The breeding of skeletal muscle-specific HuR-deficient (HuR^m−/−) mice and littermate controls (HuR^fl/fl) has been previously described (Mynatt et al., 2019). Briefly, mice floxed at elavl1, the gene encoding HuR, were purchased from the Jackson Laboratory (Ghosh et al., 2009) (Stock # 021431) and bred to mice expressing Cre recombinase under the control of the Mlc1f promoter (Bothe et al., 2000) (Jackson Laboratory, Stock # 024713) to delete HuR in skeletal muscle. All mice were on a C57BL/6 background. Mice were bred and group-housed at room temperature (RT; 22°C–24°C) under a 12:12-h light:dark cycle and allowed ad libitum access to food (Purina Mills 5015) and water. Mice were either 10–12 or 20–24 weeks of age when euthanized and the age of each group is listed for each dataset. The mice were euthanized by cervical dislocation, and tissues were collected, snap-frozen in liquid nitrogen, and stored at −80°C until subsequent analyses could be performed. Mixed gastrocnemius skeletal muscle was powdered and used for all assays presented, unless otherwise noted. The mice were euthanized according to the approved procedures of the Panel on Euthanasia of the American Veterinary Medical Association.

Body composition was measured using a Bruker NMR Minispec (Bruker Corporation, Billerica, MA, United States). Glucose tolerance tests and blood lactate measurements were performed after a 4-h fast as previously described (Warfel et al., 2017). Briefly, at 10–12 weeks of age after measuring baseline blood glucose and lactate levels via the tail vein using either a lactate plus meter (Nova Biomedical) or a blood glucose monitoring system (OneTouch Ultra 2), mice received a 0.2-mL intraperitoneal injection of 20% D-glucose (40 mg glucose per mouse; 1.5 g/kg for males and 2 g/kg for females), and blood glucose levels were subsequently monitored at 20 min, 40 min, and 60 min post-injection. Behavioral and indirect calorimetry studies were done in a 16-chamber Promethion system (Sable Systems International, Las Vegas, NV, United States) on the mice at 20–22 weeks of age. For these experiments, the mice were single-housed at RT (22°C–24°C) under a 12:12-h light:dark cycle and allowed ad libitum access to food and water. Gastrocnemius muscle was used for the measurement of the total glycogen content in HuR^m−/− and control mice. A glycogen assay kit (Abcam, ab65620) was used for glycogen measurement following the manufacturer’s instructions.

RNA was extracted from 20 to 30 mg of powdered mouse tissue using TRIzol (Thermo Fisher Scientific, Waltham, MA, United States), as previously described (Lam et al., 2014). Briefly, samples were homogenized in 300 μL TRIzol and allowed to sit at RT for 5 min, and then, 0.2 mL of chloroform was added. The samples were shaken vigorously for 15 s and allowed to sit at RT for 2–3 min before they were centrifuged (12,000 × g; 15 min; 4°C) to induce phase separation. Approximately 150 µL of the upper aqueous supernatant containing RNA was transferred to a new microcentrifuge tube to which 150 µL of 70% ethanol was added, and the samples were vortexed. RNA was then isolated using an RNeasy kit (QIAGEN, Valencia, CA, United States) with DNAse treatment, as per the manufacturer’s instructions. RNA content and quality (260/280 ratio range 1.9–2.1) were assessed using NanoDrop 1000.

Whole gastrocnemius muscle of HuR^m−/− and control mice (N = 5–7 per group) were used for total RNA extraction, as described in Section 2.3. RNA-seq libraries were constructed using Illumina’s TruSeq Stranded Total RNA Library Prep Kit with Ribo-Zero. RNA was sequenced on the Illumina NextSeq 500 using the High Output v2 Kit and paired-end sequencing forward and reverse reads (2 × 75 bp) with 75 million reads/sample. Gene-level aggregated raw counts were normalized via the TMM algorithm in edgeR (Robinson et al., 2010), and subsequent differential gene expression analysis was conducted via limma (Ritchie et al., 2015). Significantly regulated genes were defined as genes with an adjusted p-value of differential gene expression (false discovery rate or FDR) ≤ 0.05. Biological pathway enrichment was evaluated via Gene Set Enrichment Analysis (GSEA) (Mootha et al., 2003) with pathways drawn from the Kyoto Encyclopedia of Genes and Genomes (KEGG), available from Molecular Signatures Database, MSigDB (Liberzon et al., 2011), from WikiPathways (Agrawal et al., 2024), or Gene Ontology and Biological Pathways (GOBP) (Nguyen et al., 2015). For pathway analysis, pathways with an adjusted enrichment p-value (FDR) ≤ 0.1 were considered to be significantly regulated, following established methods (Griss et al., 2020; Reimand et al., 2019).

Total RNA from tissues was isolated as described above. cDNA was then synthesized using an iScript cDNA synthesis kit and used for qRT-PCR with the SYBR Green system (Bio-Rad, Hercules, CA, United States). Analysis was conducted using the Norma-gene macro, as previously described (Heckmann et al., 2011; O’Connell et al., 2017). The mouse ppib transcript was included as an additional reference gene within the Norma-gene macro to provide additional power. Primer details are given in Supplementary Table S1.

Mixed gastrocnemius muscle homogenates were prepared as previously described [14]. Fatty acid oxidation was measured as the liberation of ¹⁴CO₂ from [1–¹⁴C]-palmitate (ARC, St. Louis MO). Oxidation of palmitate at a final concentration of 200 μM unlabeled and 0.625 nCi/μL labeled was measured over the course of 30 min. Homogenates were incubated with or without 1 mM unlabeled pyruvate to measure the inhibition of palmitate oxidation. Pyruvate oxidation was measured using identical procedures, with the exception of using a final concentration of 1 mM pyruvate unlabeled and 0.625 nCi/μL [3–¹⁴C]-pyruvate (ARC, St. Louis MO) ± unlabeled palmitate (200 µM) to assess substrate competition. All reactions were conducted using media (pH 7.4) consisting of 100 mM sucrose, 60 mM EDTA, 10 mM Tris HCl, 10 mM K₂HPO₄, 80 mM KCl, 1 mM MgCl₂ 6H₂O, 1 mM L-carnitine, 0.05 mM malate, 1 mM DTT, 0.05 mM nicotinamide adenine dinucleotide, 2 mM ATP, and 0.05 mM CoA.

Using previously described procedures (Hom et al., 1984) with modifications, at 10–12 weeks of age, HuR^m−/− and control mice were fasted for 4 h and injected with either 200 µL of 20% glucose supplemented with 25 µL of 1 mCi/mL deoxy-D-glucose, 2-[1–³H] (ARC), or oral gavaged with 200 μL of olive oil supplemented with 125 μL 0.1 mCi/mL S-2-bromopalmitic acid [1–¹⁴C] (ARC). For the preparation of the palmitic acid solution, the stock solution was primarily ethanol, which was evaporated with compressed air prior to the addition of olive oil. Glucose-injected mice were euthanized after 40 min and palmitate-gavaged mice were euthanized after 3 h. Tissues were collected following euthanasia, and up to 75 mg of each tissue was placed in 1.5 mL of hexadecyltrimethylammonium bromide (Sigma) containing solubilization buffer and incubated for 6 h in a 65-C water bath with gentle shaking. Then, 10 mL of scintillation cocktail was added to each vial, and the vials were read for total ³H and ¹⁴C counts, respectively, in a scintillation counter.

Animal studies were conducted at the Pennington Biomedical Research Center’s AALAC-approved facility. All experiments were in compliance with the NIH Guide for the Care and Use of Laboratory Animals and approved by the Pennington Biomedical Research Center Institutional Animal Care and Use Committee under PBRC IACUC Protocol #1049.

Data are expressed as the mean ± SEM. For single-variable comparisons between control HuR^m−/− mice or between males and females, GraphPad Prism software was used to determine significant differences with paired, equal-variance two-tailed Student's t-tests, where normality was established using the D’Agostino–Pearson normality test. For multiple comparisons between genotypes and sex groups, GraphPad Prism software was used to determine significant differences with two-way ANOVA using Tukey’s post hoc analysis. For either test, p < 0.05 was considered significant.

We previously reported that 20- to 24-week-old HuR^m−/− male mice have increased fat mass and impaired glucose clearance (Mynatt et al., 2019; Stone et al., 2021). As adiposity is strongly linked to glucose intolerance (Goodpaster and Sparks, 2017; Goodpaster and Kelley, 2002), we sought to determine whether glucose intolerance preceded the development of adiposity. In the present study, 10- to 12-week-old HuR^m−/− male mice exhibited similar levels of body weight and fat mass as age-matched controls (Figure 1A). Interestingly, in the absence of differences in adiposity, young HuR^m−/− male mice exhibited improved glucose tolerance (Figure 1C). In addition, our previous work reported sex-specific differences in HuR^m−/− mice. Specifically, while 20- to 24-week-old female HuR^m−/− mice exhibited similar increases in adiposity as HuR^m−/− male mice, unlike their male counterparts, females had improved glucose clearance rates (Stone et al., 2021). Within the present study, we tested the link between adiposity and glucose tolerance in 10- to 12-week-old mice, which showed similar body weight and fat mass between HuR^m−/− vs. floxed controls (Figure 1B). Surprisingly, the results herein suggest that in the absence of expansion of adiposity, female HuR^M−/− mice do not exhibit improvements in glucose tolerance (Figure 1D).

To determine where glucose clearance may be enhanced in male HuR^m−/− mice, we utilized these same 10- to 12-week-old mice to monitor glucose uptake following injection with 1,2-³H, 2-deoxy-D-glucose (2DG). Male HuR^m−/− mice show a significant increase in glucose uptake in skeletal muscle when all tissues are averaged together as a whole, and this increase is especially localized to tissues with a significant proportion of type II skeletal muscle fibers such as gastrocnemius and extensor digitorum longus (EDL) (Figures 2A, C). Females on the other hand show no increases in uptake in any measured tissues but have a slightly decreased clearance in adipose tissue (Figures 2B, D). The only other tissue where significant differences in glucose uptake were noted was brown adipose tissue (BAT), where glucose uptake is significantly higher in HuR^m−/− males than in controls. Given the common origin of BAT and skeletal muscle as Myf5⁺ cells (Jung et al., 2019), we verified that mRNA of elavl1, the gene encoding HuR, was not expressed at a lower level in BAT of HuR^m−/− animals as it was it gastrocnemius compared to controls (Supplementary Figure S1). Although the result of increased glucose uptake in BAT in the absence of skeletal muscle HuR is intriguing and warrants further investigation, we did not perform further analyses of this phenomenon.

In addition to monitoring glucose uptake, we also monitored tissue uptake of 1-¹⁴C-S-2-bromopalmitate (2BP) to determine whether baseline clearances in dietary fat contributed to enhanced fat mass gain. HuR^m−/− males exhibited significantly increased uptake of 2BP only in the liver (Figures 2E, G), whereas no differences in lipid uptake were observed between genotypes in female mice (Figures 2F, H).

A second cohort of mice of 20 weeks of age (WOA) was studied for better comparison to previous studies (Mynatt et al., 2019; Stone et al., 2021). Mice from each genotype controlled for weight were used for transcriptomics analyses conducted on total RNA isolated from skeletal muscle of HuR^m−/− and control mice. We utilized gastrocnemius muscle in order to represent a sample known to contain a relatively equal number of type 1 and type 2 muscle fibers (Schiaffino and Reggiani, 2011), which also showed enhanced glucose clearance during uptake assays (Figure 2A). Reporting only genes with an FDR < 0.05, we found that the total number of differentially expressed genes between male HuR^m−/− and control samples was 384, whereas female mice had 315 differentially expressed genes (Figure 3A). Differences between male and female skeletal muscle were much more pronounced, with 2,974 genes being differentially expressed between control male and female samples. Remarkably, the removal of HuR decreased this number by 50% to 1,483 differentially expressed genes.

Scatterplots were generated with genes at FDR<0.05 by comparing Log₂ (fold change) between all groups. These comparisons were of genes that were differentially expressed in either the male HuR^fl/fl vs. male HuR^m−/− comparison (left panel, Figure 3B), female HuR^fl/fl vs. female HuR^m−/− comparison (right panel, Figure 3B), or differentially expressed in both comparisons (middle panel, Figure 3B). These plots reveal that among both common and uniquely differentially expressed genes between groups, the direction of change in expression in HuR^m−/− mice compared to controls is largely consistent regardless of sex. When similar scatterplots are created comparing sexes within a given genotype (Figure 3C), the direction of change in female mice tends to be consistent compared to males regardless of the genotype. Collectively, these results demonstrate that the removal of HuR results in few changes in gene expression relative to controls but greatly diminishes transcriptomic differences between male and female gastrocnemius muscle.

Through pathway enrichment analysis via GSEA, we determined pathways that were significantly different in group comparisons with FDR ≤ 0.1. Of the top 60 pathways identified (Supplementary Figure S2A), the Ppar signaling pathway was number 3, and the glycolysis/gluconeogenesis pathway was number 7. The Ppar signaling pathway is a known regulator of both fat metabolism and lipid biosynthesis (Christofides et al., 2021), and we have previously reported differences in expression levels of genes within this pathway between male HuR^m−/− and control male mice (Stone et al., 2021). The differential expression in Ppar signaling pathway genes thus likely relates to metabolic changes within skeletal muscle lacking HuR (Supplementary Figure S3).

Further analysis of differences in the KEGG pathway termed “glycolysis/gluconeogenesis” between groups (Figures 4A–D) indicated differences that in general followed the same trend as was found in the Ppar signaling pathway. Significant enrichment was found in female controls relative to male controls with FDR = 0.08 and in female controls relative to female HuR^m−/− with FDR = 0.03. In contrast to females, male HuR^m−/− gastrocnemius muscle showed enrichment in glycolysis/gluconeogenesis pathway gene expression with FDR = 0.08. Strikingly, the pathway enrichment found in female relative to control males was not seen between HuR^m−/− males and females (FDR = 0.68).

To further confirm these results indicating that female muscle has enhanced glycolytic gene expression relative to males, which is ablated in the absence of HuR, we used qPCR to analyze the expression of glycolytic genes within the four groups (Figure 4E). Significance calculations using two-way ANOVA suggest that 3 of the 11 genes measured are significantly decreased in expression in control male muscle relative to females, whereas only 1 gene, pfk, is significantly different between males and females of the HuR^m−/− genotype, and this is due to an increase in this gene in male over female levels. Whereas no genes are significantly different between females of different genotypes, skeletal muscle HuR knockout results in increased expression of four genes in HuR^m−/− males relative to controls. Specifically, for pgk, pgm, and eno1, knockout of HuR from skeletal muscle of males renders expression levels much more similar to those in female control and HuR^m−/− muscle. These results reinforce the suggestion that the knockout of HuR from skeletal muscle results in a more similar gene expression pattern between male and female gastrocnemius, including an increased glycolytic signature in males that is more similar to that seen in female controls.

In addition to the mRNA analyses above, we used indirect calorimetry to assess metabolic parameters for HuR^m−/− mice and weight-matched controls. As with previous results (Mynatt et al., 2019; Stone et al., 2021), we found increases in respiratory exchange ratios (RERs) for male but not female HuR^m−/− mice relative to controls (Figures 5A, B). Increases in RERs in male HuR^m−/− mice are reflective of decreases in oxygen consumption that are more robust than decreases in carbon dioxide production (Figures 5C–F). Although RER can be influenced by total activity and food intake, we did not find any differences in these parameters between groups (Supplementary Figure S4).

We then tested muscle and serum for glycogen content and lactate, respectively, as additional markers of increased carbohydrate utilization. The skeletal muscle glycogen content is increased in male HuR^m−/− mice relative to controls (Figure 5G). Despite gene expression showing no difference in levels of the ldh gene, which encodes for lactate dehydrogenase (Figure 4E), serum lactate levels were also increased in male HuR^m−/− mice relative to controls (Figure 5H). Although female HuR^m−/− mice show a trend toward increases in levels of each of these markers, they are not significantly different between HuR^m−/− females and controls, or between HuR^m−/− females and HuR^m−/− males. This reinforces that the removal of HuR from skeletal muscle results in several increased similarities between male and female mouse muscle.

Our previous results have indicated that not only are mRNA transcripts of proteins involved in fat oxidation decreased in HuR^m−/− male skeletal muscle (Stone et al., 2021) but also mRNA associated with oxidative phosphorylation proteins is decreased (Mynatt et al., 2019). We therefore sought to test whether pyruvate oxidation was also decreased, which would implicate decreased mitochondrial function following glycolysis. Relative to controls, male HuR^m−/− mice show significant decreases in base oxidation of both palmitate and pyruvate (Figures 6A, B). Substrate competition designs were used for each, and the results showed that pyruvate effectively inhibited palmitate oxidation similarly in both genotypes; however, the ability of palmitate to serve as a competing substrate against pyruvate was less robust in HuR^m−/− males than in floxed controls (Figures 6A, B). Female HuR^m−/− mice show no differences in levels of maximal pyruvate or palmitate oxidation relative to controls, or in the shift toward oxidation of the alternate substrate in either assay (Figures 6C, D). These results provide further evidence that the removal of HuR from skeletal muscle results in the decreased utilization of mitochondrial metabolic substrates in a sex-specific manner.