rate (FDR). Multivariable genus-phenotype associations were further evaluated using MaAsLin2 (Microbiome Multivariable Association with Linear Models), applying a logit transformation to proportional abundance data and adjusting for clinical covariates. MaAsLin2 reported effect sizes (β), standard errors, and both nominal (p) and FDR-adjusted (q) values. Genera were considered nominally associated at p < 0.05 and FDR-significant at q < 0.05.

Results

Study Population

We analyzed fecal cathepsin B and S activity in 95 patients with MASLD and 18 healthy controls. Characteristics are described in Table 1. The patient population was older and had a higher body mass index (BMI) compared with healthy controls. Dyslipidemia was present in 5.5% of healthy controls compared with 57.9% of patients. Diabetes, hypertension, and metabolic syndrome were only present in patients, but not in any of the healthy controls. In 77 patients, liver biopsy was performed, and patients were sub-grouped according to histological classification into MASLD (n = 23) and MASH (n = 54, Table S1). While 17.9% of patients had liver cirrhosis, their liver function was preserved, as reflected by a low Model for End-Stage Liver Disease (MELD) score of 7.9 (6.4-10.7).

Table 1.

Clinical characteristics of the study population

Characteristic	Healthy controls n = 18†	MASLD n = 95†	p-value‡
Age	31.1 (29.1, 36.5)	53.9 (41.7, 63.6)	<0.001
Male	8 (44.4%)	45 (47.4%)	0.700
Diabetes	0 (0.0%)	22 (23.2%)	0.021
Hypertension	0 (0.0%)	61 (64.2%)	<0.001
Dyslipidemia	1 (5.5%)	55 (57.9%)	<0.001
Metabolic Syndrome	0 (0.0%)	40 (42.1%)	<0.001
Alcohol consumption [g/d]	2.0 (2.0, 9.3)	1.0 (0.0, 5.0)	0.003
MASH	-	54 (56.8%)	-
Liver Cirrhosis	-	17 (17.9%)	-
BMI [kg/m2]	20.9 (19.4, 23.7)	30.2 (27.4, 33.7)	<0.001
Waist circumference [cm]	82.0 (78.0, 85.0)	108.0 (97.0, 119.0)	<0.001
Hip circumference [cm]	96.5 (95.0, 102.0)	110.0 (104.0, 117.0)	<0.001
HbA1c [%]	-	5.5 (5.2, 6.0)	-
Triglycerides [mg/dl]	67.0 (59.0, 145.0)	141.0 (101.0, 225.0)	0.003
Cholesterol [mg/dl]	152.0 (137.0, 167.0)	195.0 (168.0, 224.0)	<0.001
LDL cholesterol [mg/dl]	70.5 (59.0, 83.0)	119.5 (94.0, 150.5)	<0.001
HDL cholesterol [mg/dl]	62.0 (56.0, 71.5)	49.0 (39.5, 58.5)	0.001
FibroScan [kPa]	4.5 (3.9, 5.4)	6.1 (4.7, 11.6)	<0.001
CAP [dB/m]	195.0 (182.0, 202.0)	288.5 (259.0, 318.0)	<0.001
FIB-4	0.8 (0.6, 1.1)	1.2 (0.7, 1.8)	0.022
AST [U/l]	24.0 (17.5, 25.5)	35.0 (28.0, 51.0)	<0.001
ALT [U/l]	14.0 (10.0, 23.0)	44.0 (33.0, 76.0)	<0.001
AP [U/l]	55.0 (47.8, 66.3)	73.0 (63.0, 94.0)	<0.001
GGT [U/l]	16.0 (11.0, 22.0)	72.0 (45.0, 124.0)	<0.001
Platelets [x109/l]	236.0 (209.0, 268.5)	234.0 (183.0, 281.0)	0.600
GFR [ml/min]	116.0 (102.0, 125.0)	93.0 (80.0, 102.0)	<0.001

^†Median (Q1, Q3); n (%)

^‡Wilcoxon rank sum test; Pearson’s Chi-squared test; Fisher’s exact test

ALT, alanine aminotransferase; AP, alkaline phosphatase; AST, aspartate aminotransferase; BMI, body mass index; CAP, Controlled Attenuation Parameter; FIB-4, Fibrosis-4 index; GFR, glomerular filtration rate; GGT, gamma-glutamyl transferase; HbA1c, glycated hemoglobin A1c; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MASLD, metabolic dysfunction-associated steatotic liver disease.

High Fecal Cathepsin B and S Activity in MASLD Patients is Associated with Liver Disease Severity

Fecal cathepsin B and S activity were highly correlated with each other (Figure S1A). Fecal cathepsin B and S activity were higher in patients with MASLD compared with healthy controls (CTSB: 384 vs 60 U, p < 0.001; CTSS: 240 vs. 66 U, p = 0.010, Figure 1A). Additionally, cathepsin B activity was higher in patients with MASH compared with MASLD (424 vs 332 U, p = 0.030), while cathepsin S activity was comparable between patients with MASH and MASLD (Figure 1B). Levels of fecal cathepsin B activity positively correlated with serum levels of ALT, AST, and GGT, and cathepsin S activity showed a weak correlation with ALT and AST levels (Figure 1C, D, S1B, C). Liver stiffness was measured using vibration-controlled transient elastography (FibroScan) as a non-invasive tool to assess liver fibrosis. Patients were stratified using the cutoff value of 8 kPa, based on algorithms to non-invasively rule out advanced fibrosis (F3-F4) in MASLD.¹⁷ Patients with higher liver stiffness exhibited increased cathepsin B activity compared to those with normal or mildly elevated liver stiffness (422 vs 281 U, p = 0.017, Figure 1E). Additionally, cathepsin B activity levels correlated positively with liver stiffness, and fecal cathepsin B and S activity correlated with liver steatosis assessed by controlled attenuation parameter (CAP) (Figure 1C, F, G). In a subgroup of patients with standardized histological liver assessment, cathepsin B activity showed a non-significant (p = 0.084) positive correlation with the NAFLD Activity Score (NAS), whereas cathepsin S did not correlate with NAS (Figure S1D). Cathepsin B and S activity were not significantly elevated in patients with cirrhosis (Figure S1E).

Figure 1. Fecal cathepsin activity and association with liver disease severity in MASLD.

(A) Fecal cathepsin B (CTSB) and S (CTSS) activity in healthy controls (HC) and patients with MASLD (Mann-Whitney U test, median with IQR).

(B) Fecal CTSB and CTSS activity in HC, MASLD, and MASH (Kruskal-Wallis test with Dunn’s post hoc test, median with IQR).

(C) Spearman correlation matrix of fecal CTSB and CTSS with liver disease-related parameters. Significant correlations (p < 0.05) are shown; circle size and colour indicate Spearman’s rho.

(D) Correlations of fecal CTSB and CTSS with ALT. Spearman’s rho and p values are indicated.

(E) Fecal CTSB and CTSS activity in individuals with liver stiffness <8 vs. ≥8 kPa (FibroScan; Mann-Whitney U test, median with IQR).

(F) Correlations of fecal CTSB and CTSS with liver stiffness (kPa; FibroScan); Spearman’s rho and p values are indicated.

(G) Correlations of fecal CTSB and CTSS with controlled attenuation parameter (CAP; dB/m; FibroScan); Spearman’s rho and p values are indicated.

Statistical significance was defined as p < 0.05, with levels indicated as follows: p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), and p < 0.0001 (****).

ALT, alanine aminotransferase; AST, aspartate aminotransferase; AP, alkaline phosphatase; CAP, controlled attenuation parameter; CTSB, cathepsin B; CTSS, cathepsin S; FS, FibroScan; GGT, gamma-glutamyl transferase; HC, healthy controls; IQR, interquartile range; MASLD, metabolic dysfunction-associated steatotic liver disease; MASH, metabolic dysfunction-associated steatohepatitis; MELD, Model of End-Stage Liver Disease.

Patients with MASLD and alcohol intake had higher fecal cathepsin S activity compared to patients without any alcohol intake, and a non-significant trend towards higher cathepsin B activity (Figure S1F). Use of proton pump inhibitors (PPI) was not associated with different levels of cathepsin B or S activity (Figure S1G), and PPI use did not confound the observed associations in multivariable linear regression models adjusted for age, BMI, HbA1c, ALT, AST, and liver stiffness (CTSB: p = 0.631; CTSS: p = 0.130). There was no difference in cathepsin B or S activity detectable between wild-type PNPLA3 variant and heterozygous or homozygous p.I148M variants (Figure S1H).

Taken together, fecal cathepsin B and S activity was associated with liver disease severity in patients with MASLD, with a more pronounced association observed for cathepsin B.

Fecal Cathepsin B and S Activity is Associated with Overweight

Correlation of fecal cathepsin B and S activity with clinical parameters associated with metabolic dysfunction revealed significant correlation with parameters for overweight, including body mass index (BMI), waist circumference as an indicator of visceral adiposity, and hip circumference as an indicator of subcutaneous adiposity (Figure 2A–D). Stratification of the overall study population, or of the MASLD population, by overweight (BMI ≥ 25), showed higher cathepsin B activity in overweight individuals (overall: 429 vs 77 U, p < 0.001; MASLD: 429 vs 190 U, p = 0.036, Figure 2E). Cathepsin S activity in the overall study population was higher in overweight individuals (419 vs 68 U, p = 0.011) and showed a non-significant trend to higher activity in overweight patients (419 vs 216 U, p = 0.36, Figure 2F). Further stratification by obesity (BMI ≥ 30) did not show a difference in fecal cathepsin activity between overweight and obese individuals (Figure S2A). In an independent cohort, intestinal expression levels of CTSB and CTSS were assessed in ileal biopsies, and overweight individuals showed a tendency toward higher expression levels of CTSB (Figure 2G).

Figure 2. Association of fecal cathepsin activity with adiposity.

(A) Spearman correlation matrix; significant correlations (p < 0.05) are shown. Circle size and colour indicate effect size (Spearman’s rho).

(B) Spearman correlations of fecal CTSB and CTSS activity with BMI (kg/m²); Spearman’s rho and p values are shown.

(C) Spearman correlations of fecal CTSB and CTSS activity with waist circumference (cm); Spearman’s rho and p values are shown.

(D) Spearman correlations of fecal CTSB and CTSS activity with hip circumference (cm); Spearman’s rho and p values are shown.

(E, F) Fecal CTSB (E) and CTSS (F) activity in the overall study population (left) and in patients with MASLD (right), stratified by BMI (<25 vs. ≥25 kg/m²; Mann-Whitney U test, median with IQR).

(G) Ileal CTSB and CTSS mRNA expression, normalized to GAPDH and multiplied by 100 (Mann-Whitney U test, median with IQR).

Statistical significance was defined as p < 0.05, with levels indicated as follows: p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), and p < 0.0001 (****).

BMI, body mass index; CTSB, cathepsin B; CTSS, cathepsin S; HbA1c, glycated hemoglobin A1c; HDL, high-density lipoprotein; IQR, interquartile range; LDL, low-density lipoprotein; MASLD, metabolic dysfunction-associated steatotic liver disease.

Interestingly, fecal cathepsin activity did not correlate with HbA1c, triglycerides, cholesterols, kidney function, or age (Figure 2A, S2B–E). Additionally, the presence of metabolic syndrome, dyslipidemia, diabetes, or hypertension in patients with MASLD was not associated with increased cathepsin B or S activity; rather, these conditions showed a trend toward lower cathepsin activity, both in the overall cohort and when restricted to overweight individuals (Figure S2F–I), despite their expected association with BMI (Figure 2A).

Overall, fecal cathepsin activity was specifically associated with adiposity, highlighting overweight as a key clinical correlate of fecal cathepsin activity. Again, the observed association was stronger for cathepsin B than for cathepsin S.

Altered Gut Microbiome Composition in MASLD is Associated with Fecal Cathepsin Activity

16S sequencing of the intestinal microbiota was performed.^14,16 A Bray-Curtis principal coordinates analysis (PCoA) on genus level, followed by PERMANOVA, demonstrated a difference in overall gut microbiome composition between individuals with high versus low fecal cathepsin B or S activity (cutoff defined by the median; CTSB: R² = 0.033, p = 0.001; CTSS: R² = 0.029, p = 0.004, Figure 3A). Cathepsin B activity showed no significant association with Chao1 richness, Shannon alpha diversity, or the inverse Simpson index, whereas cathepsin S activity exhibited a negative correlation with the Chao1 index (Figure S3A–C). At the genus level, correlation analyses revealed significant associations of 9 genera with cathepsin B and 10 genera with cathepsin S activity (adjusted p < 0.05, Figure 3B). Blautia, Erysipelatoclostridium, Absiella, and Eggerthella were positively associated with both fecal cathepsins. At the same time, Holdemanella, Butyricimonas, Oscillibacter, and Prevotella showed negative associations. In multivariable microbiome association analysis using MaAsLin2, adjusting for BMI and liver stiffness, several genera showed nominal associations with cathepsin B activity (p < 0.05), including Blautia, Holdemanella, Erysipelatoclostridium, and Bifidobacterium (Figure 3C). Among these, Blautia demonstrated the strongest positive association with cathepsin B activity (β = 0.34, p < 0.001), approaching significance after FDR correction (q = 0.085). However, no genera remained significant (q < 0.05) after multiple-testing correction. In parallel, the analysis of fecal cathepsin S activity revealed several genera showing only nominal association.

Figure 3. Genus-level gut microbiome profiles in relation to fecal cathepsin activity.

(A) Bray-Curtis PCoA at the genus level comparing individuals with high versus low fecal CTSB (left) or CTSS (right) activity (median split). Ellipses indicate 95% confidence intervals.

(B) Spearman correlation analysis identifying genera associated with fecal CTSB (left) or CTSS (right) activity (FDR-adjusted p < 0.05). Spearman’s rho is shown on the x-axis and coloured by direction of association.

(C) Volcano plots from MaAsLin2 differential abundance analyses for fecal CTSB (left) or CTSS (right) activity, adjusted for BMI (kg/m²) and liver stiffness (FibroScan, kPa). Effect sizes are shown on the x-axis and −log10(p) on the y-axis. The dashed line indicates p = 0.05; genera with q < 0.10 are highlighted in red.

CTSB, cathepsin B; CTSS, cathepsin S; FDR, False Discovery Rate; MaAsLin2, Microbiome Multivariable Association with Linear Models, Version 2; PCoA, principal coordinates analysis.

Taken together, these analyses indicate that fecal cathepsin B and S activity is associated with distinct shifts in gut microbiome composition.

Discussion

In this study, we demonstrated that fecal cathepsin B and S activity was associated with liver disease severity in a well-characterized patient cohort with MASLD. Particularly, cathepsin B activity showed a robust association with serum markers of liver injury, fibrosis, hepatic steatosis, and the presence of MASH, underscoring its potential relevance in disease progression.

MASLD is increasingly recognized as a multisystem disorder and is defined by hepatic steatosis in the presence of at least one additional cardiometabolic risk factor.¹⁷ Interestingly, fecal cathepsin activity correlated strongly with overweight, but not with other assessed metabolic risk factors. Taken together with the absence of higher fecal cathepsin activity in obese compared with overweight individuals, these findings suggest that intestinal cathepsin activation may be most relevant during early stages of adiposity-associated disease development rather than in advanced obesity or other metabolic disorders.

The role of cathepsin B and S in obesity has been previously investigated and linked to adipose tissue dysfunction. Cathepsin B is activated during adipocyte hypertrophy, where it mediates adipocyte cell death and promotes subsequent macrophage infiltration into adipose tissue.^18,19 In addition, cathepsin B contributes to dysregulated lipid metabolism by impairing autophagy and thereby exacerbating lipolytic dysfunction in obese adipose tissue in preclinical models.²⁰ Cathepsin S is highly expressed in the adipose tissue of obese individuals. It facilitates adipogenesis and adipose tissue expansion in vitro, and the inhibition of cathepsin S reduces adipocyte formation and inflammatory infiltration.²¹ Circulating cathepsin S levels are increased in obesity and decline with weight loss.²² In MASLD, cathepsins have been implicated in disease pathogenesis; their role appears to be multifaceted, involving the regulation of lipotoxicity and inflammatory responses, but evidence from human studies remains limited.² In preclinical models, cathepsin B regulates hepatic lipid metabolism by cleaving liver fatty acid-binding protein, thereby reducing free fatty acid uptake and very-low-density lipoprotein secretion in hepatocytes.²³ Moreover, excess free fatty acids induce lysosomal membrane permeabilization and cytosolic activation of cathepsin B, which precedes and mediates mitochondrial dysfunction, oxidative stress, and hepatocyte injury.²⁴ Consistently, deficiency or inhibition of cathepsin B in diet-induced steatohepatitis models attenuates hepatic lipid accumulation, inflammatory cell infiltration, and fibrosis.^25,26 Despite these insights, intestinal cathepsins have not been characterized before in the context of obesity or MASLD, nor have they been mechanistically linked to disease pathogenesis in these conditions.

This study has several limitations. First, the analyses are primarily correlative and therefore do not allow conclusions regarding causality. Second, the cohort size is modest, and no independent validation cohort was available, which limits the ability to confirm the robustness of the observed associations. These limitations should be considered when interpreting the findings.

Nevertheless, recent mechanistic work provides biological plausibility for the observed associations. We recently demonstrated that macrophage-derived intestinal cathepsin B proteolytically degrades the tight-junction protein occludin and thereby promotes gut barrier dysfunction and ethanol-induced liver disease in mice.¹³ A similar mechanism may contribute to the relationship between increased intestinal cathepsin activity and MASLD observed in the present study. Future work is required to directly test this hypothesis, including studies aimed at identifying the specific cellular sources and regulatory pathways driving intestinal cathepsin expression in MASLD.

Alterations in gut microbiota composition are a hallmark of MASLD and its progressive form, MASH.^27–29 However, the identification of bacterial genera that are consistently enriched or depleted across MASLD studies remains challenging, largely due to heterogeneity in study design, patient cohorts, and analytical approaches.^30–32 In our study, correlation analyses between relative microbial abundances and fecal cathepsin activity revealed, among others, positive associations with Blautia and Bifidobacterium, and negative associations with Prevotella. Notably, Blautia and Bifidobacterium are generally considered beneficial taxa because of their roles in supporting intestinal barrier integrity and metabolic homeostasis; however, reports on changes in the abundances of these taxa in MASLD are conflicting.^29,33–35 In contrast, Prevotella has been described as enriched in MASLD^33,36, with certain strains being associated with disease progression³⁷. Although these associations did not remain statistically significant after adjustment for BMI and fibrosis stage, nominal associations persisted. This suggests a link between microbiota, fecal cathepsin activity, and liver disease progression. Since intestinal cathepsins have been implicated in tight junction destabilization, immune response, and barrier dysfunction in alcohol-associated liver disease¹³ and exert direct antimicrobial actions^38,39, fecal cathepsins could induce changes in microbiota composition and modulate intestinal barrier function, with a consequent impact on liver disease progression. Future mechanistic studies will be required to delineate causality and to assess the underlying pathophysiological pathways of fecal cathepsins in MASLD. From a clinical perspective, the integration of fecal cathepsin activity with microbiome profiling may provide a non-invasive approach to capture gut–liver axis alterations and improve risk stratification in MASLD, although prospective validation in independent cohorts is required.

Collectively, our findings show that increased fecal cathepsin B and S activity is associated with disease severity and adiposity in MASLD and is linked to gut microbiome alterations. With this, we are extending the current knowledge of the dysregulated gut-liver axis in liver disease and reinforcing emerging evidence that intestinal cathepsins represent a potential therapeutic target to restore intestinal barrier integrity in liver disease, with possible broader implications for the treatment of obesity.

Supplementary Material

The supplementary material includes additional figures (Figures S1–S3) and a supplementary table (Table S1) detailing extended analyses and additional clinical data.

Key points/ Highlights.

• Fecal cathepsin B and S activity are increased in patients with MASLD and correlate with hepatic injury and steatosis.

• Cathepsin B activity is further elevated in patients with steatohepatitis and correlates with liver stiffness.

• Fecal cathepsin activity is associated with adiposity, but not with other metabolic parameters.

• Distinct gut microbiome profiles are associated with high versus low fecal cathepsin activity.

• Intestinal cathepsins may contribute to gut–liver axis dysfunction in MASLD.

Abbreviations

ALT

alanine aminotransferase

AST

aspartate aminotransferase

AP

alkaline phosphatase

BMI

body mass index

CTSB

cathepsin B

CTSS

cathepsin S

FDR

False Discovery Rate

GFR

glomerular filtration rate

GGT

gamma-glutamyl transferase

CAP

controlled attenuation parameter

HbA1c

glycated hemoglobin A1c, HC, healthy controls

IQR

interquartile range

MaAsLin2

Microbiome Multivariable Association with Linear Models, Version 2

MASLD

metabolic dysfunction-associated steatotic liver disease

MASH

metabolic dysfunction-associated steatohepatitis

MELD

Model for End-Stage Liver Disease

NAS

Non-alcoholic fatty liver disease (NAFLD) activity score

PCoA

principal coordinates analysis

PERMANOVA

permutational multivariate analysis of variance. PNPLA3, patatin-like phospholipase domain-containing protein 3

PPI

proton pump inhibitors

Data availability statement

The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request. Access to individual-level patient data is restricted due to ethical and data protection regulations. The 16S rRNA gene sequencing data generated in this study are available in the NCBI Sequence Read Archive (SRA) under BioProject accession number PRJNA540738.