Adipose-specific HuR deletion protects against high-fat diet-induced obesity in mice through upregulating Ucp1 expression

Abstract

Background

RNA-binding proteins (RBPs) have been proved to play essential roles in post-transcriptional regulation of genes associated with adipogenesis. However, the role of the RBP human antigen R (HuR) in the pathogenesis of obesity remains to be clarified.

Methods

Adipocyte-specific HuR knockout (HuR^−/−) and HuR floxed (HuR^f/f) mice were fed a high-fat diet (HFD), or a paired normal control diet (NC) for 16 weeks. Moreover, 8-week-old HuR^−/− or HuR^f/f mice were subjected to cold exposure or CL316,243 treatments. The mouse body weight was recorded and the histological changes in adipose tissue were examined. RNA sequencing analysis and RT-qPCR were used to identify potential target genes for HuR. The regulation of HuR on the uncoupling protein 1 (Ucp1) expression was determined using RNA immunoprecipitation (RIP), RNA pull-down, and Luciferase assays.

Results

Adipocyte-specific HuR deletion inhibited body weight gain with HFD feeding, being accompanied by less BAT whitening and more WAT browning, and up-regulated expressions of adipose thermogenic genes (Pgc-1α, Ucp1, etc.). HuR could bind to the 3’UTR of the Ucp1 mRNA, and thus downregulated its expression. In addition, although the HuR expression was not changed in obesity, there was an enhanced transfer of HuR protein from the nuclear to cytoplasm, thus impacting the expression of target genes including the Ucp1.

Conclusions

These findings indicate that adipose tissue-specific HuR deletion alleviates HFD-induced obesity by promoting adipose thermogenesis through upregulating Ucp1 expression.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12944-025-02680-w.

Background

Obesity has become a major health problem worldwide with increasing burdens of associated complications, such as dyslipidemia, type 2 diabetes, nonalcoholic fatty liver disease, respiratory and cardiovascular diseases, sleep disturbance, psychological distress, certain types of cancers, etc [1]. Being a complex and heritable disorder, obesity results from the interplay between genetic susceptibility, epigenetics, metagenomics and the environment [2, 3]. With the successful cloning of ob gene three decades ago [4] and the subsequent emergence of genome-wide association studies, numerous genes influencing adipogenesis and body phenotype (body mass index, waist-to-hip ratio, and other adiposity traits) have been identified [5, 6]. The expression of these genes, at the transcriptional level, is regulated by a variety of transcriptional factors or epigenetic modification [5–7]. Recently, several microRNAs (e.g. miR-130 and miR-22) and RNA-binding proteins (RBPs) (HuR, PSPC1, Sam68, RBM4, Ybx1, Ybx2, IGF2BP2, and KSRP) have been proved to play essential roles in posttranscriptional regulation of genes associated with adipogenesis and some RBPs are mutated or dysregulated in obesity and diabetes [8–10].

Of all the RBPs, the human antigen R (HuR), also named the embryonic lethal and abnormal vision gene (ELAVL1) and ubiquitously expressed in all human tissues, is one of the best characterized. HuR has been demonstrated to be able to specifically bind to adenine and uridine rich elements (AREs) located primarily at the 3’UTR of target mRNAs via its three evolutionarily conserved RNA recognition motifs in order to stabilize them and potentially promote their translation [11, 12]. In general, HuR’s functional activity is regulated by dynamic subcellular localization. Under normal cellular and physiologic conditions, HuR is primarily located in the nucleus, but upon exposure to intrinsic and/or extrinsic stress, it translocates to the cytoplasm and increases the stability and translational efficiency of target genes [12].

Based on almost two decades of work, it is well established that HuR is essential for many cellular biological processes including proliferation, differentiation, apoptosis, senescence and survival, carcinogenesis, inflammation and stress response, and thus is associated with a number of chronic diseases [12]. Recently, it has been reported that adipose-specific deletion of HuR significantly affects mouse body weight when fed with a high-fat diet (HFD), but the results are in controversy [13–15]. HuR has been demonstrated to function as a positive regulator of lipolysis by targeting adipose triglyceride lipase (Atgl) and insulin-induced gene 1 (Ing1) to maintain their mRNA stability, and adipose-specific HuR knockout promotes HFD-induced obesity and increases fat mass [13, 14]. Inconsistently, Adipo-HuR knockout mice display a lean phenotype compared to wild-type littermate controls, with an increase in energy expenditure [15]. Therefore, the role of HuR in the pathogenesis of obesity remains to be clarified.

In contrast to white adipose tissue (WAT), brown adipose tissue (BAT) can undergo thermogenesis and maintain homeothermy in mammals mediated by mitochondrial uncoupling protein 1 (Ucp1) [16, 17]. With the progression of obesity, BAT undergoes a process known as “whitening”, characterized by a white appearance and decreased Ucp-1 expression and activity, and beginning to express leptin [18]. Meanwhile, the expression of other biomarkers for BAT is downregulated, including Prdm16, Pgc1-α, Dio2, Cidea, etc., which are maintenance factors of the brown phenotype, or key genes as translational mediators of β-adrenergic signaling in mitochondrial biogenesis and activation of thermogenesis [16–18]. Whitening or browning of the adipose tissue is an adaptive and reversible response to environmental challenges [19]. It is worthy to mention that the rodent thermoneutral zone is between 26 °C and 34 °C, whereas most of the facilities for rats and mice are operated at a temperature of 20–24 °C, implying that the temperatures usually used to house the rodents may stimulate BAT thermogenesis [19, 20]. Nonetheless, consumption of high-fat diet (HFD) markedly reduces the cooling-evoked increases in BAT sympathetic nerve activity and BAT thermogenesis in rodents, and thus chronically reducing BAT energy expenditure and thereby contributing to high-fat obesity, and the effect of HFD in promoting adaptive thermogenesis is independent of the diet composition, within the range of 20–78% of total energy from fat [20].

Therefore, in the current study, the effects of HuR deficiency on obesity and its target genes were determined in adipose-specific HuR knockout mice with HFD (60% energy from fat) feeding, and further the distribution of HuR in adipocyte nucleus and cytoplasm were examined in HFD-induced obese mice. Finally, the regulation of HuR on expression of the target gene Ucp1 and the potential mechanisms were examined under the circumstance of cold exposure or β-adrenergic agonist treatment.

Methods

Animal studies

The HuR floxed C57 BL/6J mice (HuR^flox/flox, abbreviated as HuR^f/f) (Jackson Laboratory, stock number 021431) were crossed with adipocyte-specific adiponectin-cre C57 BL/6J mice (Jackson Laboratory, stock number 028020) to generate mice with adipocyte-specific HuR knockout (HuR^−/−) (Fig. 1a). Genotyping was performed (primers are listed in Supplementary Table 1), and the male offspring with indicated genotypes were used in the present study. C57BL/6J wild type (WT) mice were purchased from the SPF (Beijing) Biotechnology Co., Ltd. (Beijing, China).

Fig. 1.

Adipocyte-specific HuR knockout inhibits body weight gain in mice fed the high fat diet. a Targeting strategy for deletion of HuR. b mRNA levels of HuR in the adipose tissue (n = 6). c Protein levels of HuR in the adipose and non-adipose tissues (n = 3). d Changes in mouse body weight fed with the NC diet (n = 6), and (e) with the HFD diet (n = 6). f Representative images of the liver, BAT and WAT depots form the NC group and (g) the HFD group, scale bar = 1 cm. h Histological changes in BAT and WAT with hematoxylin and eosin (H&E) staining, and in the liver with the Oil Red O staining in the NC group, and (i) in the HFD group. scale bar = 50 μm. j BAT stereology - Q_A [nuclei]. Error bars represent SD; *p < 0.05, **p < 0.01 and ***p < 0.001 for HuR^−/− vs. HuR^f/f

Four- to five-week-old HuR^−/− or HuR^f/f mice were fed a high-fat diet (HFD) (34.9% fat by wt., 60% energy) (No. H10060), or a paired normal control diet (NC) (4.3% fat by wt., 10% energy) (No. H10010) (Beijing Huafukang Bioscience Co. Inc., Beijing, China) for 16 weeks. The diets were made based on the formula of the high-fat diet for DIO mice (D12492) and the paired control diet (D12450B) (Research Diets, New Brunswick, NJ, USA), sterilized with γ-irradiation 25 kGy and stored at −20 °C until use. The diet formula was shown in Supplementary Table 2. All mice were housed in room temperature (22 °C) and humidity with a 12-hour (hr) light 12-hr dark cycle and cycles of air ventilation, with free access to water and food. Eight-week-old mice were subjected to cold (4 °C) exposure for 6 h, or daily injections of CL316,243 (β-adrenergic agonist) (1 mg/kg/day) for 7 days with saline as control. Then, 12-h fasted mice were anesthetized by intraperitoneal injection of Avertin (2,2,2-tribromoethanol, T-4840-2, Sigma-Aldrich Chemie GmbH) (125 mg/kg). After the blood samples drawn by heart puncture, mice were euthanized by injection of an overdose of Avertin (500 mg/kg) and decapitation to minimize suffering. Then, the inguinal white adipose tissue (iWAT), epididymal white adipose tissue (eWAT), subscapular brown adipose tissue (BAT), and the liver were immediately dissected, removed, stored in tubes and frozen in liquid N₂, and then transferred to −80 °C until use.

All experimental protocols were approved by the Committee on the Ethics of Institute of Laboratory Animal Sciences, National Institute of Occupational Health and Poison Control of China (number EAWE-2019-03), and in accordance with the Guide for the Care and Use of Laboratory Animals in China, and the Animals (Scientific Procedures) 1986 Act (UK) (amended 2013).

Histological analysis

The inguinal, epididymal and subscapular fat were fixed in adipose tissue fixative (cat. No. G1119, Servicebio Science Tech. Co., Ltd., Wuhan, China), and the hepatic tissue was fixed with 10% neutral-buffered formalin. The fixed tissues were embedded in paraffin, and cut into 6 μm sections, and stained with the hematoxylin-eosin (HE) and the oil Red O for analysis. The BAT stereology was determined using the numerical density of nuclei per area (QA [nuclei]), which is estimated by dividing the total number of nuclei counted within the test area in mm² [21].

Real-time RT-PCR

Total RNA was extracted from the targeted tissues (epididymal, inguinal and subscapular fat) and HEK293 cells using the TRIzol Reagent (cat. No.15596-018, Invitrogen, Carlsbad, CA, USA), and then reverse transcribed to cDNA using the RT kit (cat. No. AE341-02, TransGen Biotech Co., Ltd., Beijing, China) according to the procedures provided by the manufacturer. The mRNA expression of targeted genes was measured using the real-time qPCR with a CFX96 Touch Real-Time PCR Detection System (Bio-Rad Laboratories, Inc.) and Top Green qPCR SuperMix (cat. No. AQ131-01, TransGen Biotech Co., Ltd., Beijing, China). The relative expression levels of genes were determined after normalization to the expression levels of internal reference gene RPLP0 using the 2^−△Ct method. The oligonucleotide primers for the targeted genes were listed in Supplementary Table 3.

Western blot analysis

Nuclear or cytoplasmic lysates were extracted and isolated from the targeted tissues (epididymal, inguinal and subscapular fat) using a nuclear and cytoplasmic extraction kit (cat. No. P1200-50, Applygen, Beijing, China) according to the procedures provided by the manufacturer. To prepare total protein lysates, tissues were homogenized using a dounce homogenizer with 15–20 strokes in RIPA buffer (cat. No. C1053, Applygen, Beijing, China) containing proteinase inhibitor cocktail (cat. No. P1266, Applygen, Beijing, China). Subsequently incubated the lysates on ice for 30 min, followed by centrifugation at 12,000 × g, 4℃ for 15 min. The supernatant with proteins was collected and protein concentration was measured with a BCA Protein Assay Kit (cat. No. 23225, Thermo, Waltham, MA, USA). Total protein, nuclear or cytoplasmic lysates were separated by SDS-PAGE and transferred onto polyvinylidene fluoride (PVDF) membranes (cat. No. IPVH00010, Millipore, Billerica, MA, USA). Membranes were blocked with 5% milk in TBST at room temperature for 1 h and incubated overnight with primary antibodies at 4 °C. After three washes with TBST, membranes were incubated with secondary antibodies at room temperature for 1 h. Following three more washes with TBST, blots were visualized through Pierce™ ECL substrate (cat. No. 32209, Thermo, Waltham, MA, USA) and captured using the Bio-Rad image analysis system (Bio-Rad, Hercules, CA, USA). Antibodies are listed in Supplementary Table 4.

RNA immunoprecipitation (RIP)

To prepare BAT lysates, frozen BAT was homogenized using a dounce homogenizer with 15–20 strokes in NP-40 lysis buffer (50mM HEPES pH 7.5, 150 mM KCl, 2 mM EDTA, 0.5% NP-40, 0.5 mM DTT with Protease Inhibitor Cocktail and RNase-OUT). For each RIP, 5 µg Rabbit IgG (cat. No. C1755, Applygen, Beijing, China) or HuR antibody (cat. No. 11910-1-AP-150, Proteintech, Wuhan, China) was firstly incubated with 30 µl washed Dynabeads^® Protein G in 300 µl NT2 buffer (50mM Tris-HCl pH 7.5, 150 mM NaCl, 1 mM MgCl2, 0.05% NP-40) at 4℃ for overnight with rotation. Then the antibody coupled with beads were added to 2.5 mg BAT lysates diluted in 200 µl NP-40 lysis buffer and incubated for 3 h at 4℃ with gentle rotation. Beads were washed briefly five times with IP washing buffer (50mM HEPES pH 7.5, 300 mM KCl, 0.05% NP-40, 0.5 mM DTT). At the final wash, one fifth of beads were used for protein analysis and the rest of beads were incubated with DNase I and proteinase K in NT2 buffer, then added 1 ml of Trizol for RNA extraction. Co-precipitated RNAs were isolated and analyzed by RT-PCR. The oligonucleotide sequences were designed and tested for efficiency using Primer-BLAST (Supplementary Table 2). The expression levels were normalized with those from input, and then the fold enrichments were compared with IgG control.

Luciferase reporter vectors construction and luciferase reporter assay

To construct the pGL3-Ucp1-3’UTR, the Ucp1-3’UTR fragment was amplified by PCR using following primer pairs: GCTCTAGAGCAACTTGGAGGAAGAGATA and GCTCTAGAAGATGGAATTAGCAATACTTT, then inserted between the XbaI site of the pGL3-Control vector (cat. No. E1761, Promega, Madison, Wisconsin, USA). After producing pGL3-Ucp1-3’UTR vector, we used the vector as a model, and constructed pGL3-Ucp1-3’UTR-M, the HuR-binding motif (positions 1311 to 1318 inside the B fragment) UUUUUUUU was mutated to UUUGGUUU by overlapping PCR with the primers: AACCTCTTTTAATTTGGTTTAAAGGGA and CCAAATTAAAAGAGGTTTCAAAACTCT. HEK293 cells were transfected with siRNA targeting HuR (GAGGCAAUUACCAGUUUCAUU). Twenty-four hours later, cells were transfected with each of the pGL3-derived reporters together with a pRL-CMV served as an internal control and cultured for additional 48 h. Firefly and renilla luciferase activities were measured with a double luciferase assay system (cat. No. E1960, Promega, Madison, Wisconsin, USA) following the manufacturers’ instructions. All firefly luciferase measurements were normalized to renilla luciferase measurements from the same sample.

RNA pull-down

The pGL3-Ucp1-3’UTR vector was used as a template for the PCR amplification of different fragments of Ucp1 mRNA. All 5’ primers contained the T7 promoter sequence: 5’-CCAAGCTTCTAATACGACTCACTATAGGGAGA-3’ (T7). To prepare templates for the 3’UTR (positions 1160 to 1644), and 3’UTR-A (positions 1160 to 1309), -B (positions 1271 to 1424), -C (positions 1381 to 1544), -D (positions 1481 to 1644), the following primer pairs were used: (T7)GCAACTTGGAGGAAGAGATA and AGATGGAATTAGCAATACTTT for 3’UTR; (T7)GCAACTTGGAGGAAGAGATA and TAAAAGAGGTTTCAAAACTC for A; (T7)CAAGATCATTTCCAGTAGAG and GGACTTTATATAGGTATTAT for B; (T7)AGCATTCACTAATATTTTGA and GTGTTTATCGATGTAAAGGG for C; (T7)CTGATTACAGCTCAAACTAG and AGATGGAATTAGCAATACTTT for D. For RNA pull-down assays, PCR-amplified DNA was used as templates to transcribe biotinylated RNA by using T7 RNA polymerase (cat. No. EP0111, Invitrogen, Carlsbad, CA, USA) in the presence of biotin-11-UTP (cat. No. 40033, Biotium, San Francisco, USA). One microgram of purified biotinylated transcripts was incubated with 100 µg of whole cell lysates for 30 min at room temperature. Complexes were isolated with paramagnetic streptavidin-conjugated Dynabeads (cat. No. 11206D, Invitrogen, Carlsbad, CA, USA), and the pull-down material was analyzed by western blotting.

mRNA half-life measurement

To analyze the half-lives of Ucp1-3’UTR mRNAs, HEK293 cells were transfected with siRNA or NC for 24 h and then transfected with pGL3-Ucp1-3’UTR for 12 h, the total RNAs were extracted at 0, 3, and 6 h after actinomycin D (ActD, final concentration 1 µg/mL, cat. No. J608, Amresco, ID, USA) treatment. The mRNA levels at different times were analyzed using RT-qPCR. The assays were performed in triplicates, and the results were normalized to the GAPDH mRNA levels using the 2^–△Ct method. The primer sequences were as follows: Luci-F: AGAACTGCCTGCGTGAGATT, Luci-R: AAAACCGTGATGGAATGGAA; GAPDH-F: AGCCACATCGCTCAGACAC, GAPDH-R: GCCCAATACGACCAAATCC.

RNA-sequencing (RNA-seq)

RNA-seq was performed by Aksomics Inc (Shanghai, China). In brief, the total RNA was extracted from the adipose tissue using the TRIzol Reagent kit (cat. No. 15596-018, Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instruction. RNA purity and concentration were determined by a NanoDrop ND-1000 (Thermo Fisher). Total RNA was enriched by oligo-dT magnetic beads and RNA-seq library was prepared using KAPA Stranded RNA-Seq Library Prep Kit (Illumina, San Diego, CA, USA). RNA sequencing performed using an Illumina HiSeq4000 instrument (Illumina) for 150 cycles and libraries were qualified using an Agilent 2100 Bioanalyzer (Agilent), and quantified using quantitative real-time PCR (qPCR).

RNA-seq count data were normalized to FPKM (fragment per kilobase per million) by the R package Ballgown software. Furthermore, Hierarchical clustering, volcano plots and gene ontology (GO) enrichment analysis of differentially expressed genes were performed in R, Python, or Shell environments for statistical calculation and graphical display.

Statistical analysis

All statistical analyses were conducted by SPSS 21.0. The Kolmogorov-Smirnov test was used to evaluate whether the data is normally distributed. We used the unpaired t-test for the normally distributed data and the Mann–Whitney U test for the non-normally distributed data to calculate the difference between each two groups., where p < 0.05 was considered statistically significant.

Results

Adipocyte-specific HuR knockout inhibited HFD-induced body weight gain

Real-time RT-PCR and western blotting analysis confirmed the efficient and specific deletion of HuR in iWAT, eWAT and subscapular BAT, and the normal expression of HuR in non-adipose tissues including the liver, spleen and lung (Fig. 1b and c). HuR^−/− mice were born at a normal Mendelian ratio and were indistinguishable from their HuR^f/f littermates.

With NC feeding, the HuR^−/− mice showed a similar growth curve (Fig. 1d) and food intake (Supplementary Figure S1) to the HuR^f/f littermates, with no changes in the volume of the adipose tissue and liver (Fig. 1f). The HE staining demonstrated no differences in the adipocyte size in both the eWAT and iWAT between HuR^−/− mice and HuR^f/f littermates, but less multilocular adipocytes in the BAT in HuR^−/− mice than HuR^f/f littermates (Fig. 1h), indicating less whitening. Stereology data corroborates the histological findings, showing higher Q_A [nuclei] in the BAT of HuR^−/− group compared to the BAT of HuR^f/f group, confirming less whitening (Fig. 1j). The Oil Red O staining revealed no differences in hepatic triglyceride (TG) concentrations between HuR^−/− mice and HuR^f/f littermates (Fig. 1h).

With the HFD feeding, HuR^−/− mice had a less body weight gain compared to HuR^f/f littermates, and the suppressed body weight gain in HuR^−/− mice was not in keeping with food intake (Fig. 1e, Supplementary Figure S1). The adipocyte size in both the eWAT and iWAT was smaller in HuR^−/− mice than that of HuR^f/f littermates. Still, less multilocular adipocytes and more Q_A [nuclei] in the BAT were indicated in HuR^−/− mice compared to HuR^f/f littermates, indicating still less whitening in HFD feeding (Fig. 1i and j). The oil Red O staining showed decreased lipid accumulation in the liver of HuR^−/− mice (Fig. 1i).

HuR knockout promoted browning of the adipose tissue

Analysis of the RNA-seq showed that 12,701 genes were identified in epididymal WAT samples. A total of 37 genes were transcriptionally up-regulated and 101 genes were transcriptionally down-regulated of HuR^−/− mice, compared to HuR^f/f littermates (Fig. 2a). Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis of the differentially expressed genes (DEGs) from RNA-seq data further revealed that a significant enrichment of up-regulated genes involved in thermogenesis, fatty acid metabolism, PPAR signaling pathway, chemical carcinogenesis - reactive oxygen species, fatty acid biosynthesis, fatty acid degradation and adipocytokine signaling pathway (Fig. 2b). The clustered heatmap showed that, compared to the HuR^f/f mice, the HuR^−/− mice exhibited multiple thermogenesis-related genes were upregulated (Fig. 2c).

Fig. 2.

HuR knockout increases browning in the adipose tissue. a-c Analysis of the RNA-seq in epididymal white adipose tissue (eWAT) between HuR^−/− mice and HuR^f/f mice (n = 3/group). a A volcanoplot illustrating the differentially expressed genes (DEGs). Red dots represent upregulated genes, blue dots represent downregulated genes (P < 0.05, |logFC|>1.0), and gray dots represent no changed genes, in HuR^−/− mice versus HuR^f/f mice. b Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis for upregulated genes in HuR^−/− mice versus HuR^f/f mice. c The clustered heatmap view of genes expression. d RT-qPCR analysis of genes involved in adipogenesis, lipolysis, fatty acid oxidation, and thermogenesis in brown adipose tissue (BAT), inguinal white adipose tissue (iWAT) and eWAT in mice with NC feeding (n = 8/group). e RT-qPCR analysis of genes involved in adipogenesis, lipolysis, fatty acid oxidation, and thermogenesis in BAT, iWAT and eWAT in mice with HFD feeding (n = 8/group). All data are presented as the means ± SD; *p < 0.05, **p < 0.01 and ***p < 0.001 for HuR^−/− versus HuR^f/f

Further validation of the expression of genes associated with fatty acid metabolism (hydrolysis, synthesis and oxidation) and thermogenesis were conducted, and the results showed that HuR deficiency only led to an increase in the expression of thermogenic genes (Pgc-1α, Ucp1, or Dio2) (Fig. 2d), being consistent with less whitening of BAT and more browning of iWAT in morphological change in HuR^−/− mice with NC feeding. Similar results were obtained with HFD feeding. Intriguingly, the Ucp1 mRNA expression was greatly increased in iWAT of HuR^−/− mice, as compared with HuR^f/f littermates (Fig. 2e).

HuR regulated thermogenesis under cold exposure or β-agonists treatment

Cold exposure or β-adrenergic agonists, impacting thermogenesis in the adipose tissue, were used to determine their effects on adipose morphological changes and associated gene expressions in HuR^−/− mice. As shown in Fig. 3a, b and c, cold exposure (4℃) reduced whitening in BAT and increased browning in iWAT and eWAT both in HuR^−/− and HuR^f/f littermates, with a greater impact on HuR^−/− mice particularly in iWAT. Consistently, changes in the expression of thermogenic genes were greater in HuR^−/− mice than HuR^f/f littermates (Fig. 3d, e and f). Similar changes in BAT whitening, and browning of iWAT and eWAT (Fig. 3g, h and i), and associated thermogenic genes’ expression were found with CL316,243 treatment (Fig. 3j, k and l). Thus, the thermogenic function of both BAT and iWAT was enhanced by the ablation of HuR.

Fig. 3.

Changes in thermogenesis of the adipose tissue under cold exposure or β-agonists treatment in HuR^−/− mice with NC feeding. H&E staining in (a) brown adipose tissue (BAT), (b) inguinal white adipose tissue (iWAT) and (c) epididymal white adipose tissue (eWAT) in HuR^−/− and HuR^f/f mice following 4℃ exposure for 6 h, scale bar = 50 μm. Relative mRNA levels of thermogenesis genes (d) in BAT, (e) in iWAT and (f) in eWAT from mice with 4℃ exposure. H&E staining (g) in BAT, (h) in iWAT and (i) in eWAT from HuR^−/− and HuR^f/f mice with injection of CL316,243 (1 mg/kg/day) with saline as control, scale bar = 50 μm. Relative mRNA levels of thermogenesis genes (j) in BAT, (k) in iWAT and (l) in eWAT from mice with injection of CL316,243. All data are presented as the means ± SD; n = 8 in each group; *Significantly different from the HuR^f/f-RT group or HuR^f/f-saline group (p < 0.05), # Significantly different from the HuR^f/f−4℃ group or HuR^f/f- CL316,243 group (p < 0.05)

HuR destabilizes Ucp1 mRNA

Furthermore, interactions between HuR and potential target genes were examined using RNA immunoprecipitation assay. As shown in Fig. 4a, HuR was detected in anti-HuR immunoprecipitates instead of anti-IgG immunoprecipitates, indicating validity of the experimental system. The Ucp1 mRNA was significantly enriched in the immunoprecipitation sample with the HuR antibody, compared to the control IgG antibody, with no alterations in Elvol3, Dio2 or Fabp4 mRNA. Analysis on BAT lysates indicated no Ucp1 mRNA with anti-HuR immunoprecipitates in HuR^−/− mice (Fig. 4b). To determine whether HuR can bind to mouse Ucp1 mRNA, the biotinylated fragments of Ucp1 mRNA were prepared, including the full-length 3’UTR, and the 3’UTR -A, -B, -C, and -D fragments (Fig. 4c). Following the incubation of biotinylated Ucp1 mRNA fragments with the whole cell extracts of 3T3-L1 cells, RNA protein pulldown analysis demonstrated that HuR was able to interact with the 3’UTR and the 3’UTR -B fragment but not with the 3’UTR -A, -C, and -D fragments (Fig. 4d). In addition, when the UUUUUUUU motif in the Ucp1 3’UTR was mutated to UUUGGUUU (named the 3’UTR-M fragment), the HuR interaction with the 3’UTR was abolished (Fig. 4c and e). We found that in the BAT of HuR^−/− mice, Ucp1 protein expression was greatly increased compared with HuR^f/f mice (Fig. 4f). In addition, we observed that more expression of Ucp1 protein in the iWAT of HuR^−/− mice under cold exposure compared to room temperature (Fig. 4g). The above results further demonstrate that HuR maybe regulate the expression of Ucp1 protein by regulating the stability of Ucp1 mRNA.

Fig. 4.

HuR destabilizes Ucp1 mRNA. a BAT lysates were immunoprecipitated with mouse control IgG or anti-HuR antibody. The levels of Ucp1, Elvol3, Dio2, and Fabp4 mRNA in immune complex were analyzed by Real-time qPCR (n = 3; ***p < 0.001). The levels of HuR protein in immune complex were analyzed by Western blotting. b BAT lysates from HuR^f/f and HuR^−/− mice were immunoprecipitated with anti-HuR antibody. The levels of Ucp1, Elvol3, Dio2, and Fabp4 mRNA in immune complex were analyzed by Real-time qPCR (n = 3; ***p < 0.001), the levels of HuR protein in immune complex were analyzed by Western blotting. (c) Schematic representation of the Ucp1 mRNA with the four AREs and the Ucp1 mRNA fragments used for biotin pulldown assays. d, e RNA pulldown assays were performed to detect the interaction between HuR and Ucp1 mRNA. The HEK293T cell lysates were incubated with each of the biotinylated Ucp1 mRNA fragments as indicated. HuR bound to the Ucp1 mRNA fragments was detected by Western blotting. f Western blot analysis of UCP1 in the BAT of HuR^f/f and HuR^−/− mice. g Western blot analysis of UCP1 in the iWAT of HuR^−/− and HuR^f/f mice following 4℃ exposure for 6 h. h The pGL3 luciferase reporter vector fused to the indicated Ucp1 mRNA fragment was transfected into HEK293T cells. The cells were treated with the NC siRNA or with HuR siRNA. At 48 h after siRNA transfection, the luciferase activity in each cell lysate was detected. Relative luciferase activity to the corresponding control sample was presented (n = 3/group). i-k HEK293T cells were transfected with NC siRNA or HuR siRNA for 24 h and then transfected with luciferase, luciferase-Ucp1 3’UTR, luciferase-Ucp1 3’UTR-ARE mutant reporter for 12 h. Cells were exposed to Dox (1 µg/ml) and cellular RNA was prepared at the indicated time points after Dox addition. Real-time qPCR was performed to assess the half-lives of the luciferase (i), luciferase-Ucp1 3’UTR (j), luciferase-Ucp1 3’UTR-ARE mutant (k) (n = 3/group)

To test if the association of HuR with Ucp1 mRNAs was functional, pGL3-derived reporters bearing fragments of Ucp1 mRNA 3’UTR and 3’UTR-M were constructed. HEK293 cells were transfected with each of these reporters and 24 h later, they were transfected with siRNAs (control or HuR-directed) and cultured for an additional 48 h. Knockdown of HuR increased the luciferase activity of pGL3-derived vectors bearing the Ucp1 mRNA 3’UTR. However, knockdown of HuR had no effect on the activity of pGL3 and pGL3-Ucp1-3’UTR-M (Fig. 4h). Further tests on the half-lives of Ucp1 mRNAs in cells with silenced HuR showed that knockdown of HuR significantly extended the half- lives of Ucp1-3’UTR (3.4 versus 5.9 h), but did not influence the half-lives of the pGL3 (Fig. 4i and j). Moreover, the half-lives of pGL-Ucp1 3’UTR reporter mRNA with the mutated sequence was higher than that with non-mutated sequence (Fig. 4k). Taken together, HuR was capable of destabilizing the Ucp1 3’UTR chimeric transcripts and the UUUUUUUU motif in the Ucp1 3’UTR is the response element for HuR.

Alteration of HuR expression and its cellular distribution in DIO mice

As shown in Fig. 5a and b, no changes were indicated either in the mRNA expression or protein expression of HuR in the DIO mice. However, the expression of HuR protein in BAT, iWAT or eWAT was higher in the cytoplasm in DIO mice, with a concomitant reduction in the nuclear (Fig. 5c). Results from intervention with cold exposure or β-adrenergic agonist indicated that the HuR protein levels in BAT, iWAT or eWAT were reduced either with cold exposure (Fig. 5d) or CL316,243 treatment (Fig. 5e). Further examination on distribution between cytoplasm and nuclear found that the HuR protein was reduced in the cytoplasm or the nucleus of BAT, iWAT or eWAT, either with cold exposure or CL316,243 treatment (Fig. 5f and g). These data suggest that HuR may be involved in the adaptive thermogenesis of fat and decreased adipose HuR expression may impair systemic energy homeostasis.

Fig. 5.

Changes in HuR expression with HFD feeding and exposure to cold and β-agonists. a-c Three to four-week-old C57BL/6J male mice were fed a high-fat diet (HFD) for 16 weeks, with a normal control diet (NC). a mRNA levels of HuR in the adipose tissue (n = 8/group). b Total HuR protein in the adipose tissue (n = 3/group). c Cellular distribution of HuR protein in the adipose tissue (n = 3/group). d-g HuR protein expression and its cellular distribution in mice with NC feeding. d Total HuR protein in the adipose tissue with cold (4℃) exposure for 6 h (n = 3/group). e Total protein of HuR in the adipose tissue with daily injection of CL316243 (1 mg/kg/day) for 7 days with saline as control (n = 3/group). f Cellular distribution of HuR protein in the adipose tissue with cold (4℃) exposure (n = 3/group). g Cellular distribution of HuR protein in the adipose tissue with daily injection of CL316243 (n = 3/group)

Discussion

In the present study, adipocyte-specific HuR knockout did not affect the body weight in mice with NC feeding. However, under HFD feeding, adipocyte-specific HuR deficiency inhibited mouse body weight gain, being accompanied by less BAT whitening and more WAT browning. Meanwhile, the expression of adipose thermogenic genes (Pgc-1α, Ucp1, or Dio2) were up-regulated in HuR knockout mice, among which the 3’UTR -B fragment of Ucp1 mRNA could bind to HuR, suggesting that Ucp1 is the target gene for HuR. Thus, HuR may bind with the 3′UTR of Ucp1 mRNA to affect its stability, and thereby downregulate the Ucp1 mRNA expression. Furthermore, although the HuR expression was not changed in obesity, there was an alteration of HuR protein in cellular distribution of adipocytes, characterized by increased accumulation of HuR in the cytoplasm and concomitant reduction in the nuclear, indicating enhanced transfer of HuR protein from the nuclear to cytoplasm. This might impact the Ucp1 expression and is involved in the pathogenesis of obesity.

With respect to the effects of HuR in obesity, although several papers have been published using HuR knockout mice, the conclusions are still in controversy [13–15]. In keeping with our findings, it is reported that mice with an adipocyte-specific deletion of HuR are leaner and display increased energy expenditure [15]. As well, intestinal epithelium-specific HuR knockout reduces the expression of DGAT2 and MGAT2, thereby reducing the dietary fat absorption through TAG synthesis and mitigating HFD-induced non-alcoholic fatty liver disease and obesity [22]. The underlying reasons for the discrepancy in conclusion have been considered to be involved in several aspects, among which the differences in high-fat diets used and the severity of obesity may be more important factors. As far as no effects of HuR deficiency on body weight of NC-fed mice was concerned, the temperature (22 °C) used to house the mice was below their thermoneutral (30 °C) and may stimulate BAT thermogenesis [19, 20], which is probably overriding the effects of HuR on energy expenditure.

Many earlier studies have demonstrated that HuR normally targets and stabilizes ARE-containing transcripts, or destabilizes its targets in very rare cases [11, 23–25]. In the current study, the upregulated expression of Ucp1, the HuR’s target, in the knockout adipose tissue, means a transcript-destabilizing function. This is in contrast to the previous reports that HuR’s target genes (Atgl and Ingl) are downregulated in the knockout adipose tissue [13]. However, among the limited number of RBPs which have been studied in the adipose tissue, deletion of most of them (Pspc1, Sam68, Rbm4, Igf2bp2, or Ksrp) in mice shows less fat mass and lipid storage, increased energy expenditure, or WAT browning, leading to beneficial metabolic phenotypes [10]. Thus, it is sensible that adipocyte-specific HuR deletion inhibited HFD-induced obesity in the current study.

To note, HuR, like many other RNA binding proteins, should be able to recognize and influence many targets to exert its biological function through multiple biological pathways. Thus, by no means, Ucp1 is the sole target for HuR, and the lean phenotype of adipose tissue-specific HuR deficiency should probably be attributable to multiple targets. Moreover, HuR may function via changes in its abundance, subcellular localization, regulation on other stages of the RNA process, including alternative splicing, RNA translocation, and polyadenylation [26–30]. Therefore, identification of other targets for HuR and elucidation of their interactions at multiple stages will warrant further investigation.

Ucp1, being expressed in classical brown adipocytes, as well as in “beige cells” in WAT, plays an important role in energy expenditure and obesity [31, 32], particularly in BAT non-shivering thermogenesis which converts the energy of mitochondrial proton gradient into heat instead of ATP [33, 34]. Thermogenesis is a process that happens in brown adipose tissue and beige adipocytes induced in iWAT at early stage of obesity or in response to cold exposure or β3-adrenoceptor agonist stimulation [35, 36]. The expression of Ucp1 mRNA is tightly controlled by various transcription factors and co-regulators, such as Pgc1a, Prdm16, Pparγ, and Zfp516, at the promoter region of the UCP1 gene [37–39]. Our findings demonstrated that HuR acted as a post-transcriptional regulator of Ucp1 regulated gene expression, being consistent with other reports [40].

Moreover, in this study, both the mRNA and protein expression of HuR either in WAT or BAT were not changed in HFD-induced obesity. The data are inconsistent with a previous study demonstrating that the mRNA levels of HuR in eWAT, iWAT and BAT were decreased from HFD-fed mice [13]. The differences between the two results may be related to the different reference genes used [41], or the severity of obesity induced. Further analysis showed that the nucleus-cytoplasm translocation of HuR protein was increased in HFD-induced obesity. As expression levels of transcription factors and co-regulators relevant to Ucp1 expression were not much affected, it appears that nutrient enrichment facilitates cytoplasmic HuR level increased, then HuR binding to Ucp1 mRNA, resulting in its degradation.

Limitations of this study

This study investigated the role of HuR-mediated regulation of Ucp1 exclusively in the context of mouse obesity, with lack of validation in human samples. It is difficult to directly extrapolate the findings to the pathophysiological processes of human obesity, as species differences may affect the clinical relevance of the conclusions. Although it is confirmed that HuR downregulates Ucp1 expression by binding to the 3’UTR of Ucp1 mRNA, the regulatory mechanisms of other thermogenesis-related genes (such as Pgc-1α) are not thoroughly verified (e.g., whether they are direct targets of HuR or indirectly regulated through Ucp1). This may overlook other pathways by which HuR participates in thermogenesis regulation via other target genes.

Conclusions

In conclusion, adipose tissue-specific HuR deletion alleviated HFD-induced obesity through upregulating Ucp1 expression in mice. Although the HuR expression was not changed in obesity, there was an enhanced transfer of HuR protein from the nuclear to cytoplasm, that might impact the expression of target genes including Ucp1 and thereby be involved in the pathogenesis of obesity.

Abbreviations

RBP

RNA-binding protein

HuR

Human antigen R

Ucp1

Uncoupling protein 1

BAT

Brown adipose tissue

WAT

White adipose tissue

iWAT

Inguinal WAT

eWAT

Epididymal WAT

Authors’ contributions

XF participated in the study design, data accuracy, and paper writing. YW carried out mouse feeding. RW, TT and MY participated in some experiments. PL performed the statistical analysis. KQ participated in its design and revised the whole manuscript. All the authors reviewed and approved the final manuscript.

Funding

This study was supported by the National Natural Science Foundation of China (No. 81800750) and Funding for Reform and Development of Beijing Municipal Health Commission.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

All animal experiments in this study were approved by the Committee on the Ethics of Institute of Laboratory Animal Sciences, National Institute of Occupational Health and Poison Control of China. The study complied with the relevant ethical regulations pertaining to animal research, and all laboratory animals were cared for and used according to institutional guidelines.

Consent for publication

All authors have read and agreed with the submission of the manuscript to Lipids in Health and Disease.

Competing interests

The authors declare no competing interests.