A multi-functional oral small molecule targeting energy and lipid metabolism to treat obesity and related metabolic disorders

TOFA is a unique orally bioavailable ACC1/2 inhibitor

Previous studies have defined TOFA as a low micromolar inhibitor of fatty acid and cholesterol synthesis in hepatocytes at the level of ACC1 and ACC2 (31). Given TOFA’s unique structural properties that share similarities with fatty acids, TOFA has been shown to similarly be converted intracellularly into its thioester form, TOFyl-CoA, which represents a more potent ACC inhibitor than the parent TOFA compound (24, 32, 33) (Fig. 1A). As expected, our biochemical profiling of the TOFyl-CoA thioester form demonstrated markedly stronger inhibition of both ACC1 and ACC2 as compared to the free form TOFA (Fig. 1, B and C). TOFA treatment of HepG2 hepatoma cells, compared to other characterized ACC inhibitors, such as Firsocostat (19), PF-05175157 (20), and CP-640186 (34), uniquely resulted in higher oxygen consumption rates (Fig. 1D) via increased proton leak (Fig. 1E) and elevated spare mitochondrial capacity for ATP production during metabolic stress or demand (Fig. 1, F and G and fig. S1, A to C), suggesting TOFA’s distinct mechanism beyond just ACC inhibition in enhancing cellular energy expenditure. Moreover, among lipogenic enzyme modulators more broadly, TOFA exhibited a distinctly elevated mitochondrial capacity profile, especially with proton leak and spare respiratory capacity, further highlighting the distinct mechanism of TOFA as a lipid and energy metabolism modulating agent (fig. S1, D and E). To assess in vivo tolerability of TOFA, clearance of TOFA in C57BL/6 J mice following oral administration showed longer half-life clearance following oral administration (PO) compared to intraperitoneal administration (IP) (Fig. 1H), providing justification for subsequent oral administration of TOFA. In healthy, chow-fed C57BL/6 J mice orally administered with TOFA for three weeks, no significant change in weight was observed compared to vehicle treated mice (Fig. 1I). Interestingly, TOFA treatment reduced liver triglyceride levels (Fig. 1J), without an increase in serum triglyceride levels, a noted adverse class side effect of ACC inhibition (Fig. 1K). To further understand the underlying hepatic effects of TOFA in vivo, primary hepatocytes were isolated from C57BL/6 J mice treated once daily by oral gavage for one week with either vehicle or TOFA, and subsequently evaluated by mitochondrial stress testing. Hepatocytes from TOFA-treated mice exhibited higher oxygen consumption rates throughout the mitochondrial stress test, indicating higher mitochondrial respiratory function in response to TOFA treatment in mice, consistent with our in vitro findings (Fig. 1, L to P). Together, our data demonstrate a differentiated mechanism of action and outcome of TOFA compared to similar classes of lipid modulating compounds and supports TOFA’s bioavailability, efficacy, and tolerability in mice.

TOFA treatment counters metabolic derangements in a high-fat diet-induced mouse model of obesity and metabolic syndrome

Given the effects of TOFA on mitochondrial energy expenditure in hepatic cells, we wished to assess whether TOFA might also boost energy expenditure in vivo and beneficially affect obesity and related metabolic disorders. Male C57BL/6 J mice were fed a 60 kcal% fat diet (HFD) for seven weeks to induce rapid weight gain and diet-induced obesity (DIO), then orally treated twice-daily for four weeks with vehicle or TOFA. Body weight reduction was immediately observed after TOFA introduction and sustained after switching to a lower maintenance dose (to prevent rapid weight loss exceeding defined humane endpoints), equating to an average 18% decrease in body weight at the end of the treatment course (Fig. 2A). The effect on weight change was dose-dependent, in which the initiation dose conferred the strongest effects on body weight change (fig. S2A). Treatment caused no changes in food intake (Fig. 2B), and weight loss was primarily attributed to a significant fat mass reduction with no significant changes in lean mass as assessed by EchoMRI (Fig. 2C). Weight loss is known to improve type 2 diabetes–related phenotypes such as insulin resistance and glucose intolerance in obese mammals. TOFA treatment significantly lowered baseline fasting blood glucose and improved total glucose tolerance observed by earlier blood glucose level recovery response initiation starting at thirty minutes after glucose injection (Fig. 2, D and E), a concerted effect of both weight loss and prolonged insulin sensitivity (Fig. 2, F and G and fig. S3, A and B). TOFA treatment also reduced fasting insulin levels (Fig. 2H), an anti-diabetic effect. Obesity is also strongly associated with fatty liver diseases (e.g., MASLD). Strikingly, TOFA treatment significantly reduced liver triglycerides (Fig. 2I) and visibly improved liver morphology with less lipid accumulation (Fig. 2J), indicating protective effects of TOFA against hepatic steatosis. These results align with the dose-dependent efficacy of TOFA in modulating liver lipid levels (fig. S2B). We also observed a significant decrease in various circulating lipid species, including serum triglycerides (Fig. 2K) and cholesterol, in which the reduction was more pronounced in the VLDL/LDL cholesterol fraction over the HDL cholesterol fraction (Fig. 2L), consistent with beneficial remodeling of the blood lipid profile. Changes in blood lipid levels were also observed in a dose-dependent manner with TOFA treatment (fig. S2, C and D). These results present TOFA as an ACC1/2 inhibitor with a unique mechanism of action that addresses broader metabolic risk factors and complications associated with diet-induced obesity.

TOFA treatment confers metabolic and energy homeostasis benefits against diet-induced obesity

To gain a better mechanistic understanding of how TOFA confers its metabolic benefits beyond lipid regulation, energy expenditure was assessed in vivo. In order to investigate the dynamic period of weight change (as opposed to assessing treatment end outcomes), C57BL/6 J mice were fed a 60% HFD for twelve weeks followed by a short one-week course of vehicle or TOFA once daily oral treatment. At the one-week timepoint, TOFA-treated mice started to show significant body weight change compared to vehicle treated mice, with a specific reduction in fat mass and no significant changes in lean mass (Fig. 3A). Gross fecal energy content revealed no caloric differences between control and TOFA-treated mice, indicating no signs of malabsorption and the energy phenotype driven by internal cellular metabolism rather than gastrointestinal losses (fig. S4A). Mice were subsequently placed in metabolic cages. Interestingly, TOFA treated mice exhibited increased energy expenditure at all housing temperature conditions, ranging from room temperature (23°C), thermoneutral (30°C), and less pronounced in cold exposure (4°C) (Fig. 3B and fig. S4B), suggesting broad stimulation of energy expenditure beyond low temperature-dependent thermogenesis mediated by brown and beige adipose tissues. To account for differences in body composition proportions potentially contributing to metabolic energy expenditure, ANCOVA adjustment of energy expenditure (35) was carried out, which further confirmed our observation that TOFA treatment increases energy expenditure through non-thermogenic mechanisms, with significant contribution by lean mass (fig. S4, C and E and table S1). Indeed, we observed ∼18% increase in energy expenditure in TOFA-treated mice compared to control under both room temperature and thermoneutral conditions, in which thermogenic demands are minimal, whereas energy expenditure increase was minimal under cold exposure (4% increase in TOFA-treated mice), supporting the notion of genuine metabolic activation by TOFA rather than enhanced thermogenesis mechanisms (Fig. 3C). This increase in overall energy expenditure was consistent with the subcomponent increases in oxygen consumption (fig. S4, F and G) and carbon dioxide production (fig. S4, H and I), which reveal fat oxidation as the predominant energy source (fig. S4, J and K). Furthermore, the increase in energy expenditure is not attributed to significant increases in overall activity levels (fig. S4, L and M), suggesting whole body metabolic thermogenesis. Measurement of body temperature revealed no evidence of hyperthermia, an adverse effect of concern for energy metabolism modulating agents (Fig. 3D). The modest rise in energy expenditure in response to TOFA treatment alleviates concerns related to increased cardiac workload demand, as previous studies support the tolerability of 10–20% increases in energy expenditure (36, 37). The boost in energy expenditure by TOFA in vivo also aligns with the increased mitochondrial oxygen consumption rate observed from TOFA treatment in both human hepatocellular cell lines (Fig. 1D) and primary murine hepatocytes (Fig. 1L). In agreement with these findings, gene expression analysis with RT-qPCR of livers from TOFA-treated animals revealed a marked upregulation of genes involved in mitochondrial fusion, fission, mitophagy, and other mitochondria-associated bioenergetic processes, suggesting broad beneficial mitochondrial dynamic remodeling in response to TOFA treatment (Fig. 3E). Given the critical role of mitochondria in energy production and oxidative stress regulation, the impact of TOFA on mitochondrial dynamics for maintaining bioenergetic function, restoring cellular health, and adapting to metabolic needs is highly significant. Another aspect of metabolic benefits conferred by TOFA is the improvements in insulin resistance, an obesity-related hallmark of metabolic syndrome and pre-diabetes. Before tissue collection, mice were injected with PBS or insulin (1 U/kg) to assess insulin signaling via the pAKT to AKT ratio. As expected, control DIO mice were insulin resistant in multiple tissues, showing no change in pAKT levels between PBS and insulin treatment (Fig. 3, F to I). However, TOFA treatment robustly restored liver insulin sensitivity, as evidenced by a significant increase in the pAKT to AKT ratio after insulin injection (Fig. 3F). Given the remarkable beneficial metabolic effects in obese mice, we speculated that TOFA might also confer extrahepatic metabolic and energetic benefits. Of note, in an endurance exercise treadmill test, no decrease in endurance muscle function was observed (fig. S5A), consistent with our observation of preservation of lean mass with TOFA treatment. Interestingly, examination of gastrocnemius muscle, a mixed fiber type muscle, revealed upregulation of genes involved in the uncoupling response and mitochondrial biogenesis, suggesting peripheral, extrahepatic bioenergetic effects of TOFA (fig. S5B). Another potential contributor to elevated energy expenditure with TOFA is stimulation of brown adipose tissue (BAT) or the beigeing (or browning) of subcutaneous white adipose depots. However, canonical genes involved in BAT function and WAT browning (e.g., Pgc1a, Prdm16, Ucp1) showed only marginal and inconsistent increases, suggesting these processes are unlikely to explain the robust effect of TOFA on whole-body energy expenditure (fig. S5, C and D). Moreover, the stronger elevation in energy expenditure at room temperature and thermoneutral conditions as compared with cold exposure (Fig. 3C) further support a broad increase of systemic metabolism beyond classical brown or beige adipose thermogenesis. Taken together, these findings highlight the role of TOFA in modulating energy and metabolic homeostasis in various tissue depots, likely contributing to its potent anti-obesity effect.

TOFA treatment has a distinct transcriptional profile revealing upregulation of the PPAR transcriptional network

While the effects of many ACC inhibitors have been well characterized (19, 20, 22), the distinct treatment outcomes of TOFA, especially regarding serum triglycerides and its anti-obesity effects, remain unexplained. To better understand the effects of TOFA on whole genome expression, RNA-seq analysis was performed on liver tissues from DIO mice treated with vehicle or TOFA. Differentially regulated genes (DEGs) were identified, revealing 1,184 DEGs, of which 758 genes were upregulated and 426 were downregulated (fig. S6A). Intriguingly, Enrichr pathway enrichment (38) analysis of upregulated genes against the TRRUST 2019 database of transcriptional regulatory networks revealed the enrichment of the PPAR nuclear hormone receptor superfamily as being among the top enriched transcription factors in databases for mouse (Fig. 4A) as well as human (fig. S6B) in response to TOFA treatment. Amongst the PPAR family, Ppara mediates the fasting response in liver and plays a key role in regulating genes involved in lipid and energy homeostasis, strongly promoting mitochondrial fatty acid beta-oxidation (39–41). Notably, the RNA-seq analysis also revealed downregulation of the Srebp1 transcriptional network, which controls fatty acid and lipid biosynthesis (Fig. 4A), consistent with the lack of hypertriglyceridemia [a metabolic side effect after treatment with other ACC inhibitors as discussed above (22)] upon TOFA treatment. RT-qPCR analysis of liver samples showed upregulation of multiple PPAR network genes associated with the hallmark Ppara-mediated fasting response (Fig. 4B). Intriguingly, FGF21, a hepatokine regulated by PPARα with relevant pharmaceutical effects in mediating weight loss and pleiotropic metabolic effects of PPARα (42), was shown to be significantly upregulated in both mRNA transcripts as well as circulating protein in serum (Fig. 4, C and D). Further RNA-seq analysis of gastrocnemius muscle tissue also revealed strong upregulation of the PPAR signaling pathway by gene set enrichment analysis (GSEA) (43) in this metabolic tissue, as well as broad downregulation of inflammatory pathways (fig. S6, C to F). Taken together, these findings suggest that TOFA not only acts through the previously characterized ACC inhibition, but also stimulates the PPAR network, thereby conferring beneficial therapeutic effects promoting an overall favorable anti-obesity and anti-inflammatory profile.

Activation of PPARα and PPARδ is critical for the full beneficial effects of TOFA treatment

PPARs (α, δ, γ) are nuclear receptors for long-chain fatty acids involved in transcriptional control of metabolism, energy balance, and adipose function, with each subtype differing in expression patterns, ligands, and target genes (39, 41, 44). Given the structural similarity of TOFA to long-chain fatty acids, which are known ligands of PPARs (41), we examined the relationship between TOFA and PPAR family members to further characterize the specificity and mechanism underlying TOFA’s treatment outcomes. PPARα is a known therapeutic target (e.g., via fibrates) to modulate lipid homeostasis in mice and humans by decreasing serum triglyceride levels and increasing fatty acid clearance by beta-oxidation, primarily through effects in the liver (45). To evaluate the possible role of PPARα in mediating metabolic effects of TOFA, whole body Ppara knockout (KO) mice were fed a 60% HFD for ten weeks followed by a two-week oral treatment of vehicle or TOFA daily. TOFA reduced body weight in both wild-type and Ppara KO mice (Fig. 5A). However, in the wild-type mouse, TOFA treatment decreased serum triglycerides but this decrease was blunted in TOFA-treated Ppara KO mice compared to control treatment (Fig. 5B). Similarly, serum cholesterol levels presented the same pattern in which the Ppara KO condition diminished the TOFA-dependent decrease in serum cholesterol (Fig. 5C). RNA-seq analysis of liver tissue from Ppara KO mice revealed the PPAR signaling pathway as a persistently upregulated pathway in gene set enrichment analysis (Fig. 5D), as well as interferon-related responses as top downregulated gene sets (Fig. 5E). Amongst other gene sets, we observed a consistent increase in fatty acid metabolism hallmark gene signature (fig. S7A) and a decrease in inflammatory response hallmark gene signatures (fig. S7B). The upregulated PPAR signaling pathway (fig. S7C) in Ppara KO mice treated with TOFA pointed to a redundant or compensatory PPAR family member at play in the liver, such as Ppard (fig. S7D). Subsequent analysis by RT-qPCR revealed genes, such as Angptl4, Fabp4, and Cyp4a14, that share multiple PPAR subtype transcriptional regulatory elements were persistently upregulated by TOFA treatment in Ppara KO mice (fig. S7E). However, several PPAR target genes, such as Ucp2, Pdk4, and Cd36, previously upregulated by TOFA treatment in wild-type DIO mice were no longer upregulated in Ppara KO mice (fig. S7F). The attenuated yet persistent upregulation of specific PPAR target genes in Ppara KO mice suggests functional redundancy among PPAR subtypes in the liver, primarily PPARα and PPARδ due to their prominent hepatic expression (39, 46). Furthermore, analysis of gastrocnemius muscle, with strong PPARδ expression (47), showed upregulation of PPARδ target genes, such as Pdk4 and Cpt1b, by TOFA treatment (fig. S7, G and H), suggesting a role for PPARδ agonist activity inducing peripheral energy metabolism and having a positive effect on mitochondrial respiratory function (48–51). To confirm this hypothesis, we utilized a cell-based protein-protein interaction assay to assess receptor activation and subsequent co-activator recruitment. TOFA showed strong agonist activity against human PPARα (EC₅₀ = 2.095 μM) and human PPARδ (EC₅₀ = 0.594 μM), with very weak to no activity on human PPARγ (EC₅₀ > 10 μM) (Fig. 5, F to H). Furthermore, activation of PPARα and PPARδ was shown to maximize at 74% and 54%, respectively, relative to full agonist controls, implying partial PPARα/δ agonist activity of TOFA (Fig. 5, F to H). To further investigate the mechanistic relationship between TOFA and PPARα and PPARδ, biochemical thermal shift assays with purified ligand-binding domains revealed that TOFA directly interacted with both PPARα and PPARδ proteins in a concentration-dependent manner. TOFA binding stabilized the PPARα LBD, producing a significant increase in melting temperature from 1 μM onward (Fig. 5, I and J) while TOFA induced a significant decrease in the melting temperature of the PPARδ LBD from 0.1 μM onward (Fig. 5, K and L), indicative of a ligand-induced conformational rearrangement. Taken together, these findings suggest that the persistent hepatic PPAR transcriptional signal in TOFA-treated Pparɑ KO mice results from residual Pparδ activity, highlighting the importance of the PPAR network, specifically TOFA interaction with PPARα and PPARδ, in mediating the improvements in lipid and energy metabolism from TOFA treatment.

Synergistic effects of ACC1/2 inhibition and PPARα/δ agonism in treatment outcomes of TOFA

Given the dual action of TOFA, we aimed to determine the proportional contributions of ACC inhibition versus PPAR agonism on metabolic improvements. To test whether TOFA achieves greater outcomes compared to single agent treatment with mechanistic mimics, mice fed with a HFD were treated with either vehicle control, TOFA, Firsocostat (clinical ACC1/2 inhibitor), Elafibranor (clinically approved dual PPARα and PPARδ agonist), or a combination of Firsocostat and Elafibranor. Consistently, TOFA treatment outperformed both single and combination regimens, especially compared to Firsocostat, which elevated body weight and serum triglyceride levels. The negative effects of Firsocostat treatment were corrected by co-treatment with the PPARα/δ agonist Elafibranor, likely through Pparα activation (fig. S8, A to C). Notably, even as a single agent, TOFA treatment led to greater body weight reduction than Elafibranor alone or the combination of Elafibranor with Firsocostat (fig. S8, A and B). The hypertriglyceridemia in Firsocostat-treated mice was correlated with an increase in Srebp1c mRNA expression, which was not observed with other treatment regimens (fig. S8D). Furthermore, Firsocostat treatment did not impact fasting blood glucose or insulin levels while TOFA, Elafibranor, and combination treatment (FIR/ELA) reduced both, contributing to decreased risk of insulin resistance and diabetes progression, and suggesting the mechanistic dependence of PPARα and PPARδ signaling (fig. S8, E and F). With regards to liver triglyceride levels, TOFA and Firsocostat were the most effective single agents in reducing hepatic triglycerides, suggesting mechanistic ACC inhibition dependence in decreasing hepatic steatosis. In contrast, the PPARα/δ agonist Elafibranor showed only marginal effects on hepatic steatosis when used alone (fig. S8G). Accordingly, combination of Elafibranor with Firsocostat produced only a small decrease in liver triglyceride levels compared to Firsocostat alone (fig. S8G). Moreover, treatment of human Huh7 hepatoma cells with Firsocostat failed to elevate canonical PPARα/δ target genes CPT1A and VLDLR while both PPARα/δ-acting Elafibranor and TOFA do (fig. S8, H and I), demonstrating ACC-independent PPARα/δ target activation and consistent with our previous observation of PPARα/δ-dependence for energy expenditure-related transcriptional responses. Therefore, TOFA bridges the individual, beneficial actions of ACC inhibition and PPARα/δ activation and overcomes target-class shortcomings in one molecule by a synergistic dual-target mechanism.

TOFA treatment blunts MASLD/MASH development

Beyond its metabolic benefits in diet-induced obesity, glucose intolerance and insulin resistance, we evaluated TOFA in mouse models of MASLD/MASH. Male C57BL/6 J mice were fed a choline-deficient, L-amino acid-defined, high-fat diet (CDAA-HFD, a classic MASLD/MASH mouse model (52)) for eight weeks followed by treatment with vehicle or TOFA for four weeks. TOFA-treated mice demonstrated weight loss, with a significant change by day 14 that was sustained until end of treatment with an average weight loss of 8.7% (Fig. 6A). Overall, TOFA was orally tolerated as evidenced by no change in food intake (Fig. 6B). TOFA treatment also resulted in a trending decrease in serum ALT and AST, circulating markers of liver damage, compared to vehicle control (Fig. 6, C and D and fig. S9, A and B). Serum triglyceride levels were also reduced (Fig. 6E), however the lack of lipoprotein production due to dietary choline deficiency blunts the severity of hypertriglyceridemia in this model (53). Similar to our observations in DIO mice, TOFA significantly reduced liver triglyceride levels in this MASLD/MASH model (Fig. 6F and fig. S9C). The reduction of total hepatic lipid content directly correlated with ameliorated hepatic steatosis, revealed by histological staining (Fig. 6G). The expression of various inflammatory genes such as Il1b, Il6, and Tnfa, hallmarks of MASH, were also markedly and significantly reduced in the liver, consistent with a potent anti-inflammatory effect of TOFA treatment (Fig. 6H). Additionally, expression of collagen biosynthesis genes Col1a1 and Col3a was significantly reduced, indicating a key anti-fibrotic benefit of TOFA, targeting another hallmark of MASH (Fig. 6I). Complementary to the improvements in hepatic inflammation and fibrotic markers, livers of TOFA-treated mice had significantly lower hepatic hydrogen peroxide content, indicating decreased oxidative stress and damage (Fig. 6J). Overall, there was marked improvement in liver histology and decreased lipid accumulation (by H&E and Oil Red O) and fibrotic marker staining (by Sirius Red and Masson Trichrome) (Fig. 6K). Further histopathological scoring of livers was performed using a standardized scoring system for MASLD/MASH activity (54). Of the MASH components, overall decreases in inflammation and fibrosis were observed, with a modest effect on ballooning (Fig. 6L). The improvements in hepatic steatosis, inflammation, and collagen deposition in response to TOFA treatment were also found to be dose-dependent (fig. S9, D to G).

RNA-seq analysis of liver tissues from CDAA-HFD mice and subsequent Enrichr analysis using the WikiPathways 2024 Mouse database identified several lipid metabolism pathways, including PPAR signaling and fatty acid beta-oxidation, as enriched among these DEGs, consistent with the PPAR transcriptional signature observed in the DIO model (Fig. 6M). Accordingly, many of the top upregulated genes were associated with the PPAR signaling pathway, as identified by Enrichr analysis using KEGG 2021 Human dataset (Fig. 6N). Given the ongoing clinical trials of ACC inhibitors for MASLD/MASH, TOFA efficacy was benchmarked against Firsocostat efficacy in the CDAA-HFD model. TOFA-treated mice showed a more pronounced reduction in body weight (fig. S10A) without any changes in food intake (fig. S10B) and serum triglycerides (fig. S10C). TOFA treatment demonstrated similar efficacy as Firsocostat on MASLD/MASH disease markers, including serum ALT and AST (fig. S10, D and E), liver triglycerides (fig. S10F), and biomarkers of inflammation and collagen biosynthesis (fig. S10, G to J).

The MUP-uPA mouse fed a HFD is a well-established transgenic model for studying MASLD/MASH, as it induces chronic liver injury by triggering ER stress specifically in hepatocytes through the expression of urokinase-type plasminogen activator (uPA) under the liver-specific Mup promoter (55). The MUP-uPA model overcomes limitations of diet-induced models by combining metabolic features with advanced MASH progression, including bridging fibrosis, to better reflect human MASH pathological manifestations. After ten weeks of HFD feeding to accelerate disease progression, male MUP-uPA mice were treated with either vehicle control or TOFA for an additional six weeks. During the treatment period, TOFA-treated mice experienced significant weight loss by two weeks and sustained through the remainder of the treatment time course, with no changes in food intake (fig. S11, A and B). Consistent with our previous results in other diet-induced disease models, TOFA-treated mice exhibited an overall decrease in triglyceride levels, in both liver and serum, as well as a decrease in hepatic cholesterol (fig. S11, C to F). Liver histology revealed profound decreases in lipid accumulation by Oil Red O staining as well as a marked decrease in fibrosis as assessed by Sirius Red staining (fig. S11, G to I). RNA-seq analysis of liver samples further highlighted the strong upregulation of genes related to lipid metabolism and marked downregulation of a wide array of inflammatory signaling pathways (fig. S11, J and K). Taken together, TOFA was effective in addressing hepatic steatosis, inflammation, and fibrosis in this severe MASLD/MASH model.

TOFA enhances anti-obesity profile quality and persistence as a complementary agent to incretin analogs

While targeting energy intake to improve metabolic health has seen clinical success with incretin-based therapies (56, 57), targeting energy expenditure is an alternative, promising path for obesity treatment that could be complementary in combination with existing incretin-based approaches. We explored the potential for combinatorial effects of TOFA with incretin analogs in DIO mice with daily treatment of vehicle control, subcutaneous injection of minimum efficacious doses of semaglutide (Novo Nordisk) or tirzepatide (Eli Lilly), oral gavage of a sub-optimal dose of TOFA, or combinations for 24 days. Upon treatment initiation, tirzepatide and TOFA combination-treated mice experienced rapid onset weight loss requiring moving to lower maintenance dosing. Semaglutide and TOFA-treated mice demonstrated a slower, steady weight loss after a lag time of about a week. Overall, all monotherapy treatments demonstrated similar degrees of weight modulation relative to the controls (average 10% weight loss compared to controls), while the combination treatments significantly outperformed the monotherapies in weight loss (Fig. 7A). The significant changes in body weight were solely due to fat mass loss, with no significant loss of lean mass (Fig. 7B). Tirzepatide treatment alone strongly promoted improved glucose handling, and combination treatments (TOFA + TZP and TOFA + SEMA) demonstrated stronger improvements in glucose control (Fig. 7, C and D). Fasting insulin levels were also most improved in the combination treatment groups (Fig. 7E). Similarly, serum and hepatic triglycerides presented the greatest decreases in combination treatments, outperforming the single agent treatment effects (Fig. 7, F and G). Given the distinct mechanism of action between TOFA and incretin analogs, we investigated durability of weight loss by assessing time to weight rebound after treatment termination of semaglutide or TOFA fed a 60% HFD (Fig. 7H). After treatment termination, or the washout period, TOFA-treated animals were able to maintain body weight levels similar to that of control-treated mice while semaglutide-treated subjects experienced an immediate and stark body weight increase (Fig. 7I). As expected, body weight loss during semaglutide treatment was attributed to decreases in food intake, with body weight regain after treatment termination associated with significant increases in food intake; by contrast, no changes in food intake were observed between TOFA-treated and vehicle-treated mice (Fig. 7, J and K). Taken together, TOFA represents a complementary agent in combination with incretin analogs to promote greater weight loss and improve sustainability of metabolic and lipid health restoration independent of appetite-regulating mechanisms.

We also explored the use of semaglutide alone and in combination with TOFA in MASLD/MASH using the CDAA-HFD model. Consistent with previous results, the combination treatment yielded a greater degree of weight loss (fig. S12, A and B), greater decrease in lipid levels in the liver (fig. S12C), improvements in liver histopathology (fig. S12, D to G), and strong reductions in inflammation (fig. S12, H and I) and fibrosis (fig. S12, J and K) markers. These additional studies further highlight the complementary metabolic actions of TOFA to the current class of satiety-inducing incretin analogs that could be used in combination to outperform shortcomings of single agent treatments.

Experimental model and subject details

Method details

Funding:

This work was supported by discretionary funds from the University of California, Berkeley with assistance from the UCSF Liver Center (P30 DK026743) and the University of Michigan Animal Phenotyping Core (1U2CDK135066; MMPC-Live). RIS is supported by NIH/NIDDK P30 DK040561.

Author contributions:

Conceptualization: J.Y.L., C.Z., A.M.N. Methodology: J.Y.L., C.Z., A.M.N. Investigation: J.Y.L., C.Z., M.A.B., R.L.S., C.P., L.X., F.G., X.T., K.T.D., P.I., K.S., K.W. Visualization: G.B., K.C., P.I., R.I.S. Supervision: P.I., R.I.S., B.N.S., M.K., A.M.N. Writing—original draft: J.Y.L., C.Z. Writing—review & editing: J.Y.L., C.Z., R.I.S., A.M.N.

Competing interests:

A.M.N., J.Y.L., and P.I. are co-founders of ReRx Therapeutics, which develops drugs related to this work. A.M.N., J.Y.L., and C.Z. are inventors on patent application (U.S. pending patent no. PCT/US2023/028833, filed on 27 July 2023; U.S. pending patent no. PCT/US2025/013399, filed on 28 January 2025) submitted and held by the Regents of the University of California that covers the therapeutic use of TOFA in metabolic disease. All other authors declare they have no competing interests.

Data, code, and materials availability:

This study did not generate new materials. All data and code needed to evaluate and reproduce the results in the paper are present in the paper and/or the Supplementary Materials. Raw data (.fastq), processed data (count matrix.h5 files and fragment.bed files), and meta data of the RNA-seq data studies have been deposited in NCBI’s Gene Expression Omnibus under GEO Series accession number, GSE302447 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE302447).

This PDF file includes: