Dual effects of a high-protein diet on metabolic dysfunction-associated steatotic liver disease: Basic research on precision clinical interventions

Abstract

Metabolic dysfunction-associated steatotic liver disease (MASLD) is a global health challenge wherein high-protein diets (HPDs) act as a double-edged sword. Building upon the recent systematic review, this review critically examines the dual effects of HPDs on MASLD. Hepatic outcomes depend critically on protein source, intake level, and host metabolic phenotype. Plant-based proteins confer hepatoprotective benefits via gut–liver axis modulation and favorable amino acid signaling, whereas excessive animal protein intake may exacerbate steatosis through mTORC1 activation. However, critical gaps persist: The bioactive peptides in plant proteins remain functionally uncharacterized; evidence-based HPD intake thresholds stratified by disease severity and comorbidities are lacking; and multi-omics findings have yet to translate into clinical practice. Future research must prioritize functional validation of plant-derived bioactives, development of artificial intelligence-driven personalized nutritional algorithms, and long-term randomized controlled trials assessing hard clinical endpoints. Addressing these imperatives will accelerate the transition from generic dietary advice to precision nutritional strategies for MASLD management.

Key Words: High-protein diet; Metabolic dysfunction-associated steatotic liver disease; Precision nutrition; Intestinal microbiota; Plant-based protein

Core Tip: A high-protein diet has dual effects on metabolic dysfunction-associated steatotic liver disease (MASLD), with protein source, intake level, and host metabolic phenotype serving as key regulators. Although previous studies have established a framework involving the gut–liver axis and amino acid pathways, gaps persist in causal mechanisms, intake thresholds, and clinical translation of multiomic findings. Future work should prioritize validating plant-based protein bioactivities, developing personalized algorithms that integrate host–microbiota interactions, and conducting long-term randomized controlled trials for precise MASLD dietary guidance.

INTRODUCTION

Metabolic dysfunction-associated steatotic liver disease (MASLD) has rapidly ascended to become the most prevalent chronic liver disease worldwide, with an estimated global prevalence exceeding 38%, and is becoming a major indication for liver transplantation, thereby imposing a substantial and growing clinical and economic burden[1-4]. Unlike other chronic liver diseases, MASLD is closely linked to metabolic syndrome, obesity, and type 2 diabetes (T2D), making nutritional intervention a cornerstone of its management[5-7]. High-protein diet (HPD) has gained attention for its potential to improve hepatic steatosis, insulin resistance, and body composition[8-10], as evidenced by recent randomized controlled trials demonstrating that HPD significantly reduces intrahepatic lipid content and improves insulin sensitivity[11]. However, emerging evidence suggests that HPD may exert adverse effects in susceptible populations, such as exacerbating hepatic lipogenesis or impairing renal function[12,13], leading to intense debate in the field.

HPD IN MASLD: FUTURE RESEARCH ON MECHANISMS, STRATIFICATION, AND SAFETY

A recent systematic review by Yin et al[14] systematically demonstrates the “dual nature” of HPD in MASLD management, highlighting protein source and intake level as key determinants of liver-related outcomes. At the molecular level, it elucidates the biological basis for the differential effects of distinct protein sources. Furthermore, by integrating evidence from animal models and clinical studies, this review presents the contrasting effects of various protein sources on hepatic lipid metabolism, inflammatory markers, and gut microbiota composition. This work provides a valuable foundation for both mechanistic understanding and clinical application. Nevertheless, in light of current research frontiers and future clinical needs, several critical aspects merit further exploration, as detailed below.

First, the mechanisms underlying the beneficial effects of plant-based proteins remain incompletely understood. Although this review emphasizes the roles of gut microbiota remodeling and short-chain fatty acid production, the specific bioactive peptides (e.g., from legume-derived protein hydrolysates) that mediate lipid-lowering and anti-inflammatory effects require further purification and functional validation[15,16]. Lupine protein hydrolysates were identified as potential candidates in a recent study by Santos-Sánchez et al[17]. However, their molecular targets in hepatic lipid metabolism - such as peroxisome proliferator-activated receptor α or AMP-activated protein kinase (AMPK) - remain unclear. As summarized in Table 1, the beneficial effects of plant-based proteins on MASLD may be attributed not only to the protein itself but also to coexisting bioactive compounds (e.g., polyphenols, saponins, and fiber) that modulate the gut microbiota, lipid metabolism, and inflammatory pathways.

Table 1 Summary of plant protein sources, study characteristics, bioactive compounds, and metabolic dysfunction-associated steatotic liver disease-related outcomes.

Plant protein source	Study type	Intervention period	Key bioactive compounds	MASLD-related outcomes
Soy protein	Human RCT	8-12 weeks	Isoflavones, saponins	Hepatic fat↓, ALT↓, LDL-cholesterol↓[60]
Lupine protein	Animal model	12 weeks	Lupine peptides, alkaloids, fiber	Abdominal adiposity↓, Hepatic steatosis↓, Inflammation↓[17]
Pea protein	Human crossover trial	4 weeks	Legume peptides, saponins, resistant starch	Postprandial glucose↓, Satiety↑, Total cholesterol↓[61]
Chickpea protein	Animal model	8 weeks	Bioactive peptides, polyphenols, fiber	Hepatic triglyceride accumulation↓, Antioxidant enzymes↑[62]
Mixed plant proteins	Human RCT	16 weeks	Various isoflavones, peptides, fiber	Liver stiffness↓, Inflammatory markers↓[63]

MASLD: Metabolic dysfunction-associated steatotic liver disease; ALT: Alanine aminotransferase; LDL: Low-density lipoprotein; RCT: Randomized controlled trial.

Additionally, the potential adverse effects of HPD on MASLD must also be considered. Excessive intake of animal-based protein has been associated with activation of the mTORC1 pathway, which may promote hepatic de novo lipogenesis and exacerbate steatosis in susceptible individuals[18]. Some studies have also reported increased circulating levels of branched-chain amino acids (BCAAs) following HPD, which are paradoxically associated with insulin resistance and worse metabolic outcomes in individuals with obesity[19,20]. These observations underscore the importance of personalized approaches that consider both the protein source and the patient’s metabolic and renal status.

Second, no consensus guidelines currently exist to define optimal protein intake specifically for MASLD patients. While general population guidelines (e.g., ESPEN recommends 1.2-1.5 g/kg/day for older adults) provide a reference point, MASLD-specific HPD thresholds remain to be established. In this review, the intake threshold for HPD ranges from 1.2 g/kg/day to 2.0 g/kg/day. However, evidence suggests that metabolic phenotypes - such as obesity, T2D, and chronic kidney disease - influence an individual’s tolerance for protein intake[12]. For example, individuals with impaired glomerular filtration rates may be more susceptible to HPD-induced hepatic steatosis[13]. Furthermore, the presence of cardiometabolic risk factors - including hypertension, dyslipidemia, and T2D - may significantly modulate individual responses to HPD[21]. For instance, insulin resistance can alter amino acid metabolism and protein utilization efficiency[22]. These observations underscore the necessity of stratifying patients according to comorbidities in future studies. This research gap should be addressed through dose-escalation randomized controlled trials (RCTs) in which patients are stratified by MASLD severity (ranging from simple steatosis to liver fibrosis) and by comorbidities, to establish safe and effective protein intake ranges tailored to distinct patient subgroups. Such trials should also incorporate liver-specific imaging (e.g., proton magnetic resonance spectroscopy) and histopathological endpoints to complement serological markers of steatosis and inflammation[23].

Third, although this review highlights the significance of personalized nutrition, translating multiomics data (including genomics, proteomics, and microbiomics) into clinical practice remains in its early stages. Current evidence relies predominantly on observational studies and cross-sectional microbiome analyses[24], whereas longitudinal studies integrating multiomics profiling with dietary interventions are needed to identify predictive biomarkers of responses to an HPD. For example, host genetic variants (e.g., in PNPLA3 and TM6SF2) that influence amino acid metabolism or gut microbial features (e.g., an abundance of Akkermansia muciniphila) could serve as stratification tools for tailoring protein intake[25-27]. Furthermore, artificial intelligence-driven algorithms that integrate such data with clinical parameters - such as visceral adiposity and insulin resistance - hold promise for generating actionable dietary recommendations[28]. However, it must be emphasized that these AI models require rigorous external validation in diverse populations before clinical implementation to avoid overfitting and ensure generalizability[29,30]. Recent studies have highlighted the immaturity of certain AI models in the MASLD field, with many lacking prospective validation[31], highlighting the need for caution when interpreting model predictions[32,33]. Although a recent study by Kan et al[34] demonstrated the feasibility of this approach in overweight Chinese adults, it has not yet been validated in patients with MASLD.

Fourth, the long-term safety and efficacy of HPD in MASLD patients with advanced liver fibrosis or cirrhosis remain insufficiently investigated[35]. While this review notes that excessive intake of animal-based protein may contribute to renal impairment, data on these outcomes in patients with decompensated cirrhosis or hepatocellular carcinoma are notably scarce[36]. A prospective cohort study by Daftari et al[37]. revealed an association between lower animal-based protein intake and reduced mortality in individuals with cirrhosis. However, similar data for MASLD-related cirrhosis are still lacking. Future randomized controlled trials should incorporate clinical outcomes - such as the progression to MASLD-related steatohepatitis or liver-related mortality - to clarify the risk-benefit ratio of HPD in patients with MASLD[38,39].

Finally, greater attention should be given to interactions between HPD and other dietary components, such as dietary fiber and polyunsaturated fatty acids, as well as lifestyle factors, including physical exercise and circadian rhythms[40,41]. For instance, the co-ingestion of plant-based proteins with dietary fiber may modulate amino acid bioavailability and the gut microbiota composition[42,43]. The synergistic effects of HPD and physical exercise on hepatic metabolism are likely mediated through complementary mechanisms[44-46]. Resistance training enhances muscle protein synthesis and amino acid utilization, potentially improving nitrogen balance and reducing the amino acid load delivered to the liver[47,48]. Aerobic exercise activates AMPK in skeletal muscle and liver, which synergizes with HPD-induced AMPK activation to increase fatty acid oxidation and inhibit de novo lipogenesis[49]. Additionally, exercise-induced improvements in insulin sensitivity may enhance the metabolic utilization of dietary amino acids, shifting them toward muscle protein synthesis rather than hepatic gluconeogenesis or lipogenesis[50-52]. Time-restricted feeding combined with HPD may further amplify these benefits by aligning nutrient intake with circadian rhythms of metabolic regulation[23,53,54], but the efficacy of such multimodal interventions for MASLD has not been systematically evaluated. This mechanism may involve restructuring of the gut microbiota following alterations in host lifestyle and dietary habits, which subsequently induce a series of metabolic reprogramming events in the liver, thereby alleviating disease progression[55-58] (Figure 1).

Figure 1 Schematic diagram illustrating the mechanisms by which a high plantbased protein diet and moderate physical exercise alleviate metabolic dysfunction-associated steatotic liver disease. A high plantbased protein diet combined with moderate physical exercise can remodel the gut microbiota, increasing beneficial genera such as Akkermansia muciniphila (enhances gut barrier function), Faecalibacterium prausnitzii (produces butyrate and has anti-inflammatory effects), and Bifidobacterium (produces short-chain fatty acids), while reducing pathobionts such as Escherichia coli and, in certain contexts, Prevotella copri associated with branched-chain amino acid production and insulin resistance. The resulting microbial alterations subsequently drive hepatic metabolic reprogramming via the gut–liver axis, encompassing changes in lipid, glucose, and amino acid metabolism. These changes enhance fatty acid oxidation, reduce fatty acid production, and increase glycolysis, thereby reversing hepatic steatosis. FA: Fatty acids; FAO: Fatty acid oxidation. The two liver schematic diagrams in Figure 1 were created with BioGDP.com[59].

CONCLUSION

The systematic review by Yin et al[14] provides an essential and timely framework for understanding the complex interplay between HPDs and MASLD. In this Opinion Review, we have extended that foundation by critically synthesizing the underlying molecular mechanisms, delineating unresolved controversies, and outlining a forward-looking research agenda. The available evidence unequivocally demonstrates that the hepatic effects of HPDs are not uniform; rather, they are contingent upon a triad of key determinants: Protein source, intake level, and host metabolic phenotype. Plant-based proteins, particularly when accompanied by dietary fiber and bioactive phytochemicals, confer hepatoprotective benefits through favorable modulation of the gut-liver axis and activation of AMPK signaling. Conversely, excessive intake of animal-based proteins may exacerbate steatosis and insulin resistance via mTORC1 activation and BCAA dysmetabolism. Despite these mechanistic insights, critical translational barriers persist. The specific bioactive peptides within plant proteins remain functionally uncharacterized, evidence-based intake thresholds stratified by disease severity and cardiometabolic comorbidities are lacking, and the integration of multi-omics data into clinically actionable algorithms remains in its infancy. Moving forward, the field must prioritize the functional validation of plant-derived bioactives, the design of stratified dose-escalation randomized controlled trials, and the rigorous development and external validation of artificial intelligence-driven precision nutrition tools. Addressing these imperatives will enable a transition from generic dietary recommendations to personalized, evidence-based nutritional strategies capable of optimizing liver-related outcomes for the growing global population affected by MASLD.