Dietary sulfur amino acid restriction elicits a cold-like transcriptional response in inguinal but not epididymal white adipose tissue of male mice

eLife Assessment

The present study employed transcriptomics to investigate the impact of methionine restriction (MR) and cold exposure (CE) on liver and adipose tissues in mice. The authors demonstrate that responses to MR and CE are tissue-specific, while both MR and CE have a similar effect on beige adipose tissue. While these findings are somewhat descriptive, this work is considered important, as it provides a comprehensive resource for enhancing our understanding of these lifestyle interventions. The study is of high scientific quality, and the analyses are convincing.

https://doi.org/10.7554/eLife.108825.3.sa0

Significance of the findings:

Important: Findings that have theoretical or practical implications beyond a single subfield

Landmark
Fundamental
Important
Valuable
Useful

Strength of evidence:

Convincing: Appropriate and validated methodology in line with current state-of-the-art

Exceptional
Compelling
Convincing
Solid
Incomplete
Inadequate

During the peer-review process the editor and reviewers write an eLife Assessment that summarises the significance of the findings reported in the article (on a scale ranging from landmark to useful) and the strength of the evidence (on a scale ranging from exceptional to inadequate). Learn more about eLife Assessments

Abstract
Introduction
Results
Discussion
Materials and methods
Data availability
References
Article and author information
Metrics

Abstract

About 1 billion people are living with obesity worldwide. GLP-1-based drugs have massively transformed care, but long-term consequences are unclear in part due to reductions in energy expenditure with ongoing use. Diet-induced thermogenesis (DIT) and cold exposure (CE) raise EE via brown adipose tissue (BAT) activation and beiging of white adipose tissue (WAT). Methionine restriction (MetR) is a candidate DIT stimulus, but its EE effect has not been benchmarked against CE, nor have their tissue-level interactions been defined. In a 2×2 design (Control vs. MetR; room temperature, RT: 22°C vs. CE: 4°C for 24 hr), we used male C57BL/6 N mice to benchmark MetR-induced thermogenesis against CE and mapped how diet and temperature interact across tissues. Bulk RNA-seq profiled liver, iBAT, iWAT, and eWAT. Differential expression was modeled with main effects and a diet × temperature interaction. KEGG GSEA was used to assess pathway-level enrichment. MetR increased EE at RT and shifted fuel use towards lipid oxidation, supporting MetR as a bona fide DIT stimulus. CE elevated EE across diets and blunted diet differences. Transcriptomic responses were tissue-specific: in liver, CE dominated gene induction while MetR and CE cooperatively repressed genes. The combination enriched glucagon/AMPK-linked and core metabolic pathways. In iBAT, CE dominated thermogenic and lipid-oxidation programs with minimal MetR contribution. In iWAT, MetR and CE acted largely additively with high concordance, enhancing fatty-acid degradation, PPAR signaling, thermogenesis, and TCA cycle pathways. In eWAT, robust co-dependent and synergistic differential expression emerged only with MetR+CE. MetR is a genuine DIT stimulus that remodels metabolism in a tissue-specific manner. Our study provides a tissue-resolved transcriptomic resource that benchmarks diet-induced (MetR) against cold-induced thermogenesis and maps their interactions across liver, iBAT, iWAT, and eWAT.

Introduction

Obesity has become a global health crisis, with prevalence rates reaching alarming levels in many countries. The World Health Organization (WHO) estimates that over 890 million adults worldwide are obese (Phelps et al., 2024), leading to a dramatic increase in obesity-related diseases such as type 2 diabetes, cardiovascular disease, and certain cancers. As traditional weight-loss strategies often fail to provide lasting results, new therapeutic approaches are urgently needed. The increase in energy expenditure via brown adipose tissue (BAT) activation has emerged as a promising strategy for combating obesity. Unlike white adipose tissue (WAT), which stores excess energy, BAT is specialized in generating heat and driving energy expenditure via non-shivering thermogenesis, typically in response to cold exposure (CE). Pharmacologically or environmentally induced BAT activity results in improved metabolic health in mice and humans (Schlein and Heeren, 2016; Yoneshiro et al., 2013). Conversely, diet-induced thermogenesis (DIT) refers to the increase in energy expenditure associated with the digestion, absorption, and metabolism of food.

Recent studies have suggested that sulfur amino acid restriction, also referred to as methionine restriction (MetR), a dietary intervention that limits the intake of the amino acids methionine and cysteine, can enhance energy expenditure (EE) and promote metabolic health (Fang et al., 2022). Methionine and cysteine are amino acids involved in a plethora of metabolic reactions including protein synthesis, gene expression via methylation of DNA and histones, maintenance of DNA and RNA integrity via polyamine synthesis, redox balance via glutathione and H₂S metabolism, and nucleotide biosynthesis via the folate cycle (Fang et al., 2022; Forney et al., 2020).

Studies have shown that MetR is sensed in the liver by the GCN2-PERK-ATF4-mediated integrated stress response (Wanders et al., 2015). Initially, it was thought that increases in circulating FGF21 drive increases in EE by activating UCP1-driven thermogenesis in brown adipose tissue via β-adrenergic (βr) signaling (Fang et al., 2022; Babygirija and Lamming, 2021). However, more recent studies also show that at least some of the metabolic benefits develop in βr-incompetent, FGF21-KO, and UCP1-KO animals, with or without increases in EE and independently of ambient temperature (Forney et al., 2020; Wanders et al., 2015; Wanders et al., 2017; Lee et al., 2025), highlighting the complexity of the physiological response to the dietary restriction of sulfur amino acids. Recently, it has been proposed that sulfur amino acid restriction results in metabolic inefficiency resulting in the excretion of various metabolites including β-hydroxybutyrate, pyruvate, citrate, alpha-ketoglutarate, carnosine, and others (Varghese et al., 2025).

The above-mentioned studies highlight an intriguing overlap between diet- and cold-induced thermogenesis, particularly in the context of energy expenditure regulation. Both stimuli rely on the activation of brown adipose tissue (BAT) and the ‘beiging’ of inguinal white adipose tissue (iWAT) to promote UCP1 expression and increase mitochondrial activity, and thereby boost calorie dissipation, potentially via similar signaling pathways, including the sympathetic nervous system and key metabolic regulators like FGF21 (Spann et al., 2021). However, whether diet- and cold-induced thermogenesis produces additive or synergistic effects on the activation of energy and systemic metabolism is unknown. Here, we systematically compare the physiological and transcriptional responses to MetR and CE across multiple metabolically active tissues. Using RNA sequencing, we dissect additive, synergistic, and antagonistic gene regulatory patterns and assess whether combining MetR with CE produces tissue-specific concordant or discordant novel transcriptional outcomes.

We demonstrate that MetR increased EE at RT and shifted fuel use toward lipid oxidation. CE elevated EE across diets and blunted diet differences. The transcriptional responses were tissue-specific at the gene and pathway level. Taken together, our results provide a unique and comprehensive gene regulatory framework to understand how dietary and environmental cues converge to shape tissue-specific gene expression programs and metabolic adaptation.

Results

Physiological and metabolic effects of methionine restriction (MetR) and cold exposure (CE)

To investigate the physiological impact of dietary sulfur amino acid content on EE under room temperature (RT, i.e., 22°C) and cold exposure (CE, 4°C for 24 hr), male C57BL/6 N mice were placed on one of three cysteine-depleted diets containing either 0.8% Methionine (Control; Ctrl), 0.12% Methionine (Methionine-restricted; MetR) or 2% Methionine (Methionine-supplemented; MetS) for 6 and 5 days, respectively (Figure 1A, i.e. study 1). Following the diet switch from housing diet to aforementioned experimental diets, EE increased progressively in the MetR group under RT conditions, rising from approx. 0.45 kcal/hr to 0.6 kcal/hr. Conversely, EE did not change in the Ctrl and MetS groups (Figure 1B). On day 6, energy expenditure in the MetR group was significantly elevated compared to Ctrl and MetS groups, independent of starting body weight (Figure 1C, left). In contrast, CE elevated EE in all three diet groups to ~0.9 kcal/hr, thereby rescinding the differences between MetR and the other two diets at RT (Figure 1A and C right, Figure 1—figure supplement 1A). MetR-fed animals exhibited significantly greater body weight loss compared to Ctrl and MetS-fed animals over the total duration of experimental diet feeding (Figure 1D). As food intake and locomotor activity were similar across groups (Figure 1—figure supplement 1B and D), the greater weight loss in MetR-fed animals is likely attributable to increased energy expenditure under RT conditions.

Figure 1 with 1 supplement see all

Download asset Open asset

Physiological and metabolic effects of methionine restriction (MetR) and cold exposure (CE).

(A) Schematic depicting 11-day experimental setup and dietary composition for mouse experiment 1 (n=5 animals/group). (B) Energy expenditure (EE) over the entire experiment duration. (C) ANCOVA analysis of average nighttime EE on days 6 and 11 over body weight. (D) Bodyweight and body weight loss (%). (E) Respiratory exchange ratio (RER) over the entire experiment duration. (F) Average daily food and water intake over all RT (22°C) or CE (4°C) experimental days. (G) Schematic depicting 8 day experimental setup for mouse experiment 2 (n=7 animals/group). Dietary compositions for Ctrl and MetR diets are the same as in mouse experiment 1. Groups are denominated as Ctrl_RT, Ctrl_CE, MetR_RT, and MetR_CE. (H) Bodyweight and body weight loss (%). (I) Absolute organ weights for Liver, inguinal WAT, gonadal WAT, interscapular BAT. Statistics were done within tissues. (J) Blood glucose, serum triglycerides, and FGF21 levels. EE was analyzed by ANCOVA using body weight as a covariate (via https://www.mmpc.org/); p-values were Bonferroni-corrected. Bodyweight, average food and water intake were analyzed via two-way ANOVA with *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Body weight loss, serum parameters, and absolute organ weights were analyzed using one-way ANOVA with Tukey’s post hoc test. Different letters indicate statistically significant differences (p < 0.05). Sample sizes for experiment 1 are 5 animals/group and 5–7 animals/group for Experiment 2. Error bars are SEM.

Consistent with previous studies, MetR feeding also increased water intake (Figure 1—figure supplement 1C). Respiratory exchange ratios (RER), as an indicator for macronutrient fuel selection, progressively increased during light and dark phases in the Ctrl and MetS groups, indicating enhanced carbohydrate utilization, whereas MetR-fed animals demonstrated lower RER indices, indicating a preference for lipid oxidation. Upon CE, RER decreased in Ctrl and MetS-fed animals to levels comparable to those observed in MetR-fed animals (Figure 1E, Figure 1—figure supplement 1F). To further assess dietary and temperature interactions, we calculated daily averages of food and water intake, as well as locomotor activity and RER for each period. At RT, daily caloric intake was similar between groups, and CE led to a consistent increase in caloric intake in all groups (Figure 1F). Average daily water intake was unaffected by ambient temperature, whereas daily locomotor activity exhibited a downward trend in all three groups during cold exposure (Figure 1F, Figure 1—figure supplement 1E). Collectively, these data demonstrate that short-term MetR increases energy expenditure by ca. 20% at RT and, as a consequence, promotes a metabolic shift toward lipid oxidation, potentially due to a transcriptional induction of thermogenic processes and/or lipolysis in catabolic adipose tissue depots such as iWAT or BAT. However, CE as a major physiological stimulus to induce non-shivering thermogenesis (NST) was able to override this MetR-induced effect, prompting an additional ca. 60% increase in EE independent of experimental diet feeding and body weight, alongside a shift to lipid utilization.

To further investigate the potential interactions between MetR- and CE-induced thermogenesis on physiological, metabolic, and transcriptional parameters, a second experiment was conducted: Here, mice were divided into four groups and either fed Ctrl or MetR diets (see study 1) for seven days at 22°C, or additionally exposed to 4°C for 24 hr (Study 2). In the following, animals fed Ctrl or MetR diets for 7 days at 22°C are denominated as Ctrl_RT and MetR_RT, while animals fed these diets for 7 days before being subjected to 24 hr of cold are denominated as Ctrl_CE and MetR_CE (Figure 1G). All groups experienced significant body weight loss during the intervention (Figure 1H), presumably due to single-housing. Counterintuitively, Ctrl_CE did not display exacerbated weight loss compared to the Ctrl_RT group. By contrast, MetR_RT led to approximately 10% body weight loss, and the combined treatment (MetR_CE) resulted in further weight loss compared to MetR_RT alone (Figure 1H). The reductions in body weights coincided with organ-specific alterations in organ wet weights and organ/body weight ratios. Both Ctrl_CE and MetR_RT resulted in absolute (Figure 1I) and relative (Figure 1—figure supplement 1G) reductions in liver mass, and MetR_CE reduced absolute and relative liver mass further in an additive fashion. The stepwise reduction in relative liver mass indicates that liver atrophy partly accounted for the observed body weight loss, predominantly in MetR-exposed mice. In contrast, Ctrl_CE alone did not affect iWAT, gWAT, or iBAT mass. Intriguingly, under MetR_RT, all three adipose tissues showed (non-significant) upward trends in relative and absolute masses, which was exacerbated and partially significant under MetR_CE (Figure 1I, Figure 1—figure supplement 1G). Intriguingly, these data indicate that MetR_RT and Ctrl_CE synergize to promote increases in adipose tissue mass, particularly in gWAT and iBAT. To assess the associated systemic metabolic sequelae of MetR- and CE, we measured circulating metabolic parameters: Blood glucose levels did not change significantly but displayed a stepwise reduction between the groups, indicating additive effects between diet and ambient temperature (Figure 1J). In line with previous studies, serum triglycerides were reduced, albeit non-significantly, by Ctrl_CE and MetR_RT (Grefhorst et al., 2018; Hasek et al., 2013; Bartelt et al., 2011), but showed an even stronger reduction under MetR_CE, suggesting synergy (Figure 1J). By contrast, serum NEFA levels showed only minor non-significant reductions in both CE conditions and were unchanged by MetR_RT alone (Figure 1J), while serum β-hydroxybutyrate levels, a marker of hepatic fatty acid oxidation, were elevated by either CE or MetR feeding (Figure 1—figure supplement 1H). A major hormonal stimulus of EE in response to CE or MetR is mediated via ‘thermogenic’ hormones like FGF21 or IL-6. In line with the literature, Ctrl_CE and MetR_RT elevated circulating FGF21 levels even though the effect of cold was limited in this study. FGF21 levels did not further increase under MetR_CE, indicating that MetR maximally activates Fgf21 transcription / FGF21 secretion to induce EE (Figure 1J). In contrast to literature, Ctrl_CE resulted in significant reductions in serum IL-6 levels in this study (Figure 1—figure supplement 1H; Bal et al., 2017). MetR_RT also reduced serum IL-6 levels, with no further alterations under MetR_CE (Figure 1—figure supplement 1H). Together, these results suggest that while cold exposure and MetR feeding individually trigger overlapping metabolic responses, their combination produces additive effects on weight loss, liver atrophy, and adipose size, alongside specific synergistic effects on triglyceride levels.

The transcriptional responses to CE and MetR-induced thermogenesis are tissue-specific

To investigate the transcriptional basis of the adaptation to two independent, yet potentially synergistic, physiological stimulators of NST, we next performed bulk mRNA-seq on liver, iBAT, iWAT, and eWAT across all four groups. Principal Component Analysis (PCA), as expected, revealed that the samples clustered according to their tissue of origin (Figure 2—figure supplement 1A). We next performed tissue-intrinsic PCA analyses that demonstrated highly tissue-specific responses (Figure 2A). Phelps et al., 2024 In the Liver, all four groups showed distinct transcriptional responses in PC1 and PC2, whereas in Schlein and Heeren, 2016 iBAT the groups clustered mostly by temperature (PC1), indicating that the transcriptional effects of MetR feeding are mild compared to CE. By contrast, in both white adipose depots, the four groups showed significant overlap. Although, in Yoneshiro et al., 2013 iWAT, the clusters showed additive effects of MetR and CE along PC1, explaining 60% of the variation, while in Fang et al., 2022 eWAT, the clusters overlapped in PC1 and PC2. Differential gene expression analysis (cutoff: adj. pvalue <0.05, |FC| ≥ 1.5) showed that Ctrl_CE elicited stronger responses than MetR_RT in all tissues as exemplified by the higher number of differentially expressed genes (DEGs) without clear trends towards global transcriptional induction or repression. Of note, MetR was coupled to a higher number of repressed genes, potentially linked to the increased need for cellular energy conservation during amino acid restriction. The combination of both stimuli elicited the strongest transcriptional responses in iWAT, and apart from the interaction term, the transcriptional response in eWAT appeared limited (Figure 2, Figure 2—figure supplement 1).

Figure 2 with 1 supplement see all

Download asset Open asset

The transcriptional responses to cold exposure (CE) and MetR-induced thermogenesis are tissue-specific.

(A) Tissue-specific PCA plots for liver, iBAT, iWAT, and eWAT. (B) Number of differentially expressed genes (DEGs), split into induced (red) and repressed (blue) genes, per contrast (adjusted p-value <0.05, |FC|>1.5). (**C–F**) UpSet plots showing overlap of DEGs across tissues for each contrast: (C) Ctrl_CE vs Ctrl_RT, (D) MetR_RT vs Ctrl_RT, (E) MetR_CE vs Ctrl_RT, and (F) Interaction. Set size bars indicate the total number of DEGs per tissue; intersection bars indicate the number of shared DEGs between tissues. Dots and bars in orange represent tissue-specific DEGs. In the order of Ctrl_RT, Ctrl_CE, MetR_RT, and MetR_CE, the following number of samples were included in the transcriptomic analysis: 7, 7, 7, and 7 (Liver); 7, 6, 7, and 7 (iBAT); 7, 7, 7, and 7 (iWAT); and 4, 7, 6, and 7 (eWAT; see Materials and methods).

To understand to what degree the transcriptional responses are tissue/depot-specific or broadly applicable to all investigated organs, we analyzed the intersections between CE and MetR in all four tissues (Figure 2C–F, Figure 2—figure supplement 1C and D). Under Ctrl_CE and MetR_CE, more than 50% of all DEGs were regulated in a tissue-specific manner in three out of the four tissues, with the most marked transcriptional effects seen in iWAT (Figure 2C–E). The degree of tissue-specific regulation was even higher in MetR_RT and in the interaction term (Figure 2D and F). Cumulatively, these data suggest that the majority of transcriptional effects of CE and dietary MetR are tissue-specific and display a varying degree of additive and synergistic effects.

To better scrutinize the individual transcriptional effects for each tissue, and to associate the observed gene-regulatory effects with cellular and metabolic processes, we next performed binary, combinatorial analyses of differential gene expression in liver (Figure 3), iBAT (Figure 4), iWAT (Figure 5), and eWAT (Figure 6) using similar in silico approaches.

Figure 3 with 1 supplement see all

Download asset Open asset

Cold exposure (CE) drives gene induction while methionine restriction (MetR) and CE cooperatively repress genes in the liver.

(A) Volcano plots showing DEGs for each contrast (Ctrl_CE vs Ctrl_RT, MetR_RT vs Ctrl_RT, MetR_CE vs Ctrl_RT, and the diet × temperature interaction). Numbers indicate significantly up- and downregulated genes (adjusted p-value <0.05, |FC|>1.5). (B) Upset plot showing the overlap of DEGs across contrasts. Intersection bars indicate the number of shared DEGs between contrasts. Dots in orange represent contrast-specific DEGs. (C) Scatter plot of log₂FC values in Ctrl_CE vs Ctrl_RT and MetR_RT vs Ctrl_RT for DEGs identified in MetR_CE. n denominates DEGs shown. (D) Schematic highlighting gene expression profiles. Arrows indicate significant regulation for induced (up; red) or repressed (down; blue) genes. Non-significant regulation is depicted as gray dots (see Materials and methods). (E) Classification of MetR_CE DEGs based on their mode of regulation mode. (F) GSEA of the top 20 positively enriched pathways in MetR_CE vs Ctrl_RT. Dot size represents gene ratio, color denotes contrast, and significance is indicated by filled circles (adjusted p-value <0.05). (G) Heatmap of Top 30 induced genes under MetR_CE. (H) GSEA for the top 14 negatively enriched pathways in the MetR_CE vs Ctrl_RT comparison. Dot size reflects gene ratio, colors indicate contrast group, and significance is shown by filled circles (adjusted p-value <0.05). (I) Heatmap of Top 30 repressed genes under MetR_CE.

Figure 4 with 1 supplement see all

Download asset Open asset

CE dominates gene induction in iBAT with limited contribution by MetR feeding.

(A) Volcano plots showing DEGs for each contrast (Ctrl_CE vs Ctrl_RT, MetR_RT vs Ctrl_RT, MetR_CE vs Ctrl_RT, and the diet × temperature interaction). Numbers indicate significantly up- and downregulated genes (adjusted p-value <0.05, |FC|>1.5). (B) Upset plot showing the overlap of DEGs across contrasts. Intersection bars indicate the number of shared DEGs between contrasts. Dots in orange represent contrast-specific DEGs. (C) Scatter plot of log₂FC values in Ctrl_CE vs Ctrl_RT and MetR_RT vs Ctrl_RT for DEGs identified in MetR_CE. n denominates DEGs shown. (D) Classification of MetR_CE DEGs based on their mode of regulation. (E) GSEA of the top 10 positively enriched pathways in MetR_CE vs Ctrl_RT. Dot size represents gene ratio, color denotes contrast, and significance is indicated by filled circles (adjusted p-value <0.05). (F) Heatmap of Top 30 induced genes under MetR_CE. (G) GSEA for the top 20 negatively enriched pathways in the MetR_CE vs Ctrl_RT comparison. Dot size reflects gene ratio, colors indicate contrast group, and significance is shown by filled circles (adjusted p-value <0.05). (H) Heatmap of Top 30 repressed genes under MetR_CE.

Figure 5 with 1 supplement see all

Download asset Open asset

Additive and synergistic gene regulation by MetR and CE in iWAT.

(A) Volcano plots showing DEGs for each contrast (Ctrl_CE vs Ctrl_RT, MetR_RT vs Ctrl_RT, MetR_CE vs Ctrl_RT, and the diet × temperature interaction). Numbers indicate significantly up- and downregulated genes (adjusted p-value <0.05, |FC|>1.5). (B) Upset plot showing the overlap of DEGs across contrasts. Intersection bars indicate the number of shared DEGs between contrasts. Dots in orange represent contrast-specific DEGs. (C) Scatter plot of log₂FC values in Ctrl_CE vs Ctrl_RT and MetR_RT vs Ctrl_RT for DEGs identified in MetR_CE. n denominates DEGs shown. (D) Classification of MetR_CE DEGs based on their mode of regulation. (E) GSEA of the top 20 positively enriched pathways in MetR_CE vs Ctrl_RT. Dot size represents gene ratio, color denotes contrast, and significance is indicated by filled circles (adjusted p-value <0.05). (F) Heatmap of Top 30 induced genes under MetR_CE. (G) GSEA for the top 20 negatively enriched pathways in the MetR_CE vs Ctrl_RT comparison. Dot size reflects gene ratio, colors indicate contrast group, and significance is shown by filled circles (adjusted p-value <0.05). (H) Heatmap of Top 30 repressed genes under MetR_CE.

Figure 6 with 1 supplement see all

Download asset Open asset

Limited yet codependent and antagonistic gene regulation by MetR and CE in eWAT.

(A) Volcano plots showing DEGs for each contrast (Ctrl_CE vs Ctrl_RT, MetR_RT vs Ctrl_RT, MetR_CE vs Ctrl_RT, and the diet × temperature interaction). Numbers indicate significantly up- and downregulated genes (adjusted p-value <0.05, |FC|>1.5). (B) Upset plot showing the overlap of DEGs across contrasts. Intersection bars indicate the number of shared DEGs between contrasts. Dots in orange represent contrast-specific DEGs. (C) Scatter plot of log₂FC values in Ctrl_CE vs Ctrl_RT and MetR_RT vs Ctrl_RT for DEGs identified in MetR_CE. n denominates DEGs shown. (D) Classification of MetR_CE DEGs based on their mode of regulation. (E) GSEA of the top 10 positively enriched pathways in MetR_CE vs Ctrl_RT. Dot size represents gene ratio, color denotes contrast, and significance is indicated by filled circles (adjusted p-value <0.05). (F) Heatmap of Top 30 induced genes under MetR_CE. (G) GSEA for the top 20 negatively enriched pathways in the MetR_CE vs Ctrl_RT comparison. Dot size reflects gene ratio, colors indicate contrast group, and significance is shown by filled circles (adjusted p-value <0.05). (H) Heatmap of Top 30 repressed genes under MetR_CE.

CE drives gene induction while MetR and CE cooperatively repress genes in the liver

In the liver (Figure 3), Volcano plot analysis revealed substantial gene regulation induced by Ctrl_CE (977 genes upregulated, 322 genes downregulated) and MetR_RT (257 genes upregulated, 387 genes downregulated), with even more genes regulated when combined under MetR_CE (815 genes upregulated, 692 genes downregulated). While Ctrl_CE appears to be a stronger stimulus based on the number of DEGs, MetR_RT appears to elicit stronger transcriptional effects based on log₂FC (Figure 3A). Based on the number of DEGs, MetR_RT and Ctrl_CE, when combined under MetR_CE, appear to repress gene expression in an independent, additive manner and induce genes in a dependent or antagonistic manner (Figure 3A). Indeed, only a very limited number of genes were significant in diet × temperature interaction term (36 genes upregulated, 77 genes downregulated; Figure 3A) and the transcriptional effects of CE under MetR (MetR_CE vs MetR_RT) and MetR feeding under CE (MetR_CE vs Ctrl_CE) were lower compared to their respective counterparts (Figure 3—figure supplement 1A). To gain further insights into the degree of overlap between Ctrl_CE and MetR_RT, we performed Upset plot analysis. The biggest set of shared genes was found between Ctrl_CE and MetR_CE (541 DEGs), highlighting that Ctrl_CE is a major contributor to the transcriptional effects seen in MetR_CE (Figure 3B). While the direct contribution of MetR_RT alone was lower (251 DEGs), this analysis also indicated a high degree of co-dependence on MetR_RT in the regulation of 508 DEGs in the MetR_CE condition, as well as the exclusion of 494 Ctrl_CE DEGs from the combined exposure (Figure 3B). Contrasting all MetR_CE DEGs between Ctrl_CE and MetR_RT (R²=0.34), as well as contrasting all MetR_RT and Ctrl_CE DEGs (R²=0.26), revealed relatively poor correlation between the respective log₂FC values and suggests that the overall influence of Cold is stronger (Figure 3C, Figure 3—figure supplement 1B). Next, we analyzed the transcriptional interaction between Ctrl_CE and MetR_RT (Figure 3D, Figure 3—figure supplement 1C, see methods) under consideration of significance, direction of regulation, and interaction significance based on the method proposed by Pantaleón García et al., 2022. This analysis indicated that more than 50% of down-regulated genes were due to co-dependent additive regulation between Ctrl_CE and MetR_RT (brown) and MetR_RT alone (green). By contrast, more than 50% of the up-regulated genes were significant in the combined exposure due to Ctrl_CE alone (Figure 3D). Altogether, these data confirm Ctrl_CE drives the induction of genes under MetR_CE, while Ctrl_CE and MetR_RT cooperate to repress genes under MetR_CE. On the other hand, MetR_RT prevented a significant number of Ctrl_CE DEGs (n=547 genes) from regulation under MetR_CE in an additive (subtractive) fashion (Figure 3—figure supplement 1D). Finally, to understand which pathways are differentially regulated, we performed Gene set enrichment analysis (GSEA) under consideration of log₂FC values (Subramanian et al., 2005). Among the 20 most enriched pathways under MetR_CE, MetR drove the enrichment for metabolic pathways (Biosynthesis of amino acids, Glutathione or P450 cytochrome (Drug) metabolism) despite the effects of Ctrl_CE, while CE drove the enrichment in Steroid biosynthesis and cGMP-PKG signaling pathways, despite the effects of MetR_RT (Figure 3F). These results are in line with previously published datasets (Zhang et al., 2022; Ghosh et al., 2017). This analysis also revealed a number of gene sets where MetR_CE showed the highest enrichment, such as Glucagon and AMPK signaling, TCA cycle, One carbon metabolism, and amino acid biosynthesis. This indicates that MetR and CE cooperate in the regulation of gene sets related to energy provisioning and glucose metabolism in the liver (Figure 3F). The top 30 heatmap of induced genes under MetR_CE reflects these additive effects of Ctrl_CE and MetR_RT on the level of pathway enrichment and features genes such as Pck2 and Lepr (Glucagon signaling), Pck2 (TCA cycle), Mthfd1l and Aldh1l2 (TCA cycle and folate metabolism), and Psat1, Etnppl, and Acot4 (amino acid biosynthesis; Figure 3G). The only non-immune-related pathway among the 14 most negatively enriched pathways was the Ribosome gene set, where Cold and MetR feeding had opposing effects (Figure 3H). In accordance with the contrasting enrichment by MetR_RT and Ctrl_CE alone, the top 30 heatmap of MetR_CE repressed genes only features immune-related genes including Saa1, Saa2, Saa3 and Orm1 and Orm2 (acute-phase response), Lcn2 and Lrg1 (innate immunity; Figure 3I). Furthermore, additional heatmap analyses of the top 30 up- and down-regulated genes for Ctrl_CE and MetR_RT further highlight the distinct and cooperative transcriptional effects between both stimuli (Figure 3—figure supplement 1E–J).

CE dominates gene induction in iBAT with limited contribution by MetR feeding

In iBAT (Figure 4), Volcano plot analysis revealed that Ctrl_CE elicited a much stronger transcriptional response than MetR_RT (757 genes upregulated, 912 genes downregulated, versus 57 genes upregulated and 173 downregulated, respectively). The transcriptional response to MetR_CE was only marginally different from Ctrl_CE (664 genes upregulated, 1091 genes downregulated), suggesting that the majority of the transcriptional response is driven by CE (Figure 4A). Only a very small number of genes were identified in the diet ×temperature interaction term (22 upregulated, 21 downregulated), indicating that the combination of CE and MetR does not result in non-additive transcriptional effects in iBAT (Figure 4A). In line with this, CE on top of MetR (MetR_CE vs MetR_RT) still resulted in substantial gene regulation (Figure 4—figure supplement 1A). Upset plot analysis confirmed that the majority of DEGs in the combined MetR_CE condition overlapped with Ctrl_CE (1009 genes), while only a minor subset was shared with MetR_RT (85 genes; Figure 4B). Comparison of log₂FC between Ctrl_CE and MetR_RT for all MetR_CE DEGs showed a weak correlation (R²=0.24), supporting the notion that the transcriptional responses to MetR_CE are indeed driven by CE (Figure 4C). This weak correlation was confirmed when considering all MetR_RT or Ctrl_CE DEGs (Figure 4—figure supplement 1B). The classification of gene regulation confirmed that >50% of down- and up-regulated genes were driven by Ctrl_CE (blue, Figure 4D). Altogether, these data indicate that Ctrl_CE primarily drives gene regulation in iBAT, while the transcriptional effects of MetR_RT are negligible. GSEA analysis confirmed the positive enrichment of fatty acid and thermogenic pathways by Ctrl_CE, as previously published (Ghosh et al., 2017; Hao et al., 2015). The combination of both stimuli specifically benefited the positive enrichment of metabolic pathways such as fatty acid elongation, biosynthesis of unsaturated fatty acids, and PPAR signaling, while Ctrl_CE drove the enrichment in ER and glycerolipid-related gene sets. In line with the limited transcriptional response of MetR_RT on gene level, NES scores for MetR_RT were mostly lower compared to Ctrl_CE and/or non-significant, except for Steroid biosynthesis (Figure 4E). In line, the heatmap of the top 30 upregulated genes under MetR_CE underscores this enrichment by featuring important factors such as Ppargc1a, Fgf21, Gdf15, Gpr3, and Bmp8b (PPAR signaling and thermogenesis), Elovl3 and Scd3 (fatty acid elongation), and Mthfd2 (folate metabolism; Figure 4F). Many of these genes are also featured in the top 30 heatmap for Ctrl_CE regulated genes (Figure 4—figure supplement 1D). By contrast, negatively enriched gene sets under MetR_CE were largely immune- and extracellular matrix–related, including cytokine-, ECM- and neuroactive ligand-receptor interactions and cell adhesion gene sets (Col1a1, Col1a2, Ngfr, Hcar2, Adcy5). Interestingly, the negative enrichment of these pathways was dominated by MetR_RT (Figure 4G and H). Altogether, these analyses support the notion that the transcriptional effects of MetR_RT are limited and contribute little to the combined effect of MetR_CE, which is dominated by cold (Figure 4—figure supplement 1D–G).

Additive and synergistic gene regulation by MetR and CE in iWAT

In iWAT (Figure 5), Volcano plots revealed that both Ctrl_CE (992 upregulated, 1379 genes downregulated) and MetR_RT (680 upregulated, 1156 downregulated) elicited marked transcriptional changes: The combination of both stimuli (MetR_CE) resulted in an even greater number of DEGs compared to either stimuli alone (1394 upregulated, 2261 downregulated), indicating either additive or synergistic effects (Figure 5A). Only a limited number of genes were identified in the diet × temperature interaction (188 down, 55 up), suggesting few non-additive responses (Figure 5A). Upset plots suggest cooperativity between MetR_RT and Ctrl_CE, with 1234 DEGs uniquely regulated when both stimuli were combined, but also support true additive or synergistic regulation with 1210 genes being significantly regulated in all three conditions (Figure 5B). Correlation of log₂FC between Ctrl_CE and MetR_RT for all MetR_CE showed a strong positive correlation (R²=0.81), suggesting that the impact of CE and MetR feeding on gene regulation is similar in iWAT (Figure 5, Figure 5—figure supplement 1). Regulatory classification of MetR_CE DEGs revealed that 55% of up- and 75% of down-regulated genes were predominantly regulated in an additive manner, where Ctrl_CE and MetR_RT cooperated to enhance transcriptional responses. This included genes that became significant when both stimuli were combined (co-dependent regulation) as well as genes that were already significantly regulated under each condition alone but showed non-synergistic regulation under the combined treatment (concordant – additive; Figure 5D). GSEA analysis revealed in pathways consistent with expected CE- or MetR-induced responses, such as fatty acid degradation, PPAR signaling, thermogenesis, and the TCA cycle (Figure 5E; Ghosh et al., 2017; Hao et al., 2015). Notably, nearly all of these pathways showed significant negative enrichment for the interaction term, indicating that the transcriptional response to MetR_CE is less than additive. This suggests that while both Ctrl_CE and MetR_RT independently activate key metabolic pathways, their combined effect is attenuated on the gene set level, possibly due to overlapping mechanisms or transcriptional feedback limiting further activation. Heatmap analysis of the top 30 induced genes matches GSEA results and features prominent PPAR and thermogenesis-related genes including Dio2, Ucp1, Fgf21, Elovl3, Cpt1b, and Ppargc1a. The heatmap analysis also confirms the largely additive gene regulatory effects of Ctrl_CE and MetR_RT on gene level (Figure 5F, Figure 5—figure supplement 1D and E). Similar to the responses in iBAT, iWAT also showed negative enrichment in predominantly immune-related pathways, such as cytokine-cytokine receptor interactions (Ccl2, Ccl7, Ccl8, Ccl12, Ccr5, Il1rl1; Figure 5G and H).

Limited yet codependent and antagonistic gene regulation by MetR and CE in eWAT

In eWAT (Figure 6), Volcano plot analysis revealed that both CE and MetR feeding alone did not have substantial effects on transcription, as less than 189 DEGs were up- or down-regulated in either condition (Figure 6A). This combined effect under MetR_CE also featured merely 122 up- or down-regulated genes, while the interaction term identified 193 up- and 292 down-regulated DEGs, the highest among all tissues, suggesting a potential synergistic mode of gene regulation under MetR_CE in this tissue (Figure 6A). Upset plot analysis confirmed that most MetR_CE DEGs (158) are unique to the combination of CE and MetR, and that Ctrl_CE (31 DEGs) or MetR_RT (24 DEGs) alone contributed little to the joined MetR_CE effect (Figure 6B). Scatter plot analysis of log₂FC values between Ctrl_CE and MetR_RT for all MetR_CE DEGs revealed a relatively high correlation (R²=0.54) and were predominantly categorized as Co-dependent–additive (brown; Figure 6C and D). Conversely, all DEGs from either Ctrl_CE or MetR_RT correlated less with each other (R²=0.29) and were found to interact in an antagonistic fashion (Figure 6—figure supplement 1B, C). This suggests a complex gene-regulatory relationship between MetR feeding and CE in eWAT. On the one hand, MetR_RT and Ctrl_CE additively regulate genes that only become significant under MetR_CE; on the other hand, DEGs unique to MetR_RT or Ctrl_CE alone are excluded from the combined MetR_CE response in an antagonistic fashion. GSEA analysis revealed that significant positive enrichment in immune and signaling pathways such as Rap1 signaling, MAPK signaling, and cGMP-PKG signaling (Maff, Ptchd4, and Arhgef37) under the combined treatment of MetR_CE (Figure 6E and F). Here, interaction between MetR_RT and Ctrl_CE enrichment scores appeared additive in nature (non-significant enrichment for interaction DEGs). Intriguingly, negatively enriched pathways under MetR_CE were largely associated with fatty acid metabolism and oxidative phosphorylation (Figure 6G). While MetR_RT alone again did not result in significant enrichment here, Ctrl_CE often resulted in opposing (significant or non-significant) enrichment for the mentioned pathways. These enrichments match with repressed genes that are regulated in an additive manner between MetR_RT and Ctrl_CE (Cyp2e1, Cyp4f17, Ehhadh, Slc16a1, Serpina3b; Figure 6H) and genes that are regulated in an antagonistic manner (Acly, Elovl6, Angptl8, Acss2, Me1, Fasn, Scd2, Acaca, Slc25a1; Figure 6—figure supplement 1F, I). Overall, these results highlight a complex relationship between both thermogenic stimuli in eWAT comprising both additive but also antagonistic regulation at the gene and pathway level.

Discussion

In this work, we compared the systemic effects of two environmental stimuli of energy expenditure, dietary MetR and CE, in a 2x2 design, in order to assess whether both stimuli produce additive or synergistic systemic energy metabolism and transcriptional adaptations in key metabolic tissues. Our results indicate that although dietary MetR increased EE at RT and shifted substrate utilization towards lipid oxidation, CE constituted a stronger stimulus for EE and masked diet-dependent differences. Combining CE and MetR resulted in mild reductions in body weight and glucose levels and additive reductions in circulating triglyceride levels and liver mass. Notwithstanding the threshold effects seen in the activation of FGF21 and β-hydroxybutyrate (βOHB), markers for energy homeostasis and hepatic fatty acid oxidation, respectively, these results suggest that there is potential in combining both stimuli for the correction of hyperglycemia, hypertriglyceridemia, and excess body weight in disease settings.

To gain further insights into how MetR and CE converge on the transcriptional level, we performed RNA-seq on liver, iBAT, iWAT, and eWAT. We observed a high degree of depot-specificity in the transcriptional responses to MetR, CE, and the combination, both at the gene and pathway level. In the liver, CE dominated gene induction while MetR and CE cooperatively repressed genes. The combination of both factors regulated pathways involved in energy and glucose handling (glucagon signaling, TCA cycle, AMPK signaling, amino acid and carbon metabolism), potentially explaining the stepwise reduction in circulating glucose and triglyceride levels. In iBAT, CE dominated the transcriptional response, with limited contribution from MetR, and resulted in the enrichment of thermogenic and lipid-oxidation programs (thermogenesis, fatty acid metabolism). In iWAT, MetR and CE acted largely additively with high concordance, enhancing fatty-acid degradation, PPAR signaling, thermogenesis, and TCA cycle pathways. The transcriptional responses in eWAT overall were limited, producing some additional, yet mostly antagonistic effects. Whether the downregulation of fatty acid metabolism pathways is linked to the relative increase of eWAT mass under MetR_CE requires further investigation.

An intriguing finding of potential translational value was the cold-like transcriptional response elicited by MetR feeding in iWAT. Amidst the emerging appreciation that the amount and metabolic activity of ‘classical’ human brown adipocytes might not contribute to overall EE and metabolic regulation in humans (Becher et al., 2021), other fat depots such as ‘beige’ subcutaneous adipose tissue (scWAT, i.e. the functional human equivalent to murine iWAT) are moving into the limelight of fundamental research and therapeutic pursuits (Sun et al., 2021). This is spurred by the appreciation that human, just like rodent, white fat exerts important catabolic and endocrine functions (Blondin, 2023), begetting the seminal question about the nature of physiological, pharmacological, dietary, and lifestyle-associated interventions that will help achieve an increase in energy dissipation via scWAT. Here, the unsolved task is to prompt uncoupling protein 1 (UCP1)-dependent and UCP1-independent biochemical processes such as Ca²⁺/creatine cycling (Rahbani et al., 2021; Ikeda et al., 2017), or by inducing an energy-consuming combination of triacylglycerol breakdown and fatty acid re-esterification that occurs after sympathomimetic administration (Blondin et al., 2020) selectively in beige fat. Intriguingly, a recent report demonstrated that oral supplementation of high-fat diet-fed mice with a nitroalkene derivative of salicylate (SANA) can induce creatine cycling in rodent iWAT independent of UCP1, demonstrating that the activation of iWAT thermogenesis and mitochondrial respiration is exploitable using pharmacology. Noteworthy, a Phase 1 study demonstrated that human volunteers receiving SANA exhibited a modest degree of weight loss, suggesting that strategies for iWAT mobilization in rodents might translate to (obese) patients (Cal et al., 2025). To date, orthogonal nutritional approaches for iWAT activation in mice rely on negative energy balances and include calorie restriction paradigms such as intermittent fasting (Li et al., 2017), yet conclusive data on their effect in human scWAT remains limited (Roth et al., 2021; Plummer and Johnson, 2022) and might incur detrimental processes such as the unwanted mobilization of lean (muscle and bone) mass. The selective omission or removal of methionine and cysteine has proven to be efficient in activating EE via iWAT thermogenesis and involves both humoral (FGF21 secretion) and sympathetic nervous system-dependent, UCP1-independent responses (Lee et al., 2025; Varghese et al., 2025; Ghosh et al., 2017). Future studies with extended durations should clarify whether run-in dietary priming, for example via dietary MetR, can enhance adaptive thermogenic responses during prolonged cold exposure, ideally monitored by continuous measurement of core body temperature and more in-depth molecular analysis of the browning phenotype in iWAT.

An intriguing opportunity for dietary MetR might lie in maximizing treatment effects against more clinical conditions ranging from cancer (Gao et al., 2019; Sanderson et al., 2019) to obesity. Here, drugs that mimic the action of the gut hormone glucagon-like peptide-1 (GLP1), that is so-called GLP1 receptor agonists (GLP1Ra) such as semaglutide, have radically transformed obesity treatment and reduce body weight via reducing appetite (Müller et al., 2022). However, to date, these drugs have the unfortunate downside of unfavorably reducing EE (Löffler et al., 2021). As lead-in calorie restriction allows mice to maintain EE and thus enhance the degree of achievable weight loss (Petersen et al., 2024), nutritional approaches might help to break the efficacy plateau that continues to plague GLP1Ras, collectively highlighting the notion that dietary interventions such as selective amino acid restriction prior or during pharmacological treatment deserve further experimental exploration and clinical testing.

Classically, DIT is synonymous with the thermic effect of food (TEF) and describes the post-prandial rise in energy expenditure attributable to digestion, absorption, and metabolic processing of nutrients (Calcagno et al., 2019). This canonical, meal-linked DIT is transient and scales with meal energy composition (Swaminathan et al., 1985). In contrast, dietary MetR elevates EE independent of acute feeding by engaging an endocrine-neuronal axis, where hepatic amino acid sensing activates the integrated stress response, increases FGF21 expression and circulating levels, and drives thermogenesis in WAT/BAT and a state of metabolic inefficiency, thereby increasing EE (Fang et al., 2022; Forney et al., 2020; Varghese et al., 2025). Related protein-restriction paradigms similarly raise EE through FGF21-mediated thermogenic programs (Laeger et al., 2014). Our and previous data demonstrate that MetR increases EE at thermoneutral-adjacent conditions and elicits depot-specific thermogenic transcriptional programs, even when benchmarked against cold exposure. We propose that MetR represents a true form of ‘diet-induced thermogenesis’ in a mechanistic sense that is a diet composition-initiated, hormone-driven thermogenic state that is distinct from the classic meal-processing TEF.

Altogether, our study provides important initial insights into depot-specific transcriptional responses to MetR, CE, and their combination, yet several limitations remain. The relatively short duration (7 days MetR, 24 hr CE) was sufficient for assessing acute transcriptional interactions but limited our ability to investigate long-term physiological adaptations or sustained metabolic remodeling. Furthermore, applying complementary omics analyses, including proteomics and metabolomics or flux analysis, would provide deeper mechanistic insights into metabolic interactions at the protein and metabolite levels.

Materials and methods

All animal experiments were performed in accordance with the Directive 2010/63/EU from the European Union and approved by the Ministry of Environment and Agriculture Denmark (Miljø- og Fødevarestyrelsen) under license no. 2018-15-0201-01544 and reported in line with ARRIVE 2.0 guidelines.

Animal husbandry and experiments

Request a detailed protocol

All experiments were performed in male mice on a C57BL/6 N (RRID:MGI:2159965) background (Taconic, Denmark). All mice were housed under a 12 hr light/dark cycle (lights on at 06:00, lights off at 18:00) in a temperature and humidity-controlled facility (22 ± 1°C) and had ad libitum access to diets and drinking water. 8/9-week-old mice were acclimatized to our animal facility on a chow diet (NIH-31, Zeigler Brothers Inc, 8% calories from fat) and housed in groups of three to four animals per cage for the habituation period. For reproducibility reasons and to avoid biases by ‘social thermogenesis’, we single-housed all mice for the duration of the experiments and concluded the experiments with mice being 16 weeks of age.

In study 1, we performed indirect calorimetry to assess physiological effects and interactions between diets and ambient temperature. For this, 15 mice were split randomly into three groups and fed cysteine-depleted diets containing either 0.8% Methionine (Control; Ctrl), 0.12% Methionine (Methionine-restricted; MetR) or 2% Methionine (Methionine-supplemented; MetS) for 6 days at 22°C and 5 days at 4°C. Diets were from Research Diets (New Brunswick, NJ, USA). Exact compositions can be retrieved under cat. no. A11051301B (MetR), A11051302B (Ctrl), and A21060801 (MetS).

In study 2, we performed detailed analyses of physiological (body and organ weights) and molecular parameters (RNA-seq, colorimetric/enzymatic assays) to investigate the molecular adaptations and interactions between diet and temperature. For this, 28 mice were split randomly into four groups and were either fed previously mentioned Ctrl or MetR diets at 22°C for 7 days or housed an additional 8th day at 4°C. Blood glucose levels were determined just prior to sacrifice in the cage using a handheld glucometer. The mice were then euthanized by carbon dioxide asphyxiation followed by cervical dislocation. Blood was collected by cardiac puncture and stored on ice for the duration of the sacrifice. Serum was collected after centrifugation by 2000 ×g for 15 min and stored at –80°C. Liver, eWAT, iWAT, and BAT were weighed and snap-frozen in liquid nitrogen and then stored at –80°C. All sacrifices were performed during the light phase between 09:00 and 11:00. Investigators were not blinded during animal handling and tissue collection. Downstream molecular and computational analyses were performed using coded sample identifiers to minimize bias. Sample sizes were selected based on prior experience with comparable metabolic and transcriptomic studies and were sufficient to detect biologically meaningful differences. No formal a priori power calculation was performed. No animals were excluded from physiological analyses.

Indirect calorimetry

Request a detailed protocol

Indirect calorimetry was conducted using the PhenoMaster NG 2.0 Home Cage System (TSE systems, Bad Homburg, Germany). Prior to the experiment, a complete calibration protocol for the gas analyzers was run according to the manufacturer’s recommendations, and the mice were weighed. Mice were housed individually and acclimated to the new environment for 3 days prior to the experiment. The machine was set to maintain 50% humidity throughout the experiment and a 12 hr light/dark cycle, with ad libitum access to food and water. During the experiment, energy expenditure, respiratory exchange ratio, food and water intake were recorded every 60 s and datapoints were filtered for outliers and then averaged per hour for the analysis. Indirect calorimetry data are presented as mean ± SEM.

RNA sequencing and analysis

Request a detailed protocol

Total RNAs from tissues were isolated using the TRI reagent (Sigma) followed by clean-up with RPE buffer (Qiagen, Germany). The quality of RNA was validated by the Agilent RNA 6000 Nano-Kit in Agilent 2100 Bioanalyzer according to the manufacturer’s protocol (Agilent Technologies, Waldbronn, Germany). Only samples with RNA integrity number (RIN)≥8.5 were used for library preparation. mRNA sequencing was performed in-house. After quality control, 500 ng RNA in a final volume of 25 μL DEPC-treated water was prepared and sample preparation was performed as described in the NEBNext Poly(A) mRNA Magnetic Isolation Module kit and NEBNext Ultra II RNA Library Prep Kit (cat no: #E7770, New England Biolabs, Ipswich, MA, USA). The amplified libraries were validated by Agilent 2100 Bioanalyzer using a DNA 1000 kit (Agilent Technologies, Inc, Santa Clara, CA, USA) and quantified by qPCR using the KaPa Library Kits (KaPa Biosystems, Wilmington, MA, USA). Hereafter, 2×50 bp paired-end sequencing was performed on Illumina Novaseq 6000. Sequencing targeted approximately 20 million paired-end reads per sample. Seven samples per group and tissue were submitted to sequencing. One sample in iWAT and iBAT was lost during the sample preparation.

RNAseq data analysis

Request a detailed protocol

Sequencing reads were aligned to the mouse reference genome mm10/GRCm38 (GENCODE vM25; RRID:SCR_014966; Mudge et al., 2025) using STAR aligner (v2.7.9a; RRID:SCR_004463; Dobin et al., 2013) with the following parameters: --outSAMunmapped Within, --outFilterType BySJout, --outSAMattributes NH HI AS NM MD, --outFilterMultimapNmax 10, --outFilterMismatchNoverReadLmax 0.04, --alignIntronMin 20, --alignIntronMax 1000000, --alignMatesGapMax 1000000, --alignSJoverhangMin 8, and --alignSJDBoverhangMin 1. Gene-level quantification was performed using featureCounts (v2.0.3; RRID:SCR_012919; Liao et al., 2014). In parallel, transcript-level quantification was performed using Salmon (v1.9.0; Patro et al., 2017) and Gencode vM25, with parameters --seqBias, --useVBOpt, and --numBootstraps 30. Transcript abundance estimates were imported and summarized to the gene level using the tximport package (Soneson et al., 2015). Quality control was summarized with MultiQC v1.19 (RRID:SCR_014982). All RNA-seq samples passed the QC. Differential gene expression analysis was performed using DESeq2 (v1.46.0; RRID:SCR_000154; Love et al., 2014) with shrinkage of fold changes using apeglm (Zhu et al., 2019). Genes with fewer than 10 total counts across all samples were excluded prior to normalization. The following interaction model was applied: ~ diet + temperature + diet:temperature, allowing detection of diet effects, temperature effects, and their statistical interaction. Specifically, the model tested: (i) the main effect of diet (MetR vs. Ctrl), (ii) the main effect of temperature (CE vs. RT), and (iii) the diet × temperature interaction (whether the effect of MetR differs between RT and CE). Genes with an adjusted p-value <0.05 and an absolute log₂FC ≥0.585 were considered significantly differentially expressed. Gene set enrichment analysis (GSEA; RRID:SCR_003199) was performed using the gseKEGG() function from the clusterProfiler package (v4.14.6; RRID:SCR_016884; Yu et al., 2012) with parameters: organism = "mmu", pvalueCutoff =1, pAdjustMethod = "BH", minGSSize = 0, seed = TRUE, and eps = 0. Quality control metrics and preprocessing approaches were done as previously described (Davidsen et al., 2024).

RNAseq outlier detection

Request a detailed protocol

Outlier detection was performed upon visual inspection of PCA plots within tissue analyses and was supported by robustPCA from the rrcov package (v1.7.7) (Todorov, 2004). In the order of Ctrl_RT, Ctrl_CE, MetR_RT, and MetR_CE, the following number of samples were excluded in the transcriptomic analysis: 0, 0, 0, and 0 (Liver); 0, 1, 0, and 0 (iBAT); 0, 0, 0, and 0 (iWAT); and 3, 0, 1, and 0 (eWAT).

Classification of transcriptional interaction

Request a detailed protocol

Differentially expressed genes were classified according to their regulation across single and combined stimuli using a custom R script adapted from Pantaleón García et al., 2022. The four relevant contrasts were defined as Ctrl_CE vs Ctrl_RT (A), MetR_RT vs Ctrl_RT (B), MetR_CE vs Ctrl_RT (AB), and the interaction term (A×B). Only genes significant (padj <0.05 and |FC| > 1.5) in at least one comparison were included. For each gene, the direction of regulation (Up, Down, Unchanged) was assigned for every contrast. Genes were then first classified as cooperative when both single exposures (A and B) changed in the same direction or remained unchanged, or as competitive when they differed (e.g. one Up, one Down, or only one significant). Based on these combinations, genes were grouped into distinct regulation categories: co-dependent (significant only in AB but not in A or B), concordant (both A and B regulated in the same direction and AB responded similarly), discordant (A and B regulated in the same direction but AB changed oppositely), Ctrl_CE- or MetR_RT-dominant (competitive single exposures where AB followed A or B), and masked (Figure 3D). Masked genes were defined as significant in one or both single exposures but not in AB: Ctrl_CE-masked (significant only in A), MetR_RT-masked (significant only in B), or mutually masked (significant in both A and B but lost in AB, indicating opposing regulation under the combined stimulus; Figure 3—figure supplement 1C). To evaluate synergy, an expected additive effect was computed as the sum of log₂ fold changes from the single exposures (expected{A+B} = log₂FC{A} + log₂FC{B}). The synergy score was calculated as log₂ of the absolute ratio between the observed combined effect (log₂FC{AB}) and the expected additive effect, that is log₂|log₂FC{AB}/expected{A+B}|. When this expected additive effect was approximately zero, synergy was considered undefined for that gene. Genes without significant interaction effects were classified as ‘additive’. Genes with significant interaction effects were further assigned as ‘synergistic’ (synergy score >0) or ‘antagonistic’ (synergy score <0), reflecting whether the observed combined effect surpassed or fell short of additivity. The regulation patterns are exemplified in Figure 3D, Figure 3—figure supplement 1C.

Serum measurements

Request a detailed protocol

Serum NEFAs were determined using the NEFA-HR kit (cat. no. 434–91795 and 436–91995; FUJIFILM Wako Chemicals Europe GmbH). Serum triglycerides were determined using the LabAssay Triglyceride kit (cat no. 291–94501; FUJIFILM Wako Chemicals Europe GmbH). Serum β-hydroxybutyrate levels were determined using the β-HBA kit (cat. no. 2940; Instruchemie, Delfzijl, the Netherlands). Serum FGF21 and IL-6 levels were determined using the Luminex Multiplex platform (R&D Systems).

Statistics

Exact n values and statistical tests are reported in the figure legends. Data are presented as mean ± SEM unless stated otherwise. Statistical analysis of the transcriptomics data was performed as described in the previous paragraph. Serum, body, and organ weight data were analyzed using GraphPad Prism (v. 10.4.1; RRID:SCR_002798). Prior to the analysis, the assumptions of normality and homogeneity of variance were assessed. These assumptions were met in all presented parameters. Plasma parameters, absolute and relative organ weights, and bodyweight change (%) were analyzed using an unpaired one-way ANOVA with Tukey’s HSD test to correct for multiple comparisons. Statistical differences between the groups were depicted in a compact letter display, where groups sharing the same letter are not significantly different. Bodyweight data (start/finished) and average daily locomotor activity and food and water intake (RT/CE) were analyzed using a paired two-way ANOVA with Tukey’s HSD test to correct for multiple comparisons. Statistical differences between the groups were depicted as * p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001. Energy expenditure, food and water intake data from the indirect calorimetry experiment was analyzed by ANCOVA with body weight or locomotor activity as covariate. The ANCOVA analysis was done pairwise using the regression tool by NIDDK Mouse Metabolic Phenotyping Centers (MMPC, https://www.mmpc.org).

Data availability

The RNA-seq data generated in this study have been deposited in the NCBI Gene Expression Omnibus (GEO) under accession number GSE310044. The dataset includes raw FASTQ files, processed count matrices, and metadata for all experimental conditions. In addition lists of all DEGs and GSEA analysis from each tissue were provided in the supplementary files. Custom scripts and processed data required to reproduce the transcriptomic analyses and figures are publicly available via Zenodo https://doi.org/10.5281/zenodo.19024718.

The following data sets were generated

(2025) NCBI Gene Expression Omnibus
ID GSE310044. Dietary sulfur amino acid restriction elicits a cold-like transcriptional response in inguinal but not epididymal white adipose tissue of male mice.

https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE310044
1. Ruppert P
(2026) Zenodo
MetRvsCE-transcriptomics.

https://doi.org/10.5281/zenodo.19024718

References

1. Babygirija R
2. Lamming DW
(2021) The regulation of healthspan and lifespan by dietary amino acids
Translational Medicine of Aging 5:17–30.

https://doi.org/10.1016/j.tma.2021.05.001
- PubMed
- Google Scholar
1. Bal NC
2. Maurya SK
3. Pani S
4. Sethy C
5. Banerjee A
6. Das S
7. Patnaik S
8. Kundu CN
(2017) Mild cold induced thermogenesis: are BAT and skeletal muscle synergistic partners?
Bioscience Reports 37:BSR20171087.

https://doi.org/10.1042/BSR20171087
- PubMed
- Google Scholar
1. Bartelt A
2. Bruns OT
3. Reimer R
4. Hohenberg H
5. Ittrich H
6. Peldschus K
7. Kaul MG
8. Tromsdorf UI
9. Weller H
10. Waurisch C
11. Eychmüller A
12. Gordts PLSM
13. Rinninger F
14. Bruegelmann K
15. Freund B
16. Nielsen P
17. Merkel M
18. Heeren J
(2011) Brown adipose tissue activity controls triglyceride clearance
Nature Medicine 17:200–205.

https://doi.org/10.1038/nm.2297
- PubMed
- Google Scholar
1. Becher T
2. Palanisamy S
3. Kramer DJ
4. Eljalby M
5. Marx SJ
6. Wibmer AG
7. Butler SD
8. Jiang CS
9. Vaughan R
10. Schöder H
11. Mark A
12. Cohen P
(2021) Brown adipose tissue is associated with cardiometabolic health
Nature Medicine 27:58–65.

https://doi.org/10.1038/s41591-020-1126-7
- PubMed
- Google Scholar
1. Blondin DP
2. Nielsen S
3. Kuipers EN
4. Severinsen MC
5. Jensen VH
6. Miard S
7. Jespersen NZ
8. Kooijman S
9. Boon MR
10. Fortin M
11. Phoenix S
12. Frisch F
13. Guérin B
14. Turcotte ÉE
15. Haman F
16. Richard D
17. Picard F
18. Rensen PCN
19. Scheele C
20. Carpentier AC
(2020) Human brown adipocyte thermogenesis is driven by β2-AR stimulation
Cell Metabolism 32:287–300.

https://doi.org/10.1016/j.cmet.2020.07.005
- PubMed
- Google Scholar
1. Blondin DP
(2023) Human thermogenic adipose tissue
Current Opinion in Genetics & Development 80:102054.

https://doi.org/10.1016/j.gde.2023.102054
- PubMed
- Google Scholar
1. Cal K
2. Leyva A
3. Rodríguez-Duarte J
4. Ruiz S
5. Santos L
6. Garat MP
7. Colella L
8. Ingold M
9. Benitez-Rosendo A
10. Pérez-Torrado V
11. Vilaseca C
12. Galliussi G
13. Ziegler L
14. Peclat TR
15. Bresque M
16. Handy RM
17. King R
18. Menezes Dos Reis L
19. Alves JM
20. Espasandín C
21. de la Sovera V
22. Breining P
23. Dapueto R
24. Lopez A
25. Thompson KL
26. Lino CA
27. França JV
28. Vieira TS
29. Rattan R
30. Agorrody G
31. DeVallance E
32. Haag J
33. Meadows E
34. Lewis SE
35. Santana Barbosa GC
36. Lai de Souza LO
37. Chichierchio MS
38. Valez V
39. Aicardo A
40. Contreras P
41. Vendelbo MH
42. Jakobsen S
43. Kamaid A
44. Porcal W
45. Calliari A
46. Verdes JM
47. Du J
48. Wang Y
49. Hollander JM
50. White TA
51. Radi R
52. Moyna G
53. Quijano C
54. O’Doherty R
55. Moraes-Vieira P
56. Giri S
57. Holloway GP
58. Festuccia WT
59. Leiria LO
60. Leonardi R
61. Mori MA
62. Camacho-Pereira J
63. Kelley EE
64. Duran R
65. López GV
66. Chini EN
67. Batthyány C
68. Escande C
(2025) A nitroalkene derivative of salicylate, SANA, induces creatine-dependent thermogenesis and promotes weight loss
Nature Metabolism 7:1550–1569.

https://doi.org/10.1038/s42255-025-01311-z
- PubMed
- Google Scholar
1. Calcagno M
2. Kahleova H
3. Alwarith J
4. Burgess NN
5. Flores RA
6. Busta ML
7. Barnard ND
(2019) The thermic effect of food: a review
Journal of the American College of Nutrition 38:547–551.

https://doi.org/10.1080/07315724.2018.1552544
- PubMed
- Google Scholar
1. Davidsen LI
2. Hagberg CE
3. Goitea V
4. Lundby SM
5. Larsen S
6. Ebbesen MF
7. Stanic N
8. Topel H
9. Kornfeld J-W
(2024) Mouse vascularized adipose spheroids: an organotypic model for thermogenic adipocytes
Frontiers in Endocrinology 15:1396965.

https://doi.org/10.3389/fendo.2024.1396965
- Google Scholar
1. Dobin A
2. Davis CA
3. Schlesinger F
4. Drenkow J
5. Zaleski C
6. Jha S
7. Batut P
8. Chaisson M
9. Gingeras TR
(2013) STAR: ultrafast universal RNA-seq aligner
Bioinformatics 29:15–21.

https://doi.org/10.1093/bioinformatics/bts635
- Google Scholar
1. Fang H
2. Stone KP
3. Wanders D
4. Forney LA
5. Gettys TW
(2022) The origins, evolution, and future of dietary methionine restriction
Annual Review of Nutrition 42:201–226.

https://doi.org/10.1146/annurev-nutr-062320-111849
- PubMed
- Google Scholar
1. Forney LA
2. Fang H
3. Sims LC
4. Stone KP
5. Vincik LY
6. Vick AM
7. Gibson AN
8. Burk DH
9. Gettys TW
(2020) Dietary methionine restriction signals to the brain through fibroblast growth factor 21 to regulate energy balance and remodeling of adipose tissue
Obesity 28:1912–1921.

https://doi.org/10.1002/oby.22919
- PubMed
- Google Scholar
1. Gao X
2. Sanderson SM
3. Dai Z
4. Reid MA
5. Cooper DE
6. Lu M
7. Richie JP Jr
8. Ciccarella A
9. Calcagnotto A
10. Mikhael PG
11. Mentch SJ
12. Liu J
13. Ables G
14. Kirsch DG
15. Hsu DS
16. Nichenametla SN
17. Locasale JW
(2019) Dietary methionine influences therapy in mouse cancer models and alters human metabolism
Nature 572:397–401.

https://doi.org/10.1038/s41586-019-1437-3
- PubMed
- Google Scholar
1. Ghosh S
2. Forney LA
3. Wanders D
4. Stone KP
5. Gettys TW
(2017) An integrative analysis of tissue-specific transcriptomic and metabolomic responses to short-term dietary methionine restriction in mice
PLOS ONE 12:e0177513.

https://doi.org/10.1371/journal.pone.0177513
- PubMed
- Google Scholar
(2018) Multiple effects of cold exposure on livers of male mice
The Journal of Endocrinology 238:91–106.

https://doi.org/10.1530/JOE-18-0076
- PubMed
- Google Scholar
1. Hao Q
2. Yadav R
3. Basse AL
4. Petersen S
5. Sonne SB
6. Rasmussen S
7. Zhu Q
8. Lu Z
9. Wang J
10. Audouze K
11. Gupta R
12. Madsen L
13. Kristiansen K
14. Hansen JB
(2015) Transcriptome profiling of brown adipose tissue during cold exposure reveals extensive regulation of glucose metabolism
American Journal of Physiology. Endocrinology and Metabolism 308:E380–E392.

https://doi.org/10.1152/ajpendo.00277.2014
- PubMed
- Google Scholar
1. Hasek BE
2. Boudreau A
3. Shin J
4. Feng D
5. Hulver M
6. Van NT
7. Laque A
8. Stewart LK
9. Stone KP
10. Wanders D
11. Ghosh S
12. Pessin JE
13. Gettys TW
(2013) Remodeling the integration of lipid metabolism between liver and adipose tissue by dietary methionine restriction in rats
Diabetes 62:3362–3372.

https://doi.org/10.2337/db13-0501
- PubMed
- Google Scholar
1. Ikeda K
2. Kang Q
3. Yoneshiro T
4. Camporez JP
5. Maki H
6. Homma M
7. Shinoda K
8. Chen Y
9. Lu X
10. Maretich P
11. Tajima K
12. Ajuwon KM
13. Soga T
14. Kajimura S
(2017) UCP1-independent signaling involving SERCA2b-mediated calcium cycling regulates beige fat thermogenesis and systemic glucose homeostasis
Nature Medicine 23:1454–1465.

https://doi.org/10.1038/nm.4429
- PubMed
- Google Scholar
1. Laeger T
2. Henagan TM
3. Albarado DC
4. Redman LM
5. Bray GA
6. Noland RC
7. Münzberg H
8. Hutson SM
9. Gettys TW
10. Schwartz MW
11. Morrison CD
(2014) FGF21 is an endocrine signal of protein restriction
The Journal of Clinical Investigation 124:3913–3922.

https://doi.org/10.1172/JCI74915
- PubMed
- Google Scholar
1. Lee AH
2. Orliaguet L
3. Youm Y-H
4. Maeda R
5. Dlugos T
6. Lei Y
7. Coman D
8. Shchukina I
9. Andhey PS
10. Smith SR
11. Ravussin E
12. Stadler K
13. Chen B
14. Artyomov MN
15. Hyder F
16. Horvath TL
17. Schneeberger M
18. Sugiura Y
19. Dixit VD
(2025) Cysteine depletion triggers adipose tissue thermogenesis and weight loss
Nature Metabolism 7:1204–1222.

https://doi.org/10.1038/s42255-025-01297-8
- PubMed
- Google Scholar
1. Li G
2. Xie C
3. Lu S
4. Nichols RG
5. Tian Y
6. Li L
7. Patel D
8. Ma Y
9. Brocker CN
10. Yan T
11. Krausz KW
12. Xiang R
13. Gavrilova O
14. Patterson AD
15. Gonzalez FJ
(2017) Intermittent fasting promotes white adipose browning and decreases obesity by shaping the gut microbiota
Cell Metabolism 26:672–685.

https://doi.org/10.1016/j.cmet.2017.08.019
- PubMed
- Google Scholar
1. Liao Y
2. Smyth GK
3. Shi W
(2014) featureCounts: an efficient general purpose program for assigning sequence reads to genomic features
Bioinformatics 30:923–930.

https://doi.org/10.1093/bioinformatics/btt656
- PubMed
- Google Scholar
1. Löffler MC
2. Betz MJ
3. Blondin DP
4. Augustin R
5. Sharma AK
6. Tseng Y-H
7. Scheele C
8. Zimdahl H
9. Mark M
10. Hennige AM
11. Wolfrum C
12. Langhans W
13. Hamilton BS
14. Neubauer H
(2021) Challenges in tackling energy expenditure as obesity therapy: From preclinical models to clinical application
Molecular Metabolism 51:101237.

https://doi.org/10.1016/j.molmet.2021.101237
- PubMed
- Google Scholar
1. Love MI
2. Huber W
3. Anders S
(2014) Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2
Genome Biology 15:550.

https://doi.org/10.1186/s13059-014-0550-8
- PubMed
- Google Scholar
1. Mudge JM
2. Carbonell-Sala S
3. Diekhans M
4. Martinez JG
5. Hunt T
6. Jungreis I
7. Loveland JE
8. Arnan C
9. Barnes I
10. Bennett R
11. Berry A
12. Bignell A
13. Cerdán-Vélez D
14. Cochran K
15. Cortés LT
16. Davidson C
17. Donaldson S
18. Dursun C
19. Fatima R
20. Hardy M
21. Hebbar P
22. Hollis Z
23. James BT
24. Jiang Y
25. Johnson R
26. Kaur G
27. Kay M
28. Mangan RJ
29. Maquedano M
30. Gómez LM
31. Mathlouthi N
32. Merritt R
33. Ni P
34. Palumbo E
35. Perteghella T
36. Pozo F
37. Raj S
38. Sisu C
39. Steed E
40. Sumathipala D
41. Suner M-M
42. Uszczynska-Ratajczak B
43. Wass E
44. Yang YT
45. Zhang D
46. Finn RD
47. Gerstein M
48. Guigó R
49. Hubbard TJP
50. Kellis M
51. Kundaje A
52. Paten B
53. Tress ML
54. Birney E
55. Martin FJ
56. Frankish A
(2025) GENCODE 2025: reference gene annotation for human and mouse
Nucleic Acids Research 53:D966–D975.

https://doi.org/10.1093/nar/gkae1078
- PubMed
- Google Scholar
(2022) Anti-obesity drug discovery: advances and challenges
Nature Reviews. Drug Discovery 21:201–223.

https://doi.org/10.1038/s41573-021-00337-8
- PubMed
- Google Scholar
1. Pantaleón García J
2. Kulkarni VV
3. Reese TC
4. Wali S
5. Wase SJ
6. Zhang J
7. Singh R
8. Caetano MS
9. Kadara H
10. Moghaddam SJ
11. Johnson FM
12. Wang J
13. Wang Y
14. Evans SE
(2022) OBIF: an omics-based interaction framework to reveal molecular drivers of synergy
NAR Genomics and Bioinformatics 4:lqac028.

https://doi.org/10.1093/nargab/lqac028
- Google Scholar
1. Patro R
2. Duggal G
3. Love MI
4. Irizarry RA
5. Kingsford C
(2017) Salmon provides fast and bias-aware quantification of transcript expression
Nature Methods 14:417–419.

https://doi.org/10.1038/nmeth.4197
- PubMed
- Google Scholar
1. Petersen J
2. Merrild C
3. Lund J
4. Holm S
5. Clemmensen C
(2024) Lead-in calorie restriction enhances the weight-lowering efficacy of incretin hormone-based pharmacotherapies in mice
Molecular Metabolism 89:102027.

https://doi.org/10.1016/j.molmet.2024.102027
- PubMed
- Google Scholar
1. Phelps NH
2. Singleton RK
3. Zhou B
4. Heap RA
5. Mishra A
6. Bennett JE
7. Paciorek CJ
8. Lhoste VP
9. Carrillo-Larco RM
10. Stevens GA
11. Rodriguez-Martinez A
12. Bixby H
13. Bentham J
14. Di Cesare M
15. Danaei G
16. Rayner AW
17. Barradas-Pires A
18. Cowan MJ
19. Savin S
20. Riley LM
21. Aguilar-Salinas CA
22. Baker JL
23. Barkat A
24. Bhutta ZA
25. Branca F
26. Caixeta RB
27. Cuschieri S
28. Farzadfar F
29. Ganapathy S
30. Ikeda N
31. Iotova V
32. Kengne AP
33. Khang YH
34. Laxmaiah A
35. Lin HH
36. Ma J
37. Mbanya JCN
38. Miranda JJ
39. Pradeepa R
40. Rodríguez-Artalejo F
41. Sorić M
42. Turley M
43. Wang L
44. Webster-Kerr K
45. Aarestrup J
46. Abarca-Gómez L
47. Abbasi-Kangevari M
48. Abdeen ZA
49. Abdrakhmanova S
50. Abdul Ghaffar S
51. Abdul Rahim HF
52. Abdurrahmonova Z
53. Abu-Rmeileh NM
54. Abubakar Garba J
55. Acosta-Cazares B
56. Adam I
57. Adamczyk M
58. Adams RJ
59. Adu-Afarwuah S
60. Aekplakorn W
61. Afsana K
62. Afzal S
63. Agbor VN
64. Agdeppa IA
65. Aghazadeh-Attari J
66. Ågren Å
67. Aguenaou H
68. Agyemang C
69. Ahmad MH
70. Ahmad NA
71. Ahmadi A
72. Ahmadi N
73. Ahmadi N
74. Ahmed I
75. Ahmed SH
76. Ahrens W
77. Aitmurzaeva G
78. Ajlouni K
79. Al-Hazzaa HM
80. Al-Hinai H
81. Al-Lahou B
82. Al-Lawati JA
83. Al-Raddadi R
84. Al Asfoor D
85. Al Hourani HM
86. Al Qaoud NM
87. Alarouj M
88. AlBuhairan F
89. AlDhukair S
90. Aldwairji MA
91. Alexius S
92. Ali MM
93. Alieva AV
94. Alkandari A
95. Alkerwi A
96. Alkhatib BM
97. Allin K
98. Alomary SA
99. Alomirah HF
100. Alshangiti AM
101. Alvarez-Pedrerol M
102. Aly E
103. Amarapurkar DN
104. Amiano Etxezarreta P
105. Amoah J
106. Amougou N
107. Amouyel P
108. Andersen LB
109. Anderssen SA
110. Androutsos O
111. Ängquist L
112. Anjana RM
113. Ansari-Moghaddam A
114. Anufrieva E
115. Aounallah-Skhiri H
116. Araújo J
117. Ariansen I
118. Aris T
119. Arku RE
120. Arlappa N
121. Aryal KK
122. Assefa N
123. Aspelund T
124. Assah FK
125. Assembekov B
126. Assunção MCF
127. Aung MS
128. Aurélio de Valois CJM
129. Auvinen J
130. Avdičová M
131. Avi S
132. Azad K
133. Azevedo A
134. Azimi-Nezhad M
135. Azizi F
136. Babu BV
137. Bacopoulou F
138. Bæksgaard Jørgensen M
139. Baharudin A
140. Bahijri S
141. Bajramovic I
142. Bakacs M
143. Balakrishna N
144. Balanova Y
145. Bamoshmoosh M
146. Banach M
147. Banegas JR
148. Baran J
149. Baran R
150. Barbagallo CM
151. Barbosa Filho V
152. Barceló A
153. Baretić M
154. Barnoya J
155. Barrera L
156. Barreto M
157. Barros AJ
158. Barros MVG
159. Bartosiewicz A
160. Basit A
161. Bastos JL
162. Bata I
163. Batieha AM
164. Batista AP
165. Batista RL
166. Battakova Z
167. Baur LA
168. Bayauli PM
169. Beaglehole R
170. Bel-Serrat S
171. Belavendra A
172. Ben Romdhane H
173. Benedek T
174. Benedics J
175. Benet M
176. Benitez Rolandi GE
177. Benzeval M
178. Bere E
179. Berger N
180. Bergh IH
181. Berhane Y
182. Berkinbayev S
183. Bernabe-Ortiz A
184. Bernotiene G
185. Berrios Carrasola X
186. Bettiol H
187. Beutel ME
188. Beybey AF
189. Bezerra J
190. Bhagyalaxmi A
191. Bharadwaj S
192. Bhargava SK
193. Bi H
194. Bi Y
195. Bia D
196. Biasch K
197. Bika Lele EC
198. Bikbov MM
199. Bista B
200. Bjelica DJ
201. Bjerregaard AA
202. Bjerregaard P
203. Bjertness E
204. Bjertness MB
205. Björkelund C
206. Bloch KV
207. Blokstra A
208. Blychfeld Magnazu M
209. Bo S
210. Bobak M
211. Boddy LM
212. Boehm BO
213. Boer JM
214. Boggia JG
215. Bogova E
216. Boissonnet CP
217. Bojesen SE
218. Bonaccio M
219. Bongard V
220. Bonilla-Vargas A
221. Bopp M
222. Borghs H
223. Botomba S
224. Bourne RR
225. Bovet P
226. Boymatova K
227. Braeckevelt L
228. Braeckman L
229. Bragt MC
230. Braithwaite T
231. Brajkovich I
232. Breckenkamp J
233. Breda J
234. Brenner H
235. Brewster LM
236. Brian GR
237. Briceño Y
238. Brinduse L
239. Bringolf-Isler B
240. Brito M
241. Brophy S
242. Brug J
243. Bruno G
244. Bugge A
245. Buoncristiano M
246. Burazeri G
247. Burns C
248. Cabrera de León A
249. Cacciottolo J
250. Cai H
251. Cama T
252. Cameron C
253. Camolas J
254. Can G
255. Cândido AC
256. Cañete F
257. Capanzana MV
258. Čapková N
259. Capuano E
260. Capuano R
261. Capuano V
262. Cardol M
263. Cardoso VC
264. Carlsson AC
265. Carmuega E
266. Carvalho J
267. Casajús JA
268. Casanueva FF
269. Casas M
270. Celikcan E
271. Censi L
272. Cervantes-Loaiza M
273. Cesar JA
274. Chamnan P
275. Chamukuttan S
276. Chan A
277. Chan Q
278. Charchar FJ
279. Charles MA
280. Chaturvedi HK
281. Chaturvedi N
282. Che Abdul Rahim N
283. Chee ML
284. Chen CJ
285. Chen F
286. Chen H
287. Chen LS
288. Chen S
289. Chen Z
290. Cheng CY
291. Cheng YJ
292. Cheraghian B
293. Chetrit A
294. Chikova-Iscener E
295. Chinapaw MJ
296. Chinnock A
297. Chiolero A
298. Chiou ST
299. Chirita-Emandi A
300. Chirlaque MD
301. Cho B
302. Christensen K
303. Christofaro DG
304. Chudek J
305. Cifkova R
306. Cilia M
307. Cinteza E
308. Cirillo M
309. Claessens F
310. Clare P
311. Clarke J
312. Clays E
313. Cohen E
314. Cojocaru CR
315. Colorado-Yohar S
316. Compañ-Gabucio LM
317. Concin H
318. Confortin SC
319. Cooper C
320. Coppinger TC
321. Corpeleijn E
322. Cortés LY
323. Costanzo S
324. Cottel D
325. Cowell C
326. Craig CL
327. Crampin AC
328. Cross AJ
329. Crujeiras AB
330. Cruz JJ
331. Csányi T
332. Csilla S
333. Cucu AM
334. Cui L
335. Cureau FV
336. Czenczek-Lewandowska E
337. D’Arrigo G
338. d’Orsi E
339. da Silva AG
340. Dacica L
341. Dahm CC
342. Dallongeville J
343. Damasceno A
344. Damsgaard CT
345. Dankner R
346. Dantoft TM
347. Dasgupta P
348. Dastgiri S
349. Dauchet L
350. Davletov K
351. de Assis Guedes de Vasconcelos F
352. de Assis MAA
353. De Backer G
354. De Bacquer D
355. De Bacquer J
356. de Bont J
357. De Curtis A
358. de Fragas Hinnig P
359. de Gaetano G
360. De Henauw S
361. De Miguel-Etayo P
362. De Neve JW
363. Duarte de Oliveira P
364. De Ridder D
365. De Ridder K
366. de Rooij SR
367. de Sá ACM
368. De Smedt D
369. Deepa M
370. Deev AD
371. DeGennaro VJ
372. Delisle H
373. Delpeuch F
374. Demarest S
375. Dennison E
376. Dereń K
377. Deschamps V
378. Devrishov RD
379. Dhimal M
380. Di Castelnuovo A
381. Dias-da-Costa JS
382. Díaz-Sánchez ME
383. Diaz A
384. Díaz Fernández P
385. Díez Ripollés MP
386. Dika Z
387. Djalalinia S
388. Djordjic V
389. Do HT
390. Dobson AJ
391. Dominguez L
392. Donati MB
393. Donfrancesco C
394. Dong G
395. Dong Y
396. Donoso SP
397. Döring A
398. Dorobantu M
399. Dorosty AR
400. Dörr M
401. Doua K
402. Dragano N
403. Drygas W
404. Du S
405. Duan JL
406. Duante CA
407. Duboz P
408. Duleva VL
409. Dulskiene V
410. Dumith SC
411. Dushpanova A
412. Dwyer T
413. Dyussupova A
414. Dzerve V
415. Dziankowska-Zaborszczyk E
416. Ebrahimi N
417. Echeverría G
418. Eddie R
419. Eftekhar E
420. Efthymiou V
421. Egbagbe EE
422. Eggertsen R
423. Eghtesad S
424. Eiben G
425. Ekelund U
426. El-Khateeb M
427. El Ammari L
428. El Ati J
429. Eldemire-Shearer D
430. Elliott P
431. Enang O
432. Endevelt R
433. Engle-Stone R
434. Erasmus RT
435. Erem C
436. Ergor G
437. Eriksen L
438. Eriksson JG
439. Escobedo-de la Peña J
440. Eslami S
441. Esmaeili A
442. Evans A
443. Evans RG
444. Faeh D
445. Fagherazzi G
446. Fakhradiyev I
447. Fakhretdinova AA
448. Fall CH
449. Faramarzi E
450. Farjam M
451. Farrugia Sant’Angelo V
452. Farzi Y
453. Fattahi MR
454. Fawwad A
455. Fawzi WW
456. Felix-Redondo FJ
457. Ferguson TS
458. Fernandes RA
459. Fernández-Bergés D
460. Ferrante D
461. Ferrao T
462. Ferrari G
463. Ferrari M
464. Ferrario MM
465. Ferreccio C
466. Ferreira HS
467. Ferrer E
468. Ferrieres J
469. Figueiró TH
470. Fijalkowska A
471. Fink G
472. Fisberg M
473. Fischer K
474. Foo LH
475. Forsner M
476. Fottrell EF
477. Fouad HM
478. Francis DK
479. do Franco MC
480. Fras Z
481. Fraser B
482. Frontera G
483. Fuchs FD
484. Fuchs SC
485. Fujiati II
486. Fujita Y
487. Fumihiko M
488. Furdela V
489. Furusawa T
490. Gabriela SA
491. Gaciong Z
492. Gafencu M
493. Galán Cuesta M
494. Galbarczyk A
495. Galcheva SV
496. Galenkamp H
497. Galeone D
498. Galfo M
499. Galvano F
500. Gao J
501. Gao P
502. Garcia-de-la-Hera M
503. García Mérida MJ
504. García Solano M
505. Gareta D
506. Garnett SP
507. Gaspoz JM
508. Gasull M
509. Gaya ACA
510. Gaya AR
511. Gazzinelli A
512. Gehring U
513. Geiger H
514. Geleijnse JM
515. George R
516. Gerdts E
517. Ghaderi E
518. Ghamari SH
519. Ghanbari A
520. Ghasemi E
521. Gheorghe-Fronea OF
522. Gialluisi A
523. Giampaoli S
524. Gianfagna F
525. Gieger C
526. Gill TK
527. Giovannelli J
528. Gironella G
529. Giwercman A
530. Gkiouras K
531. Glushkova N
532. Godara R
533. Godos J
534. Gogen S
535. Goldberg M
536. Goltzman D
537. Gómez G
538. Gómez Gómez JH
539. Gomez LF
540. Gómez SF
541. Gomula A
542. Gonçalves Cordeiro da Silva B
543. Gonçalves H
544. Gonçalves M
545. González-Alvarez AD
546. Gonzalez-Chica DA
547. González-Gil EM
548. Gonzalez-Gross M
549. González-Leon M
550. González-Rivas JP
551. González-Villalpando C
552. González-Villalpando ME
553. Gonzalez AR
554. Gottrand F
555. Graça AP
556. Grafnetter D
557. Grajda A
558. Grammatikopoulou MG
559. Gregg EW
560. Gregor RD
561. Gregório MJ
562. Grøholt EK
563. Grøntved A
564. Grosso G
565. Gruden G
566. Gu D
567. Guajardo V
568. Gualdi-Russo E
569. Guallar-Castillón P
570. Gualtieri A
571. Gudmundsson EF
572. Gudnason V
573. Guerchet M
574. Guerrero R
575. Guessous I
576. Guimaraes AL
577. Gujral UP
578. Gulliford MC
579. Gunnlaugsdottir J
580. Gunter MJ
581. Guo XH
582. Guo Y
583. Gupta PC
584. Gupta R
585. Gureje O
586. Gurinović MA
587. Gutiérrez González E
588. Gutierrez L
589. Gutzwiller F
590. Gwee X
591. Ha S
592. Hadaegh F
593. Hadjigeorgiou CA
594. Haghshenas R
595. Hakimi H
596. Halkjær J
597. Hambleton IR
598. Hamzeh B
599. Hanekom WA
600. Hange D
601. Hanif AA
602. Hantunen S
603. Hao J
604. Hardman CM
605. Hardy L
606. Hari Kumar R
607. Harmer Lassen T
608. Harooni J
609. Hashemi-Shahri SM
610. Hassapidou M
611. Hata J
612. Haugsgjerd T
613. Hayes AJ
614. He J
615. He Y
616. He Y
617. Heidinger-Felső R
618. Heier M
619. Heinen M
620. Hejgaard T
621. Hendriks ME
622. dos Henrique RS
623. Henriques A
624. Hernandez Cadena L
625. Herrala S
626. Herrera-Cuenca M
627. Herrera VM
628. Herter-Aeberli I
629. Herzig KH
630. Heshmat R
631. Heude B
632. Hill AG
633. Ho SY
634. Ho SC
635. Hobbs M
636. Höfelmann DA
637. Holdsworth M
638. Homayounfar R
639. Homs C
640. Hoogendijk E
641. Hopman WM
642. Horimoto AR
643. Hormiga CM
644. Horta BL
645. Houti L
646. Howitt C
647. Htay TT
648. Htet AS
649. Htike MMT
650. Hu Y
651. Huerta JM
652. Huhtaniemi IT
653. Huiart L
654. Huidumac Petrescu C
655. Husseini A
656. Huu CN
657. Huybrechts I
658. Hwalla N
659. Hyska J
660. Iacoviello L
661. Iakupova EM
662. Ibarluzea J
663. Ibrahim MM
664. Ibrahim Wong N
665. Igland J
666. Ijoma C
667. Ikram MA
668. Iñiguez C
669. Irazola VE
670. Ishida T
671. Isiguzo GC
672. Islam M
673. Islam SMS
674. Islek D
675. Ittermann T
676. Ivanova-Pandourska IY
677. Iwasaki M
678. Jääskeläinen T
679. Jackson RT
680. Jacobs JM
681. Jadoul M
682. Jafar T
683. Jallow B
684. James K
685. Jamil KM
686. Jamrozik K
687. Jan N
688. Jansson A
689. Janszky I
690. Janus E
691. Jarani J
692. Jarnig G
693. Jarvelin MR
694. Jasienska G
695. Jelaković A
696. Jelaković B
697. Jennings G
698. Jiang CQ
699. Jimenez RO
700. Jöckel KH
701. Joffres M
702. Jokelainen JJ
703. Jonas JB
704. Jonnagaddala J
705. Jøran Kjerpeseth L
706. Jørgensen T
707. Joshi P
708. Joshi R
709. Josipović J
710. Joukar F
711. Jóźwiak JJ
712. Judge DS
713. Juolevi A
714. Jurak G
715. Jurca Simina I
716. Juresa V
717. Kaaks R
718. Kaducu FO
719. Kadvan AL
720. Kafatos A
721. Kaj M
722. Kajantie EO
723. Kakutia N
724. Kállayová D
725. Kalmatayeva Z
726. Kalter-Leibovici O
727. Kameli Y
728. Kanala KR
729. Kannan S
730. Kapantais E
731. Karaglani E
732. Karakosta A
733. Kårhus LL
734. Karki KB
735. Karlsson O
736. Kassi Anicet A
737. Katchunga PB
738. Katibeh M
739. Katz J
740. Katzmarzyk PT
741. Kauhanen J
742. Kaur P
743. Kavousi M
744. Kazakbaeva GM
745. Kaze FF
746. Kazembe BM
747. Ke C
748. Keil U
749. Keinan Boker L
750. Keinänen-Kiukaanniemi S
751. Kelishadi R
752. Kelleher C
753. Kemper HC
754. Keramati M
755. Kerimkulova A
756. Kersting M
757. Key T
758. Khader YS
759. Khaledifar A
760. Khalili D
761. Kheiri B
762. Kheradmand M
763. Khosravi A
764. Khouw IM
765. Kiechl-Kohlendorfer U
766. Kiechl SJ
767. Kiechl S
768. Killewo J
769. Kim HC
770. Kim J
771. Kindblom JM
772. Kingston A
773. Klakk H
774. Klanarong S
775. Klanova J
776. Klimek M
777. Klimont J
778. Klumbiene J
779. Knoflach M
780. Kobel S
781. Koirala B
782. Kolle E
783. Kolo SM
784. Kolsteren P
785. König J
786. Korpelainen R
787. Korrovits P
788. Korzycka M
789. Kos J
790. Koskinen S
791. Kouda K
792. Koussoh Simone M
793. Kovács É
794. Kovacs VA
795. Kovalskys I
796. Kowlessur S
797. Koziel S
798. Kratenova J
799. Kratzer W
800. Kriaucioniene V
801. Kriemler S
802. Kristensen PL
803. Krizan H
804. Kroker-Lobos MF
805. Krokstad S
806. Kromhout D
807. Kruger HS
808. Kruger R
809. Kryst Ł
810. Kubinova R
811. Kuciene R
812. Kujala UM
813. Kujundzic E
814. Kulaga Z
815. Kulimbet M
816. Kulothungan V
817. Kumar RK
818. Kumari M
819. Kunešová M
820. Kurjata P
821. Kusuma YS
822. Kutsenko V
823. Kuulasmaa K
824. Kyobutungi C
825. La QN
826. Laamiri FZ
827. Laatikainen T
828. Labadarios D
829. Lachat C
830. Lackner KJ
831. Lai D
832. Laid Y
833. Lall L
834. Lam TH
835. Landaeta Jimenez M
836. Landais E
837. Lankila T
838. Lanska V
839. Lappas G
840. Larijani B
841. Larissa SP
842. Lateva MP
843. Latt TS
844. Laurenzi M
845. Lauria L
846. Lazo-Porras M
847. Le Coroller G
848. Le Nguyen Bao K
849. Le Port A
850. Le TD
851. Lee J
852. Lee J
853. Lee PH
854. Lehtimäki T
855. Lemogoum D
856. Leong E
857. Leskošek B
858. Leszczak J
859. Leth-Møller KB
860. Leung GM
861. Levitt NS
862. Li Y
863. Liivak M
864. Lilly CL
865. Lim C
866. Lim WY
867. Lima-Costa MF
868. Lin X
869. Lind L
870. Lingam V
871. Linkohr B
872. Linneberg A
873. Lissner L
874. Litwin M
875. Liu J
876. Liu L
877. Liu L
878. Liu X
879. Lo WC
880. Loit HM
881. Long KQ
882. Longo Abril G
883. Lopes L
884. Lopes MS
885. Lopes O
886. Lopez-Garcia E
887. Lopez T
888. Lotufo PA
889. Lozano JE
890. Lukrafka JL
891. Luksiene D
892. Lundqvist A
893. Lunet N
894. Lunogelo C
895. Lustigová M
896. Łuszczki E
897. M’Buyamba-Kabangu JR
898. Ma G
899. Ma X
900. Machado-Coelho GL
901. Machado-Rodrigues AM
902. Macia E
903. Macieira LM
904. Madar AA
905. Madraisau S
906. Madsen AL
907. Maestre GE
908. Maggi S
909. Magliano DJ
910. Magnacca S
911. Magriplis E
912. Mahasampath G
913. Maire B
914. Majer M
915. Makdisse M
916. Mäki P
917. Malekpour MR
918. Malekzadeh F
919. Malekzadeh R
920. Malhotra R
921. Mallikharjuna Rao K
922. Malta DC
923. Malyutina SK
924. Maniego LV
925. Manios Y
926. Mann JI
927. Mannix MI
928. Mansour-Ghanaei F
929. Manyanga T
930. Manzato E
931. Mapatano MA
932. Marcil A
933. Margozzini P
934. Maria-Magdalena R
935. Mariño J
936. Markaki A
937. Markey O
938. Markidou Ioannidou E
939. Marques-Vidal P
940. Marques LP
941. Marrugat J
942. Martin-Prevel Y
943. Martin R
944. Martorell R
945. Martos E
946. Maruf FA
947. Maruszczak K
948. Marventano S
949. Masala G
950. Mascarenhas LP
951. Masinaei M
952. Masoodi SR
953. Mathiesen EB
954. Mathur P
955. Matijasevich A
956. Matłosz P
957. Matsha TE
958. Matsudo V
959. Matteo G
960. Maulik PK
961. Mavrogianni C
962. Mazur A
963. McFarlane SR
964. McGarvey ST
965. McKee M
966. McLean RM
967. McLean SB
968. McNairy ML
969. McNulty BA
970. Mediene Benchekor S
971. Medzioniene J
972. Mehlig K
973. Mehrparvar AH
974. Meirhaeghe A
975. Meisfjord J
976. Meisinger C
977. Melgarejo JD
978. Melkumova M
979. Mello J
980. Méndez F
981. Mendivil CO
982. Menezes AMB
983. Menon GR
984. Mensink GB
985. Menzano MT
986. Meshram II
987. Meto DT
988. Meyer HE
989. Mi J
990. Michaelsen KF
991. Michels N
992. Mikkel K
993. Miłkowska K
994. Miller JC
995. Milushkina O
996. Minderico CS
997. Mini G
998. Miquel JF
999. Mirjalili MR
1000. Mirkopoulou D
1001. Mirrakhimov E
1002. Mišigoj-Duraković M
1003. Mistretta A
1004. Mocanu V
1005. Modesti PA
1006. Moghaddam SS
1007. Mohamed SF
1008. Mohammad K
1009. Mohammadi MR
1010. Mohammadi Z
1011. Mohammadifard N
1012. Mohammadpourhodki R
1013. Mohan V
1014. Mohanna S
1015. Mohd Yusoff MF
1016. Mohebbi I
1017. Moitry M
1018. Møllehave LT
1019. Møller NC
1020. Molnár D
1021. Momenan A
1022. Mondo CK
1023. Monroy-Valle M
1024. Montenegro Mendoza RA
1025. Monterrubio-Flores E
1026. Monyeki KDK
1027. Moon JS
1028. Moosazadeh M
1029. Mopa HT
1030. Moradpour F
1031. Moreira LB
1032. Morejon A
1033. Moreno LA
1034. Morey F
1035. Morgan K
1036. Morin SN
1037. Mortensen EL
1038. Moschonis G
1039. Moslem A
1040. Mosquera M
1041. Mossakowska M
1042. Mostafa A
1043. Mostafavi SA
1044. Mota-Pinto A
1045. Mota J
1046. Motlagh ME
1047. Motta J
1048. Moura-dos-Santos MA
1049. Movsesyan Y
1050. Mridha MK
1051. Msyamboza KP
1052. Mu TT
1053. Muc M
1054. Muca F
1055. Mugoša B
1056. Muiesan ML
1057. Müller-Nurasyid M
1058. Münzel T
1059. Mursu J
1060. Murtagh EM
1061. Musa KI
1062. Musić Milanović S
1063. Musil V
1064. Musinguzi G
1065. Muyer MT
1066. Nabipour I
1067. Nagel G
1068. Najafi F
1069. Nakamura H
1070. Nalecz H
1071. Námešná J
1072. Nang EEK
1073. Nangia VB
1074. Nankap M
1075. Narake S
1076. Narayan KV
1077. Nardone P
1078. Naseri T
1079. Nathalie M
1080. Neal WA
1081. Neelapaichit N
1082. Nejatizadeh A
1083. Nekkantti C
1084. Nelis K
1085. Nenko I
1086. Neovius M
1087. Nervi F
1088. Ng TP
1089. Nguyen CT
1090. Nguyen ND
1091. Nguyen QN
1092. Ni MY
1093. Nicolescu R
1094. Nie P
1095. Nieto-Martínez RE
1096. Nikitin YP
1097. Ning G
1098. Ninomiya T
1099. Nishi N
1100. Nishtar S
1101. Noale M
1102. Noboa OA
1103. Nogueira H
1104. Nordendahl M
1105. Nordestgaard BG
1106. Norton KI
1107. Noto D
1108. Nowak-Szczepanska N
1109. Nsour MA
1110. Nuhoğlu I
1111. Nunes B
1112. Nurk E
1113. Nuwaha F
1114. Nyirenda M
1115. O’Neill TW
1116. O’Reilly D
1117. Obreja G
1118. Ochimana C
1119. Ochoa-Avilés AM
1120. Oda E
1121. Odili AN
1122. Oh K
1123. Ohara K
1124. Ohlsson C
1125. Ohtsuka R
1126. Olafsson Ö
1127. Oldenburg B
1128. Olinto MTA
1129. Oliveira IO
1130. Omar MA
1131. Omar SM
1132. Onat A
1133. Ong SK
1134. Onland-Moret NC
1135. Ono LM
1136. Onodugo O
1137. Ordunez P
1138. Ornelas R
1139. Ortiz AP
1140. Ortiz PJ
1141. Osler M
1142. Osmond C
1143. Ostojic SM
1144. Ostovar A
1145. Otero JA
1146. Ottendahl CB
1147. Otu A
1148. Overvad K
1149. Owusu-Dabo E
1150. Oyeyemi AY
1151. Oyeyemi AL
1152. Paccaud FM
1153. Padez CP
1154. Pagkalos I
1155. Pahomova E
1156. de Paiva KM
1157. Pająk A
1158. Pajula N
1159. Palloni A
1160. Palmieri L
1161. Pan WH
1162. Panda-Jonas S
1163. Pandey A
1164. Pang Z
1165. Panza F
1166. Paoli M
1167. Papadopoulou SK
1168. Papandreou D
1169. Pareja RG
1170. Park SW
1171. Park S
1172. Parnell WR
1173. Parsaeian M
1174. Pascanu IM
1175. Pasquet P
1176. Patel ND
1177. Pattussi M
1178. Pavlyshyn H
1179. Pechlaner R
1180. Pećin I
1181. Pednekar MS
1182. Pedro JM
1183. Peer N
1184. Peixoto SV
1185. Peltonen M
1186. Pereira AC
1187. Peres MA
1188. Perez-Londoño A
1189. Pérez CM
1190. Peterkova V
1191. Peters A
1192. Petkeviciene J
1193. Petrauskiene A
1194. Petrovna Kovtun O
1195. Pettenuzzo E
1196. Peykari N
1197. Pfeiffer N
1198. Phall MC
1199. Pham ST
1200. Phiri FP
1201. Pichardo RN
1202. Pierannunzio D
1203. Pierre-Marie P
1204. Pigeot I
1205. Pikhart H
1206. Pilav A
1207. Piler P
1208. Pilotto L
1209. Pistelli F
1210. Pitakaka F
1211. Piwonska A
1212. Pizarro AN
1213. Plans-Rubió P
1214. Platonova AG
1215. Poh BK
1216. Pohlabeln H
1217. Polka NS
1218. Pop RM
1219. Popkin BM
1220. Popovic SR
1221. Porta M
1222. Posch G
1223. Poudyal A
1224. Poulimeneas D
1225. Pouraram H
1226. Pourfarzi F
1227. Pourshams A
1228. Poustchi H
1229. Price AJ
1230. Price JF
1231. Prista A
1232. Providencia R
1233. Puder JJ
1234. Pudule I
1235. Puhakka S
1236. Puiu M
1237. Punab M
1238. Qadir MS
1239. Qasrawi RF
1240. Qiao Q
1241. Qorbani M
1242. Quintana HK
1243. Quiroga-Padilla PJ
1244. Quoc Bao T
1245. Rach S
1246. Radic I
1247. Radisauskas R
1248. Rahimikazerooni S
1249. Rahman M
1250. Rahman M
1251. Raitakari O
1252. Raj M
1253. Rajabov T
1254. Rakhmatulloev S
1255. Rakovac I
1256. Ramachandra Rao S
1257. Ramachandran A
1258. Ramadan OP
1259. Ramires VV
1260. Ramirez-Zea M
1261. Ramke J
1262. Ramos E
1263. Ramos R
1264. Rampal L
1265. Rampal S
1266. Ramsay SE
1267. Rangelova LS
1268. Rarra V
1269. Rascon-Pacheco RA
1270. Rashidi MM
1271. Rech CR
1272. Redon J
1273. Reganit PFM
1274. Regecová V
1275. Renner JD
1276. Repasy JA
1277. Reuter CP
1278. Revilla L
1279. Reynolds A
1280. Rezaei N
1281. Rezaianzadeh A
1282. Rho Y
1283. Ribas-Barba L
1284. Ribeiro R
1285. Riboli E
1286. Rigo F
1287. Rigotti A
1288. Rinaldo N
1289. Rinke de Wit TF
1290. Risérus U
1291. Rito AI
1292. Ritti-Dias RM
1293. Rivera JA
1294. Roa RG
1295. Robinson L
1296. Roccaldo R
1297. Rodrigues D
1298. Rodriguez-Perez DMC
1299. Rodríguez-Villamizar LA
1300. Rodríguez AY
1301. Roggenbuck U
1302. Rohloff P
1303. Rohner F
1304. Rojas-Martinez R
1305. Rojroongwasinkul N
1306. Romaguera D
1307. Romeo EL
1308. Rosario RV
1309. Rosengren A
1310. Rouse I
1311. Rouzier V
1312. Roy JG
1313. Ruano MH
1314. Rubinstein A
1315. Rühli FJ
1316. Ruidavets JB
1317. Ruiz-Betancourt BS
1318. Ruiz-Castell M
1319. Ruiz Moreno E
1320. Rusakova IA
1321. Rusek W
1322. Russell Jonsson K
1323. Russo P
1324. Rust P
1325. Rutkowski M
1326. Saamel M
1327. Saar CG
1328. Sabanayagam C
1329. Sabbaghi H
1330. Sacchini E
1331. Sachdev HS
1332. Sadjadi A
1333. Safarpour AR
1334. Safi S
1335. Safiri S
1336. Saghi MH
1337. Saidi O
1338. Saieva C
1339. Sakata S
1340. Saki N
1341. Šalaj S
1342. Salanave B
1343. Salazar Martinez E
1344. Salhanova A
1345. Salmerón D
1346. Salomaa V
1347. Salonen JT
1348. Salvetti M
1349. Samoutian M
1350. Sánchez-Abanto J
1351. Sánchez Rodríguez I
1352. Sans S
1353. Santa-Marina L
1354. Santacruz E
1355. Santos DA
1356. Santos IS
1357. Santos LC
1358. Santos MP
1359. Santos O
1360. Santos R
1361. Santos TR
1362. Saramies JL
1363. Sardinha LB
1364. Sarrafzadegan N
1365. Sathish T
1366. Saum KU
1367. Savva S
1368. Savy M
1369. Sawada N
1370. Sbaraini M
1371. Scazufca M
1372. Schaan BD
1373. Schaffrath Rosario A
1374. Schargrodsky H
1375. Schienkiewitz A
1376. Schindler K
1377. Schipf S
1378. Schmidt B
1379. Schmidt CO
1380. Schmidt IM
1381. Schneider A
1382. Schnohr P
1383. Schöttker B
1384. Schramm S
1385. Schramm S
1386. Schröder H
1387. Schultsz C
1388. Schultz G
1389. Schulze MB
1390. Schutte AE
1391. Sebert S
1392. Sedaghattalab M
1393. Selamat R
1394. Sember V
1395. Sen A
1396. Senbanjo IO
1397. Sepanlou SG
1398. Sequera G
1399. Serra-Majem L
1400. Servais J
1401. Ševčíková Ľ
1402. Sewpaul R
1403. Shalnova S
1404. Shamah-Levy T
1405. Shamshirgaran SM
1406. Shanthirani CS
1407. Sharafkhah M
1408. Sharma SK
1409. Sharman A
1410. Shaw JE
1411. Shayanrad A
1412. Shayesteh AA
1413. Shengelia L
1414. Shi Z
1415. Shibuya K
1416. Shimizu-Furusawa H
1417. Shimony T
1418. Shiri R
1419. Shrestha N
1420. Si-Ramlee K
1421. Siani A
1422. Siantar R
1423. Sibai AM
1424. Sidossis LS
1425. Silitrari N
1426. Silva AM
1427. de Silva CM
1428. Silva DAS
1429. Silva KS
1430. Sim X
1431. Simon M
1432. Simons J
1433. Simons LA
1434. Sjöberg A
1435. Sjöström M
1436. Skoblina EV
1437. Skoblina NA
1438. Slazhnyova T
1439. Slowikowska-Hilczer J
1440. Slusarczyk P
1441. Smeeth L
1442. So HK
1443. Soares FC
1444. Sobek G
1445. Sobngwi E
1446. Sodemann M
1447. Söderberg S
1448. Soekatri MY
1449. Soemantri A
1450. Sofat R
1451. Solfrizzi V
1452. Solovieva YV
1453. Somi MH
1454. Sonestedt E
1455. Song Y
1456. Soofi S
1457. Sørensen TI
1458. Sørgjerd EP
1459. Sossa Jérome C
1460. Soto-Rojas VE
1461. Soumaré A
1462. Sousa-Poza A
1463. Sovic S
1464. Sparboe-Nilsen B
1465. Sparrenberger K
1466. Spencer PR
1467. Spinelli A
1468. Spiroski I
1469. Staessen JA
1470. Stamm H
1471. Stang A
1472. Starc G
1473. Staub K
1474. Stavreski B
1475. Steene-Johannessen J
1476. Stehle P
1477. Stein AD
1478. Steinsbekk S
1479. Stergiou GS
1480. Stessman J
1481. Stevanović R
1482. Stieber J
1483. Stöckl D
1484. Stokwiszewski J
1485. Stoyanova E
1486. Stratton G
1487. Stronks K
1488. Strufaldi MW
1489. Sturua L
1490. Suárez-Medina R
1491. Suarez-Ortegón MF
1492. Suebsamran P
1493. Sugiyama M
1494. Suka M
1495. Sulo G
1496. Sun CA
1497. Sun L
1498. Sund M
1499. Sundström J
1500. Sung YT
1501. Sunyer J
1502. Suriyawongpaisal P
1503. Sweis NWG
1504. Swinburn BA
1505. Sy RG
1506. Sylva RC
1507. Szponar L
1508. Tabone L
1509. Tai ES
1510. Takuro F
1511. Tambalis KD
1512. Tammesoo ML
1513. Tamosiunas A
1514. Tan EJ
1515. Tang X
1516. Tanrygulyyeva M
1517. Tanser F
1518. Tao Y
1519. Tarawneh MR
1520. Tarp J
1521. Tarqui-Mamani CB
1522. Taxová Braunerová R
1523. Taylor A
1524. Taylor J
1525. Tchibindat F
1526. Te Velde S
1527. Tebar WR
1528. Tell GS
1529. Tello T
1530. Tessema M
1531. Tham YC
1532. Thankappan K
1533. Theobald H
1534. Theodoridis X
1535. Thomas N
1536. Thorand B
1537. Thrift AG
1538. Tichá Ľ
1539. Timmermans EJ
1540. Tjandrarini DH
1541. Tjonneland A
1542. Tolonen HK
1543. Tolstrup JS
1544. Tomaszewski M
1545. Topbas M
1546. Topór-Mądry R
1547. Torheim LE
1548. Tornaritis MJ
1549. Torrent M
1550. Torres-Collado L
1551. Toselli S
1552. Touloumi G
1553. Traissac P
1554. Tran TTH
1555. Tremblay MS
1556. Triantafyllou A
1557. Trichopoulos D
1558. Trichopoulou A
1559. Trinh OT
1560. Trivedi A
1561. Tshepo L
1562. Tsigga M
1563. Tsintavis P
1564. Tsugane S
1565. Tuitele J
1566. Tuliakova AM
1567. Tulloch-Reid MK
1568. Tullu F
1569. Tuomainen TP
1570. Tuomilehto J
1571. Twig G
1572. Tynelius P
1573. Tzala E
1574. Tzotzas T
1575. Tzourio C
1576. Udoji N
1577. Ueda P
1578. Ugel E
1579. Ukoli FA
1580. Ulmer H
1581. Unal B
1582. Usupova Z
1583. Uusitalo HM
1584. Uysal N
1585. Vaitkeviciute J
1586. Valdivia G
1587. Vale S
1588. Valvi D
1589. van Dam RM
1590. van den Born BJ
1591. Van der Heyden J
1592. van der Schouw YT
1593. Van Herck K
1594. Van Lippevelde W
1595. Van Minh H
1596. Van Schoor NM
1597. van Valkengoed IG
1598. Vanderschueren D
1599. Vanuzzo D
1600. Varbo A
1601. Varela-Moreiras G
1602. Vargas LN
1603. Varona-Pérez P
1604. Vasan SK
1605. Vasques DG
1606. Vatasescu R
1607. Vega T
1608. Veidebaum T
1609. Velasquez-Melendez G
1610. Velika B
1611. Verloigne M
1612. Veronesi G
1613. Verschuren WM
1614. Victora CG
1615. Viegi G
1616. Viet L
1617. Vik FN
1618. Vilar M
1619. Villalpando S
1620. Vioque J
1621. Viriyautsahakul N
1622. Virtanen JK
1623. Visser M
1624. Visvikis-Siest S
1625. Viswanathan B
1626. Vladulescu M
1627. Vlasoff T
1628. Vocanec D
1629. Vollenweider P
1630. Völzke H
1631. Vourli G
1632. Voutilainen A
1633. Vrijheid M
1634. Vrijkotte TG
1635. Vuletić S
1636. Wade AN
1637. Waldhör T
1638. Walton J
1639. Wambiya EO
1640. Wan Bebakar WM
1641. Wan Mohamud WN
1642. de Wanderley Júnior RS
1643. Wang C
1644. Wang H
1645. Wang MD
1646. Wang N
1647. Wang Q
1648. Wang X
1649. Wang YX
1650. Wang YW
1651. Wannamethee SG
1652. Wareham N
1653. Wartha O
1654. Weber A
1655. Wedderkopp N
1656. Weghuber D
1657. Wei W
1658. Weres A
1659. Werner B
1660. Westbury LD
1661. Whincup PH
1662. Wichstrøm L
1663. Wickramasinghe K
1664. Widhalm K
1665. Widyahening IS
1666. Więcek A
1667. Wild PS
1668. Wilks RJ
1669. Willeit J
1670. Willeit P
1671. Williams J
1672. Wilsgaard T
1673. Wirth JP
1674. Wojtyniak B
1675. Woldeyohannes M
1676. Wolf K
1677. Wong-McClure RA
1678. Wong A
1679. Wong EB
1680. Wong JE
1681. Wong TY
1682. Woo J
1683. Woodward M
1684. Wu FC
1685. Wu HY
1686. Wu J
1687. Wu LJ
1688. Wu S
1689. Wyszyńska J
1690. Xu H
1691. Xu L
1692. Yaacob NA
1693. Yamborisut U
1694. Yan L
1695. Yan W
1696. Yang L
1697. Yang X
1698. Yang Y
1699. Yardim N
1700. Yasuharu T
1701. Yépez García M
1702. Yiallouros PK
1703. Yngve A
1704. Yoosefi M
1705. Yoshihara A
1706. Yotov Y
1707. You QS
1708. You SL
1709. Younger-Coleman NO
1710. Yu YL
1711. Yu Y
1712. Yusof SM
1713. Yusoff AF
1714. Zaccagni L
1715. Zafiropulos V
1716. Zainuddin AA
1717. Zakavi SR
1718. Zamani F
1719. Zambon S
1720. Zampelas A
1721. Zamrazilová H
1722. Zapata ME
1723. Zargar AH
1724. Zaw KK
1725. Zayed AA
1726. Zdrojewski T
1727. Żegleń M
1728. Zejglicova K
1729. Zeljkovic Vrkic T
1730. Zeng Y
1731. Zentai A
1732. Zhang B
1733. Zhang L
1734. Zhang ZY
1735. Zhao D
1736. Zhao MH
1737. Zhao W
1738. Zhecheva YV
1739. Zhen S
1740. Zheng W
1741. Zheng Y
1742. Zholdin B
1743. Zhou M
1744. Zhu D
1745. Zimmet P
1746. Zins M
1747. Zitt E
1748. Zocalo Y
1749. Zoghlami N
1750. Zuñiga Cisneros J
1751. Zuziak M
1752. Ezzati M
(2024) Worldwide trends in underweight and obesity from 1990 to 2022: a pooled analysis of 3663 population-representative studies with 222 million children, adolescents, and adults
The Lancet 403:1027–1050.

https://doi.org/10.1016/S0140-6736(23)02750-2
- Google Scholar
1. Plummer JD
2. Johnson JE
(2022) Intermittent methionine restriction reduces IGF-1 levels and produces similar healthspan benefits to continuous methionine restriction
Aging Cell 21:e13629.

https://doi.org/10.1111/acel.13629
- PubMed
- Google Scholar
1. Rahbani JF
2. Roesler A
3. Hussain MF
4. Samborska B
5. Dykstra CB
6. Tsai L
7. Jedrychowski MP
8. Vergnes L
9. Reue K
10. Spiegelman BM
11. Kazak L
(2021) Creatine kinase B controls futile creatine cycling in thermogenic fat
Nature 590:480–485.

https://doi.org/10.1038/s41586-021-03221-y
- PubMed
- Google Scholar
1. Roth CL
2. Molica F
3. Kwak BR
(2021) Browning of white adipose tissue as a therapeutic tool in the fight against atherosclerosis
Metabolites 11:319.

https://doi.org/10.3390/metabo11050319
- PubMed
- Google Scholar
1. Sanderson SM
2. Gao X
3. Dai Z
4. Locasale JW
(2019) Methionine metabolism in health and cancer: a nexus of diet and precision medicine
Nature Reviews. Cancer 19:625–637.

https://doi.org/10.1038/s41568-019-0187-8
- PubMed
- Google Scholar
1. Schlein C
2. Heeren J
(2016) Implications of thermogenic adipose tissues for metabolic health
Best Pract Res Clin Endocrinol Metab 30:487–496.

https://doi.org/10.1016/j.beem.2016.08.002
- Google Scholar
(2015) Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences
F1000Research 4:1521.

https://doi.org/10.12688/f1000research.7563.2
- PubMed
- Google Scholar
(2021) The nuanced metabolic functions of endogenous FGF21 depend on the nature of the stimulus, tissue source, and experimental model
Frontiers in Endocrinology 12:802541.

https://doi.org/10.3389/fendo.2021.802541
- Google Scholar
1. Subramanian A
2. Tamayo P
3. Mootha VK
4. Mukherjee S
5. Ebert BL
6. Gillette MA
7. Paulovich A
8. Pomeroy SL
9. Golub TR
10. Lander ES
11. Mesirov JP
(2005) Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles
PNAS 102:15545–15550.

https://doi.org/10.1073/pnas.0506580102
- PubMed
- Google Scholar
1. Sun W
2. Modica S
3. Dong H
4. Wolfrum C
(2021) Plasticity and heterogeneity of thermogenic adipose tissue
Nature Metabolism 3:751–761.

https://doi.org/10.1038/s42255-021-00417-4
- PubMed
- Google Scholar
1. Swaminathan R
2. King RF
3. Holmfield J
4. Siwek RA
5. Baker M
6. Wales JK
(1985) Thermic effect of feeding carbohydrate, fat, protein and mixed meal in lean and obese subjects
The American Journal of Clinical Nutrition 42:177–181.

https://doi.org/10.1093/ajcn/42.2.177
- PubMed
- Google Scholar
Software
1. Todorov V
(2004) rrcov: scalable robust estimators with high breakdown point
CRAN.

https://CRAN.R-project.org/package=rrcov
1. Varghese A
2. Gusarov I
3. Gamallo-Lana B
4. Dolgonos D
5. Mankan Y
6. Shamovsky I
7. Phan M
8. Jones R
9. Gomez-Jenkins M
10. White E
11. Wang R
12. Jones DR
13. Papagiannakopoulos T
14. Pacold ME
15. Mar AC
16. Littman DR
17. Nudler E
(2025) Unravelling cysteine-deficiency-associated rapid weight loss
Nature 643:776–784.

https://doi.org/10.1038/s41586-025-08996-y
- PubMed
- Google Scholar
1. Wanders D
2. Burk DH
3. Cortez CC
4. Van NT
5. Stone KP
6. Baker M
7. Mendoza T
8. Mynatt RL
9. Gettys TW
(2015) UCP1 is an essential mediator of the effects of methionine restriction on energy balance but not insulin sensitivity
FASEB Journal 29:2603–2615.

https://doi.org/10.1096/fj.14-270348
- PubMed
- Google Scholar
1. Wanders D
2. Forney LA
3. Stone KP
4. Burk DH
5. Pierse A
6. Gettys TW
(2017) FGF21 mediates the thermogenic and insulin-sensitizing effects of dietary methionine restriction but not its effects on hepatic lipid metabolism
Diabetes 66:858–867.

https://doi.org/10.2337/db16-1212
- PubMed
- Google Scholar
1. Yoneshiro T
2. Aita S
3. Matsushita M
4. Kayahara T
5. Kameya T
6. Kawai Y
7. Iwanaga T
8. Saito M
(2013) Recruited brown adipose tissue as an antiobesity agent in humans
The Journal of Clinical Investigation 123:3404–3408.

https://doi.org/10.1172/JCI67803
- PubMed
- Google Scholar
1. Yu G
2. Wang LG
3. Han Y
4. He QY
(2012) clusterProfiler: an R package for comparing biological themes among gene clusters
OMICS 16:284–287.

https://doi.org/10.1089/omi.2011.0118
- PubMed
- Google Scholar
1. Zhang Z
2. Cheng L
3. Ma J
4. Wang X
5. Zhao Y
(2022) Chronic cold exposure leads to daytime 895 preference in the circadian expression of hepatic metabolic genes
Frontiers in Physiology 13:865627.

https://doi.org/10.3389/fphys.2022.865627
- Google Scholar
1. Zhu A
2. Ibrahim JG
3. Love MI
(2019) Heavy-tailed prior distributions for sequence count data: removing the noise and preserving large differences
Bioinformatics 35:2084–2092.

https://doi.org/10.1093/bioinformatics/bty895
- PubMed
- Google Scholar

Article and author information

Author details

Philip MM Ruppert

Functional Genomics and Metabolism Unit, Department for Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark

Contribution
Conceptualization, Data curation, Software, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing – original draft, Project administration, Writing – review and editing

For correspondence
pmr96@cornell.edu

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0002-4028-8200
Aylin S Gueller

Functional Genomics and Metabolism Unit, Department for Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark

Contribution
Resources, Formal analysis, Investigation

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0009-0000-3636-2773
Marcus Skjæveland

Functional Genomics and Metabolism Unit, Department for Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark

Contribution
Formal analysis, Investigation

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0009-0009-7393-4721
Natasa Stanic

Functional Genomics and Metabolism Unit, Department for Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark

Contribution
Resources, Investigation

Competing interests
No competing interests declared
Jan-Wilhelm Kornfeld
1. Functional Genomics and Metabolism Unit, Department for Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark
2. Novo Nordisk Foundation Center for Adipocyte Signaling, University of Southern Denmark, Odense, Denmark
Contribution
Conceptualization, Resources, Supervision, Funding acquisition, Validation, Methodology, Project administration, Writing – review and editing

For correspondence
janwilhelmkornfeld@bmb.sdu.dk

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0002-6802-4442

Funding

European Molecular Biology Organization (ALTF 676-2021)

Philip MM Ruppert

Novo Nordisk Fonden (NNF21OC0070263)

Philip MM Ruppert
Jan-Wilhelm Kornfeld

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We thank all the members of the Kornfeld lab for fruitful discussions. We gratefully acknowledge the assistance of the FGM-seq team for their help with the sequencing measurements and Victor Goitea for helpful discussions around outlier detection methods of RNA-seq samples. All of the computation done for this project was performed on the UCloud interactive HPC system, which is managed by the eScience Center at the University of Southern Denmark. This work was supported by an EMBO Long-Term Fellowship (ALTF 676-2021) to PMMR and a Novo Nordisk Foundation call NNF21OC0070263 to PMMR and JWK.

Ethics

Version history

Preprint posted: August 9, 2025
Sent for peer review: August 26, 2025
Reviewed Preprint version 1: October 13, 2025
Reviewed Preprint version 2: December 4, 2025
Version of Record published: March 17, 2026

Cite all versions

You can cite all versions using the DOI https://doi.org/10.7554/eLife.108825. This DOI represents all versions, and will always resolve to the latest one.

Copyright

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.