Cold‐inducible lncRNA266 promotes browning and the thermogenic program in white adipose tissue

Abstract

Cold‐induced nonshivering thermogenesis has contributed to the improvement of several metabolic syndromes caused by obesity. Several long noncoding RNAs (lncRNAs) have been shown to play a role in brown fat biogenesis and thermogenesis. Here we show that the lncRNA lnc266 is induced by cold exposure in inguinal white adipose tissue (iWAT). In vitro functional studies reveal that lnc266 promotes brown adipocyte differentiation and thermogenic gene expression. At room temperature, lnc266 has no effects on white fat browning and systemic energy consumption. However, in a cold environment, lnc266 promotes white fat browning and thermogenic gene expression in obese mice. Moreover, lnc266 increases core body temperature and reduces body weight gain. Mechanistically, lnc266 does not directly regulate Ucp1 expression. Instead, lnc266 sponges miR‐16‐1‐3p and thus abolishes the repression of miR‐16‐1‐3p on Ucp1 expression. As a result, lnc266 promotes preadipocyte differentiation toward brown‐like adipocytes and stimulates thermogenic gene expression. Overall, lnc266 is a cold‐inducible lncRNA in iWAT, with a key role in white fat browning and the thermogenic program.

Lnc266 is a cold‐inducible long noncoding RNA in mouse inguinal white adipose tissue. By sponging miR‐16‐1‐3p, lnc266 stimulates Ucp1 expression, drives the thermogenic program, and promotes white fat browning in obese mice.

Introduction

Thermogenesis is an indispensable way for endotherms to maintain their body temperature. Cold ambient temperatures stimulate both nonshivering thermogenesis and shivering thermogenesis in endotherms, which is called adaptive thermogenesis (Lowell & Spiegelman, 2000; Johann et al, 2019). Shivering thermogenesis is mediated by rhythmic contraction and relaxation of skeletal muscle, while nonshivering thermogenesis is mainly mediated by metabolic pyrexia of brown adipose tissue (BAT) (Lee et al, 2014; Chouchani et al, 2019). Nonshivering thermogenesis is characterized by continuous thermogenesis compared to shivering thermogenesis, which is important for maintaining body temperature in a moderately cold environment (Puigserver et al, 1998; van Marken Lichtenbelt et al, 2009). A growing number of evidences have shown that nonshivering thermogenesis has benefits against metabolic disorders such as obesity, type 2 diabetes, hyperlipidemia, and fatty liver (Bartelt & Heeren, 2014; Suarez‐Zamorano et al, 2015; Palmer & Clegg, 2017; Betz & Enerback, 2018; Chouchani & Kajimura, 2019; Scheele & Wolfrum, 2020).

Adipose tissue, especially BAT, plays a key role in nonshivering thermogenesis (Dawkins & Scopes, 1965; Brooks et al, 1980; Rosen & Spiegelman, 2014). Traditionally, BAT locates in the shoulder blade of newborn infants, which is important for maintaining body temperature and gradually faded with age. In adults, however, there are still small amounts of brown adipose tissue in the neck and collarbone (Cypess et al, 2009; van Marken Lichtenbelt et al, 2009; Virtanen et al, 2009). Recently, a new type of adipocytes called beige adipocytes were found to be scattered in white adipose tissue (WAT), and they could be transformed into brown‐like cells with external stimuli such as prolonged cold exposure or elevated intracellular cyclic AMP (cAMP) (Guerra et al, 1998; Petrovic et al, 2010; Wu et al, 2012). At this regard, strategies by inducing brown‐like adipocytes from beige adipocytes (or termed as browning) may hold great therapeutic potentials for treating excessive energy storage associated metabolic diseases including obesity, insulin resistance, and type 2 diabetes (Ishibashi & Seale, 2010; Harms & Seale, 2013). Indeed, numerous studies have shown that browning has benefits for rectifying metabolic dysfunctions from rodents to humans (Dempersmier et al, 2015; Hanssen et al, 2015; Suarez‐Zamorano et al, 2015; Min et al, 2016).

Long noncoding RNAs (lncRNAs), a type of non‐coding RNA with a length of more than 200 nucleotides, play an important role in cell division, growth, and differentiation (Palazzo & Koonin, 2020). lncRNAs have a variety of regulatory modes and manipulate gene expression at both transcriptional and post‐transcriptional levels (Schmitt & Chang, 2016). In general, lncRNAs could be functioned as signals, decoys, guides, and/or scaffolds (Wang & Chang, 2011). With a profound understanding of lncRNAs, multiple lncRNAs have been shown to be involved in regulation of BAT thermogenesis and WAT browning in different ways. For example, Blnc1 forms ribonucleoprotein complex with transcription factor EBF2 to promote brown and beige adipocyte differentiation (Zhao et al, 2014). AK079912, a brown adipocyte‐enriched lncRNA, is capable of driving the expression of thermogenic genes in white preadipocytes (Xiong et al, 2018). LINC00473 coordinates the binding of lipid droplets during mitochondrial fission, which is necessary for thermogenesis in adipocytes (Tran et al, 2020). Hence, lncRNAs could be important regulators for orchestrating adipocyte differentiation and thermogenesis.

In the present study, lnc266 was identified as a cold‐inducible lncRNA in inguinal WAT (iWAT). Ectopic expression of lnc266 promotes preadipocyte differentiation toward brown adipocytes, and on the contrary, knockdown of lnc266 impedes this differentiation. In vivo experiments showed that lnc266 increases mouse core temperature and stimulates thermogenic gene expression in iWAT in a cold environment. On the contrary, knockdown of lnc266 in iWAT reduces thermogenic gene expression and white fat browning. Mechanistically, lnc266 interacts with miR‐16‐1‐3p and thus attenuates the repression of miR‐16‐1‐3p on Ucp1 expression.

Results

Cold exposure stimulates lipid catabolism and thermogenesis in mouse iWAT

To explore the biological processes induced by cold exposure in white adipose tissue, mice were placed at a cold environment (4°C) for 4, 8 and 16 days. Prior to cold exposure, 3‐day cold acclimation was performed to avoid acute cold stress (Fig 1A). Inguinal white adipose tissue (iWAT) is a unique WAT, in which a subset of UCP1⁺ cells hold high propensity for beiging (brown adipose tissue‐like phenotype change) in response to β‐adrenergic stimulation (Sakers et al, 2022). For this reason, iWAT was chosen for microarray analysis. The whole experimental scheme is illustrated in Fig 1A. The gross morphology of iWAT is presented in Fig 1B. The color of iWAT switched from white to brown along with the extended cold treatment. Totally, there are 6,645 differentially expressed genes in iWAT after cold exposure (Fig 1C; Appendix Fig S1A). The gene ontology‐biological process (GO‐BP) enrichment analysis revealed that these differentially expressed genes are closely associated with lipid catabolism, such as oxidation–reduction process, lipid metabolic process, transport, fatty acid metabolic process, fatty acid beta‐oxidation, and mitochondrial translation (Fig 1D). In agreement with these findings, we noticed that genes involved in lipid transport and lipolysis are upregulated in iWAT with cold exposure (Fig 1E; Appendix Fig S1B–D). The qRT‐PCR data validated the microarray results (Fig 1F). These results indicate that cold exposure evokes browning and thermogenic gene expression in iWAT.

Figure 1. Cold exposure stimulates thermogenesis and lipid catabolism in iWAT.

1. Timeline of experimentation. After 3‐day cold acclimation, mice were housed at 4°C for 4, 8, and 16 days, respectively. Mice in control group were kept at 23°C throughout all the experiment. Inguinal white adipose tissue (iWAT) was sampled for Microarray. Results are the mean ± SEM of 3–4 biological replicates.
2. The gross morphology of iWAT.
3. Number of differentially expressed genes induced by cold exposure in mouse iWAT.
4. The gene ontology‐biological process (GO‐BP) enrichment analysis for differentially expressed genes at three time points (4, 8, and 16 days) induced by cold exposure.
5. Heatmap for thermogenic genes induced by cold exposure. Blue and red shading indicates downregulation and upregulation, respectively. The color key at right side is Row Z‐Score.
6. Confirmation of the microarray data by qRT‐PCR. 18S rRNA was used as a house‐keeping gene. Results are the mean ± SEM of three biological replicates.

Data information: *P < 0.05, **P < 0.01, ***P < 0.001 (two‐tailed Student's t test).

Source data are available online for this figure.

lnc266 is a cold‐inducible lncRNA in iWAT and involved in brown adipocyte differentiation

In addition to coding RNAs, we noticed there are many differentially expressed lncRNAs in the Microarray data. Ten upregulated lncRNAs with the biggest changes were listed (Fig 2A). To confirm these Microarray data, we reanalyzed these upregulated lncRNAs by qRT‐PCR, and found that six lncRNAs (lnc266, lnc433, lnc103, lnc940, lnc767, and lnc797) exhibited similar expression patterns with comparison of the heatmap (Fig 2B). Of note, only lnc266 and lnc433 were stably upregulated with cold exposure (Fig 2B), and for this reason, we focused our interests on these two lncRNAs in the following experiments. First, we examined these lncRNA expression in different types of adipose tissues in mice, i.e., epididymal WAT (eWAT), iWAT, and BAT. The results showed that lnc266 and lnc433 were highly expressed in iWAT and eWAT; whereas in BAT, their expressions are relatively low (Fig 2C). Second, we analyzed their expression profiles during adipocyte differentiation. To this aim, C3H10T1/2 cells were subjected to differentiation induction toward brown adipocytes. Both lnc266 and lnc433 were gradually increased during the induction (Fig 2D). Compared with lnc433, the changes of lnc266 during the differentiation are much higher (Fig 2D). Therefore, it was chosen for further investigation. The gene of lnc266 is in chromosome 2 with two exons (Appendix Fig S2A). Due to overlap between lnc266 and Kcnb1, we examined the specificity of primers and our data showed that the used primers are specific for amplifying lnc266 in qPCR analysis (Appendix Fig S2B and C). Lnc266 was widely expressed in various tissues including liver, spleen, kidney, stomach, lung, heart, muscle, and fat (Appendix Fig S2D). Lnc266 expression in iWAT was induced by cold exposure according to the fluorescence in situ hybridization data (Fig 2E). Even in eWAT and BAT, lnc266 expression was also in a cold‐inducible manner (Fig 2F). Moreover, the expression of lnc266 was decreased in iWAT, eWAT, and BAT in high‐fat induced obese mice (Fig 2G). These findings strongly suggest that lnc266 may play a role in thermogenesis in iWAT with cold exposure. In addition, we found that lnc266 could be reversely aligned to the 3′‐UTR of Kcnb1 (potassium voltage‐gated channel, Shab‐related subfamily, member 1) by blasting (ranking from 3,443 to 3,069 and 4,664 to 4,178), suggesting lnc266 is an overlapping‐antisense lncRNA, and these two genes are transcribed from the same gene. Most importantly, similar to lnc266, Kcnb1 expression in iWAT was gradually increased by cold exposure as evidenced by the Microarray data and qRT‐PCR results (Appendix Fig S3A and B).

Figure 2. Cold exposure induces lncRNAs in iWAT.

1. Heatmap for differentially expressed lncRNAs induced by cold exposure. Blue and red shading indicates downregulation and upregulation, respectively. The color key at right side is Row Z‐Score.
2. Confirmation of the microarray data by qRT‐PCR. Results are the mean ± SEM of three biological replicates.
3. Lnc266 and lnc043 expression in eWAT, iWAT and BAT. Results are the mean ± SEM of four biological replicates.
4. Lnc266 and lnc043 expression profiles in brown adipocyte differentiation. C3H10T1/2 cells were subjected to differentiation towards brown adipocytes. lncRNA expression was analyzed at different time points as indicated. Results are the mean ± SEM of three biological replicates.
5. Fluorescence in situ hybridization analysis showing lnc266 was induced by cold exposure in iWAT. Signals for lnc266 were in red color, and the nuclei were stained with DAPI in blue color. Scale bar = 50 μm.
6. Lnc266 expression was induced by cold exposure in eWAT and BAT. Results are the mean ± SEM of three biological replicates.
7. The expression of lnc266 was decreased in adipose tissue of obese mice. Results are the mean ± SEM of 4–7 biological replicates.

Data information: eWAT: epididymal white adipose tissue; iWAT: inguinal white adipose tissue; BAT: brown adipose tissue. SD: standard diet; HFD: high‐fat diet. Gene expression was analyzed by qRT‐PCR, and 18S rRNA was used as a house‐keeping gene. *P < 0.05, **P < 0.01, ***P < 0.001 (two‐tailed Student's t test).

Source data are available online for this figure.

lnc266 induces preadipocyte differentiation toward brown adipocytes

Our above data showed that lnc266 in iWAT is involved in cold adaption, raising a possibility that lnc266 plays a role in white fat browning. To verify the potentials of lnc266 in brown adipogenesis, C3H10T1/2 cells were transfected with a plasmid expressing lnc266 and then cells were subjected to differentiation induction toward brown adipocytes. Lnc266 expression was increased by the transfection (Fig 3A). Overexpression of lnc266 increased lipid droplet accumulation in differentiated adipocytes (Fig 3B and C). Meanwhile, more mitochondria were observed in cells overexpressing lnc266 (Fig 3D). Brown adipocyte markers including Ucp1, Cidea, Ppara, Elovl3 were modestly upregulated by lnc266, while Cpt1b and Ppargc1a were not altered (Fig 3E). Meanwhile, lnc266 also induced pan adipocyte marker gene expression such as Pparg, Fabp4, Atgl, and Adipoq (Fig 3F). The protein levels of UCP1 were increased in lnc266‐transfected cells (Fig 3G and H). To further confirm the roles of lnc266 in brown adipocyte induction, we decreased lnc266 expression by adenovirus carrying lnc266 shRNA (ADV‐lnc266 shRNA) (Fig 4A). Knockdown of lnc266 impaired adipogenesis as evidenced by reduced lipid droplets (Fig 4B and C). The number of mitochondria in differentiated adipocytes was also decreased by lnc266 knockdown (Fig 4D). Gene expression analysis revealed that brown adipocyte specific genes including Ucp1, Ppara, Elovl3, and Cpt1b were decreased by lnc266 knockdown (Fig 4E). Other genes as general adipocyte markers such as Pparg, Fabp4, Atgl, Hsl, and Adipoq also were reduced by lnc266 knockdown (Fig 4F). UCP1 protein was reduced by lnc266 shRNA (Fig 4G and H). To confirm these findings, we synthesized siRNAs targeting lnc266 and re‐examined loss‐of‐function studies. Of note, these lnc266 siRNAs target different regions as compared to the used lnc266 shRNA (Appendix Fig S4). Of these siRNAs, lnc266 siRNA‐2 exhibited the best knockdown efficiency (Appendix Fig S5A), which was chosen for following experiments. Our data clearly showed that, as well as lnc266 shRNA, lnc266 siRNA‐2 reduced adipogenesis, mitochondrial number, UCP1 expression, brown adipocyte specific gene expression, and lnc266 expression in differentiated adipocytes (Appendix Fig S5B–H). These data indicate that lnc266 plays a key role in preadipocyte differentiation toward brown adipocytes. In addition, overexpression of lnc266 induced Pparg, Atgl, Hsl, and Adipoq expression in differentiated white adipocytes (Appendix Fig S6A and B); and knockdown of lnc266 reduced Fabp4, Adipoq, and Retn expression (Appendix Fig S6C and D), suggesting lnc266 may play a role in white adipocyte differentiation.

Figure 3. Lnc266 induces preadipocyte differentiation towards brown adipocytes.

C3H10T1/2 cells were transfected with a plasmid expressing lnc266 by electroporation and subjected to differentiation induction towards brown adipocytes for 6 days.

• A
  Lnc266 expression was increased by transfection with the plasmid carrying lnc266.
• B, C
  Lnc266 promotes adipogenesis. Lipid droplets were stained with Bodipy (green; B) and Oil Red O (red; C). Scale bar = 50 μm.
• D
  Lnc266 improves mitochondrial number. Mitochondrial cytopainter was used to evaluate mitochondrial number. Scale bar = 50 μm.
• E
  Lnc266 increases thermogenic gene expression.
• F
  Lnc266 induces gene expression involved in adipocyte differentiation.
• G
  Lnc266 enhances UCP1 expression.
• H
  Quantification of the western blot data as shown in (G).

Data information: Gene expression was analyzed by qRT‐PCR, and 18S rRNA was used as a house‐keeping gene. Protein levels were analyzed by western blot analysis, and Actin was used as a loading control. EV: empty vector. Results are the mean ± SEM of three biological replicates; *P < 0.05, **P < 0.01, ***P < 0.001 (two‐tailed Student's t test).

Source data are available online for this figure.

Figure 4. Knockdown of lnc266 impedes preadipocyte differentiation towards brown adipocytes.

C3H10T1/2 cells were transduced with adenovirus expressing lnc266 shRNA (ADV‐lnc266 shRNA) or control virus (ADV‐scramble shRNA). Cells were then induced to differentiation towards brown adipocytes for 6 days.

• A
  Lnc266 expression was decreased by ADV‐lnc266 shRNA.
• B, C
  Knockdown of lnc266 impairs lipid droplet accumulation in differentiated adipocytes. Lipid droplets were stained with Bodipy (green; B) and Oil Red (red; C). Scale bar = 50 μm (B); Scale bar = 20 μm (C).
• D
  Knockdown of lnc266 reduces mitochondrial number. Mitochondria were stained by mitochondrial cytopainter. Scale bar = 20 μm.
• E
  Knockdown of lnc266 inhibits thermogenic gene expression in differentiated adipocytes.
• F
  Knockdown of lnc266 impedes gene expression involved in adipocyte differentiation.
• G
  Knockdown of lnc266 reduces the protein level of UCP1 in differentiated adipocytes.
• H
  Quantification of the western blot data as shown in (G).

Data information: Gene expression was analyzed by qRT‐PCR and 18S rRNA was used as a house‐keeping gene. Protein levels were analyzed by western blot analysis, and Actin was used as a loading control. Results are the mean ± SEM of three biological replicates; *P < 0.05, **P < 0.01, ***P < 0.001 (two‐tailed Student's t test).

Source data are available online for this figure.

lnc266 has no effects on energy homeostasis in obese mice under room temperature

The ability of lnc266 for inducing brown adipocyte differentiation prompt us to examine the potential roles of lnc266 on energy homeostasis in vivo. To this end, mice were fed on high fat diet (HFD) for consecutive 60 days, and then mice were received adeno‐associated virus expressing lnc266 (AAV‐lnc266) or control virus in iWAT via in situ injection; 42 days post‐virus injection, mice were sacrificed and samples were collected for further analyses (Fig 5A). Of note, the whole experiment was carried out at 22–24°C. The food intake was not altered by AAV‐lnc266 (Fig 5B). The body weight of AAV‐lnc266 mice is larger than that of control mice, but there is no statistical difference (Fig 5C). The body weight growth rate was slightly reduced by AAV‐lnc266 without statistical significance (Fig 5D). The core temperature was also not affected (Fig 5E). The adipocyte size in iWAT is decreased in AAV‐lnc266 treated mice based on the hematoxylin–eosin (H&E) staining (Fig 5F and G). Immunohistochemical analysis showed that the expression of UCP1 in iWAT was not affected by lnc266 (Fig 5H). The mRNA levels of UCP1 were slightly increased without statistical significance (Fig 5I). As for the expression of lnc266, it was markedly increased in AAV‐lnc266‐treated iWAT (Fig 5I). Liver glycerol and triglyceride (TG) were slightly reduced (Fig 5J). These data indicate that overexpression of lnc266 in iWAT fails to affect energy homeostasis in obese mice at room temperature.

Figure 5. Lnc266 has no effect on white adipose tissue browning in obese mice.

1. Timeline of experimentation. Mice were fed on a high‐fat diet (HFD) for consecutive 60 days. Adeno‐associated virus expressing lnc266 (AAV‐lnc266) or control virus was administered in bilateral iWAT via in situ injection. 42 days post‐virus injection, mice were sacrificed and samples were collected for further analysis. Results are the mean ± SEM of six biological replicates.
2. Food intake. Six mice in each group were placed in one cage, and averaged food intake for each mouse was calculated and presented at each time point.
3. Body weight curve. Results are the mean ± SEM of six biological replicates.
4. Body weight (BW) growth rate. Results are the mean ± SEM of six biological replicates.
5. Rectal core body temperature. Results are the mean ± SEM of six biological replicates.
6. H&E staining for iWAT. Scale bar = 50 μm.
7. Quantified adipocyte size based on H&E staining as shown in (F). Results are the mean ± SEM of four biological replicates.
8. Immunohistochemical analysis of UCP1 in iWAT. Scale bar = 50 μm.
9. Ucp1 and lnc266 expression in iWAT. Results are the mean ± SEM of five biological replicates.
10. The triglyceride (TG) and glycerol levels in the liver. Results are the mean ± SEM of four biological replicates.

Data information: Gene expression was analyzed by qRT‐PCR, and 18S rRNA was used as a house‐keeping gene. ***P < 0.001 (two‐tailed Student's t test for D, E, G, I, J; two‐way ANOVA analysis for C).

Source data are available online for this figure.

lnc266 stimulates the thermogenic program in iWAT in a cold environment

Due to lnc266 has no effects on energy homeostasis under room temperature, we next examined whether it affects the thermogenic program in a cold environment. To this end, mice were fed on HFD for consecutive 60 days, and AAV‐lnc266 was directly injected into bilateral iWAT. 3 weeks post‐injection, mice were placed in a cold environment for seven consecutive days (Fig 6A). Cold exposure resulted in a declining trend in body weight and an inclining trend in food intake, although no statistical significances were observed (Fig 6B–D; Appendix Table S1), indicating an increased energy demand for combating cold exposure. Compared with the control group, AAV‐lnc266‐treated mice consumed less food but held higher core temperature, indicating they utilized more energy from themselves (Fig 6B and E). For this reason, AAV‐lnc266‐treated mice lost more body weight (Fig 6C and D). In iWAT, in situ injection of AAV‐lnc266 really increased the expression of lnc266 (Fig 6F). Of note, this increase induced by AAV‐lnc266 is higher than that shown in Fig 5I. This discrepancy is likely due to the differences in circumstance (22°C vs 4°C) and/or the time after AAV‐lnc266 injection (42 days vs 35 days). Genes involved in brown adipocyte differentiation including Ucp1, Ppara, Cpt1b, Cidea, Cox8b, and Ppargc1a were enhanced by lnc266 in iWAT (Fig 6G). H&E staining showed that lnc266 induced brown‐like adipocytes in iWAT as evidenced by small and multilocular lipid droplets (Fig 6H). Adipocyte size in iWAT became smaller in AAV‐lnc266‐treated mice (Fig 6I). Immunohistochemical analysis revealed UCP1 expression was increased by lnc266 in iWAT (Fig 6J). Western blot data further confirmed that UCP1 expression was upregulated in AAV‐lnc266‐treated iWAT (Fig 6K). Liver glycerol was reduced by lnc266, and liver triglyceride was decreased slightly without statistical significance (Fig 6L). All these data show that lnc266 has a positive role in white fat browning and the thermogenic program in mice in a cold environment.

Figure 6. lnc266 increases thermogenesis in iWAT of obese mice with cold exposure.

• A
  Timeline of experimentation. Mice were fed on a high‐fat diet (HFD) for consecutive 60 days. Adeno‐associated virus expressing lnc266 (AAV‐lnc266) was administered in bilateral iWAT via in situ injection.
• B
  Food intake. Results are the mean ± SEM of 4–5 biological replicates.
• C
  Body weight curve of mice. Results are the mean ± SEM of 4–5 biological replicates.
• D
  Body weight (BW) growth rate. Results are the mean ± SEM of 4–5 biological replicates.
• E
  Rectal core body temperature curve of mice. Results are the mean ± SEM of 4–5 biological replicates.
• F, G
  Lnc266 and other thermogenesis‐related gene expression in inguinal white adipose tissues (iWAT). Gene expression was analyzed by qRT‐PCR and 18S rRNA was used as a house‐keeping gene. Results are the mean ± SEM of 4–5 biological replicates.
• H
  H&E staining for iWAT. Scale bar = 50 μm.
• I
  Quantified adipocyte size based on H&E staining as shown in (H). Results are the mean ± SEM of 4–5 biological replicates.
• J
  Immunohistochemical analysis of UCP1. Scale bar = 50 μm.
• K
  The protein levels of UCP1 in iWAT. UCP1 expression was analyzed by western blot, and Actin was used as a loading control. Results are the mean ± SEM of three biological replicates.
• L
  The glycerol and triglyceride (TG) levels in the liver. Results are the mean ± SEM of three biological replicates.

Data information: *P < 0.05, **P < 0.01, ***P < 0.001 (two‐tailed Student's t test for D, F, G, I, K, L; two‐way ANOVA analysis for B, C, E).

Source data are available online for this figure.

Knockdown of lnc266 reduces white fat browning and thermogenic gene expression in iWAT

Next, we asked whether loss of lnc266 affects browning and the thermogenic program in iWAT. To this end, we generated AAV expressing lnc266 shRNA (AAV‐lnc266 shRNA), which was then in situ injected into bilateral iWAT in mice. 14‐week post injection, we found that the body weight growth rate was increased by AAV‐lnc266 shRNA, although it was not significantly affected (Fig 7A and B). The core body temperature was reduced in mice treated with lnc266 shRNA under normal condition (22–24°C) (Fig 7C). Next these mice were challenged with cold exposure (4°C) to examine whether lnc266 knockdown affects the thermogenic program in a cold environment. As shown in Fig 7D, the body temperature of mice was markedly reduced with 1‐day cold challenge, and it became constant thereafter. As compared to mice treated with scramble shRNA, mice treated with lnc266 shRNA exhibited lower body temperature (Fig 7D). H&E staining showed that many adipocytes with multiocular lipid droplets were appeared in iWAT after 7‐day cold exposure; however, these changes were not occurred in mice with reduced expression of lnc266 (Fig 7E). The adipocyte size in iWAT was increased by lnc266 knockdown (Fig 7F). Moreover, the histochemical analysis indicated that high levels of UCP1 expression can be observed in iWAT of cold exposed mice treated with scramble shRNA, whereas knockdown of lnc266 largely prevented this trend (Fig 7G). Western blot analysis also showed that the protein levels of UCP1 in iWAT were reduced by knockdown of lnc266 (Fig 7H and I). Gene expression results further confirmed the effects of lnc266 knockdown on Ucp1 expression (Fig 7J). In addition, some other brown adipocyte marker genes including Ppara, Cpt1b, Elovl3, Cidea, Dio2, and Cox8b were decreased by lnc266 shRNA (Fig 7J). The expression of lnc266 in iWAT was successfully decreased by AAV‐lnc266 shRNA (Fig 7K). These data strongly support the notion that lnc266 plays a key role in white fat browning and the thermogenic program in iWAT.

Figure 7. Knockdown of lnc266 reduces thermogenesis in iWAT.

AAV‐lnc266 shRNA was in situ injected into bilateral iWAT to reduce lnc266 expression. Mice in control group were received AAV‐scramble shRNA (scram. shRNA) at the same dosage.

1. Body weight has an ascending trend at 14‐week post virus injection. Results are the mean ± SEM of five biological replicates.
2. Body weight (BW) growth rate was increased by knockdown of lnc266 in iWAT. Results are the mean ± SEM of five biological replicates.
3. Core body temperature was decreased in mice treated with lnc266 shRNA under normal condition. Results are the mean ± SEM of five biological replicates.
4. Core body temperature was decreased by lnc266 shRNA under cold condition. Results are the mean ± SEM of five biological replicates.
5. Knockdown of lnc266 impairs adipocyte transformation towards brown adipocytes induced by cold exposure. Samples of iWAT were subjected to H&E staining. Scale bar = 50 μm.
6. Quantified adipocyte size based on H&E staining as shown in (E). Results are the mean ± SEM of five biological replicates.
7. Histochemical analysis showing cold‐induced UCP1 expression in iWAT was attenuated by knockdown of lnc266. Scale bar = 50 μm.
8. The protein levels of UCP1 in iWAT were reduced by lnc266 shRNA.
9. Quantification of western blot data as shown in (G). Results are the mean ± SEM of five biological replicates.
10. Knockdown of lnc266 decreases brown adipocyte marker gene expression in iWAT. Results are the mean ± SEM of four biological replicates.
11. The expression of lnc266 in iWAT was decreased by lnc266 shRNA. Results are the mean ± SEM of four biological replicates.

Data information: Gene expression was analyzed by qRT‐PCR, and 18S rRNA was used as a house‐keeping gene. Protein was analyzed by western blot, and actin was used as loading control. *P < 0.05, **P < 0.01, ***P < 0.001 (two‐tailed Student's t test for A, B, C, F, I, J, K; two‐way ANOVA analysis for D).

Source data are available online for this figure.

lnc266 targets miR‐16‐1‐3p and stimulates the thermogenic program in iWAT

The above data indicate that lnc266 is required for cold‐induced white fat browning and restores the browning capacity in dietary obese mice in a cold environment. To explore the involved mechanisms, we first examined the cellular distribution of lnc266. Fluorescence in situ hybridization showed that lnc266 was expressed in the nucleus and cytoplasm (Fig 8A). The qRT‐PCR data further confirmed this conclusion (Fig 8B). Since UCP1 is a key factor for adipose tissue browning and thermogenesis (Rosen & Spiegelman, 2014), we next examined whether lnc266 directly regulates Ucp1 expression. To this end, 293T cells were transfected with a GFP reporter driven by the Ucp1 promoter. Meanwhile, the plasmid carrying lnc266 or Prdm16 was co‐transfected into the cells. As expected, as a potent inducer of Ucp1 (Seale et al, 2008), Prdm16 increased GFP expression (Fig 8C). Overexpression of lnc266 resulted in a marginal effect on GFP expression, suggesting lnc266 may not directly regulate Ucp1 expression (Fig 8C). Luciferase assay further confirmed this notion that lnc266 has no direct regulation on Ucp1 expression (Fig 8D). One of mechanisms for lncRNAs is to absorb miRNAs and thus repress their biological functions (Cech & Steitz, 2014). Based on bioinformatics analysis, 151 miRNAs are identified as potential targets of lnc266 by using LNCediting (http://bioinfo.life.hust.edu.cn/LNCediting/), and 123 miRNAs are predicted as regulators of Ucp1 expression by using TargetScan (https://www.targetscan.org/vert_71/) (Fig 8E). Of these miRNAs, five miRNAs including miR‐6359, 7065‐3p, 511‐3p, 295‐5p, and 16‐1‐3p have both roles as mentioned above (Fig 8E). Next, we synthesized these miRNA mimics, which were transfected into 293T cells. At the same time, the plasmid expressing Ucp1 was co‐transfected. Our data showed that the expression of Ucp1 was reduced by miR‐295‐5p and miR‐16‐1‐3p (Figs 8F and EV1). Furthermore, we examined the effects of miR‐295‐5p and miR‐16‐1‐3p on Ucp1 expression in differentiated adipocytes. To this aim, C3H10T1/2 cells were transfected with miR‐295‐5p or miR‐16‐1‐3p mimic by electroporation, and then cells were subjected to differentiation induction toward brown adipocytes. Gene expression analysis showed that miR‐16‐1‐3p reduced Ucp1 expression, while miR‐295‐5p had no such an effect (Figs 8G and EV2A and B). In addition, we also analyzed the protein level of UCP1 and found that, indeed, miR‐16‐1‐3p largely decreased UCP1 in differentiated adipocytes (Fig 8H). These data suggest that lnc266 stimulates white adipocyte browning and thermogenic gene expression, which may rely on sponging miR‐16‐1‐3p.

Figure 8. Lnc266 targets miR‐16‐1‐3p and stimulates Ucp1 expression in adipocytes.

1. Fluorescence in situ hybridization showing lnc266 distributes in the cytoplasm and nucleus in C3H10T1/2 cells. Scale bar = 20 μm.
2. Cellular distribution of lnc266 in C3H10T1/2 cells (left panel) and differentiated adipocytes from C3H10T1/2 cells (right panel). Gene expression was analyzed by qRT‐PCR. 18S rRNA was served as a cytoplasmic gene, and U6 was served as a nuclear gene.
3. Lnc266 has no direct effect on Ucp1 expression. 293T cells were co‐transfected with the plasmid expressing GFP driven by the promoter of Ucp1, and the plasmids expressing lnc266 or Prdm16 as indicated. Prdm16 was used as a positive control and empty vector (EV) pcDNA3.1 was used as a negative control. Scale bar = 50 μm.
4. Lnc266 alone fails to affect Ucp1 expression. 293T cells were transfected with the luciferase reporter constructed with the promoter of Ucp1 and the plasmid expressing lnc266 as indicated.
5. Predicted microRNAs that can bind to lnc266 and Ucp1. Bioinformatic analysis showing there are 151 miRNAs targeted by lnc266 (blue) and 123 miRNAs targets Ucp1 (orange), in which five miRNAs belong to both (deep blue).
6. Effects of selected miRNAs on Ucp1 expression in 293T cells. 293T cells were co‐transfected with the plasmid expressing Ucp1 and each miRNA mimic as indicated.
7. Effects of miR‐295‐5p and miR‐16‐1‐3p on Ucp1 expression in differentiated adipocytes. C3H10T1/2 cells were transfected with miR‐295‐5p or miR‐16‐1‐3p mimic by electroporation. Cells were then subjected to differentiation induction towards brown adipocytes. After 8 day‐induction, cells were harvested for Ucp1 expression analysis.
8. miR‐16‐1‐3p reduces UCP1 in differentiated adipocytes. C3H10T1/2 cells were transfected with miR‐16‐3‐1‐3p mimic by electroporation. Cells were then subjected to differentiation induction towards brown adipocytes. After 8 day‐induction, cells were harvested for UCP1 expressing analysis.

Data information: Gene expression was analyzed by qRT‐PCR and 18S rRNA was used as a house‐keeping gene. Protein levels were analyzed by western blot analysis and Actin was used as a loading control. EV: empty vector. Results are the mean ± SEM of three biological replicates; *P < 0.05, **P < 0.01 (two‐tailed Student's t test).

Source data are available online for this figure.

Figure EV1. Transfection efficiencies for miRNAs.

The expression of miR‐295‐5p, miR‐6359, miR‐511‐3p, miR‐16‐1‐3p, and miR‐7065. 293T cells were co‐transfected with the plasmid expressing Ucp1 and each miRNA mimic as indicated. 24 h post‐transfection, cells were harvested for analyzing miRNA expression by qRT‐PCR, U6 was used as a house‐keeping gene.

Data information: Results are the mean ± SEM of three biological replicates; *P < 0.05, **P < 0.01, and ***P < 0.001 (two‐tailed Student's t test).

Figure EV2. Transfection efficiencies for miR‐295‐5p and miR‐16‐1‐3p.

• A, B
  The expression of miR‐295‐5p (A) and miR‐16‐1‐3p (B). C3H10T1/2 cells were transfected with miR‐295‐5p or miR‐16‐1‐3p mimic by electroporation. Cells were then subjected to differentiation induction towards brown adipocytes. After 8 day‐induction, cells were harvested for miRNA expression analysis by qRT‐PCR. U6 was used as a house‐keeping gene.

Data information: Results are the mean ± SEM of three biological replicates; *P < 0.05 (two‐tailed Student's t test).

In most cases, the 3′ untranslated region (3′‐UTR) is a regulatory region of miRNAs. Therefore, Ucp1 3′‐UTR or mutated Ucp1 3′‐UTR was cloned to luciferase reporter (Fig 9A). In 293T cells, the generated luciferase reporter together with miR‐16‐1‐3p or mutated miR‐16‐1‐3p were transfected. The results showed that the wildtype 3′‐UTR of Ucp1 fused luciferase activity was decreased by miR‐16‐1‐3p; whereas mutated miR‐16‐1‐3p had no such an effect (Fig 9B). As expected, the mutated 3′‐UTR of Ucp1 fused luciferase activity driven by was not altered by miR‐16‐1‐3p (Fig 9B). Moreover, we also generated a luciferase reporter fused with lnc266 (Fig 9C). Subsequent luciferase reporter assay showed that miR‐16‐1‐3p reduced the luciferase activity, while mutated miR‐16‐1‐3p failed to inhibit the luciferase activity (Fig 9D). These data indicate that lnc266 can be targeted by miR‐16‐1‐3p, resulting in destabilization of luciferase mRNA and reduced luciferase activity. Later, we found that there are two more sites in lnc266 (2 and 3) with potential interaction with miR‐16‐1‐3p (Appendix Fig S7). Therefore, we generated biotin‐labeled lnc266 wild type (lnc266‐WT), mutated lnc266 at the site 1 (lnc266‐mut (1)), and mutated lnc266 at the sites 1, 2, and 3 (lnc266‐mut (1–3)). Following RNA pull‐down experiments revealed that the presence of miR‐16‐1‐3p was reduced in the pull‐down complex, especially in the lnc266‐mut (1–3) (Fig 9E and F). These data further confirmed the direct interaction between miR‐16‐1‐3p and lnc266. In differentiated adipocytes from C3H10T1/2 cells, miR‐16‐1‐3p mimic repressed Ucp1 expression; however, the presence of lnc266 abolished this repression (Figs 9G and EV3A and B). These data clearly indicate that lnc266 specifically interacts with miR‐16‐1‐3p and thus counteracts its repression on Ucp1 expression. To further verify this notion, we synthesized miR‐16‐1‐3p inhibitor, which was then transfected into C3H10T1/2 cells by electroporation. After 8 day‐differentiation induction, gene analysis showed that several genes associated with brown adipocyte differentiation and thermogenesis such as Ppara, Ucp1, Elovl3, and Cpt1b were upregulated by miR‐16‐1‐3p inhibitor (Fig 9H). Protein expression analysis showed that UCP1 was enhanced by miR‐16‐1‐3p inhibitor (Fig 9I). Furthermore, we asked whether miR‐16‐1‐3p inhibitor affects lnc266 induced UCP1 expression. To this end, the plasmid of lnc266 was transfected into C3H10T1/2 cells with or without miR‐16‐1‐3p inhibitor, and then cells were induced towards brown adipocytes. As expected, cells treated with lnc266 or miR‐16‐1‐3p inhibitor alone increased Ucp1 expression; and no synergistic effect was observed in cells treated with lnc266 and miR‐16‐1‐3p inhibitor together (Appendix Fig S8A). Similar changes were also occurred in UCP1 protein levels (Appendix Fig S8B). Comparable expression of lnc266 in the presence or absence of miR‐16‐1‐3p inhibitor was observed (Appendix Fig S8C). Overall, these data clearly indicate that lnc266 sponges miR‐16‐1‐3p and abolishes its repression on Ucp1 expression, leading to white adipocyte browning and thermogenic gene expression.

Figure 9. Lnc266 combined with miR‐16‐1‐3p to counteract its negative effects.

• A
  Constructs of luciferase reporters with the wild type or mutated 3′‐untranslated region (3′‐UTR) of Ucp1. Meanwhile, wild type and mutated miR‐16‐1‐3p mimic were designed as indicated.
• B
  miR‐16‐3p targets the 3′‐UTR of Ucp1 and represses the expression of Ucp1. 293T cells were transfected with the luciferase reporter, miR‐16‐1‐3p, and their mutated forms as indicated. 48 h post‐transfection, cells were harvested for luciferase reporter assay.
• C
  Constructs of luciferase reporters with lnc266. Meanwhile, wild type and mutated miR‐16‐1‐3p mimic were designed as indicated.
• D
  miR‐16‐1‐3p inhibits the expression of lnc266. 293T cells were transfected with the luciferase reporter with lnc266, mimic for wild type and mutated miR‐16‐1‐3p as indicated. 48 h post‐transfection, cells were harvested for luciferase reporter assay.
• E, F
  Lnc266 pulldown showing the interaction between lnc266 and miR‐16‐1‐3p. Biotin‐labeled wild type lnc266 (lnc266‐WT) and mutated lnc266 (lnc266‐mut) was transcribed and incubated with cell lysates from C3H10T1/2 cells. The pulldown samples were subjected to miRNA‐specific qRT‐PCR analysis to detect endogenously associated miR‐16‐1‐3p with lnc266.
• G
  Lnc266 counteracts the reduced Ucp1 expression induced by miR‐16‐1‐3p. C3H10T1/2 cells were transfected with the plasmid expressing lnc266 and miR‐16‐1‐3p mimic by electroporation. Cells were then subjected to differentiation induction towards brown adipocytes. After 8 day‐induction, cells were harvested for Ucp1 expression analysis.
• H
  miR‐16‐1‐3p inhibitor stimulates the expression of Ucp1 and other thermogenesis related genes. C3H10T1/2 cells were transfected with miR‐16‐1‐3p inhibitor by electroporation. Cells were then subjected to differentiation induction towards brown adipocytes. After 8 day‐induction, cells were harvested for gene expression analysis.
• I
  The protein levels of UCP1 in differentiated adipocytes from C3H10T1/2 cells. Cell treatments were described in (H).

Data information: Gene expression was analyzed by qRT‐PCR, and 18S rRNA was used as a house‐keeping gene. Protein levels were analyzed by western blot analysis, and actin was used as a loading control. Results are the mean ± SEM of three biological replicates. *P < 0.05, **P < 0.01, ***P < 0.001 (two‐tailed Student's t test).

Source data are available online for this figure.

Figure EV3. Transfection efficiencies for miR‐16‐1‐3p and lnc266.

• A, B
  The expression of miR‐16‐1‐3p (A) and lnc266 (B). C3H10T1/2 cells were transfected with the plasmid expressing lnc266 and miR‐16‐1‐3p mimic by electroporation. Cells were then subjected to differentiation induction towards brown adipocytes. After 8 day‐induction, cells were harvested for gene expression analysis by qRT‐PCR. U6 or 18S rRNA were used as a house‐keeping gene.

Data information: Results are the mean ± SEM of three biological replicates; ***P < 0.001 (two‐tailed Student's t test). ns stands for no significance.

Discussion

In the present study, lnc266 was identified as a cold‐inducible lncRNA with promising effects on white fat browning and subsequent the thermogenic program. For the involved mechanism, lnc266 interacts with miR‐16‐1‐3p as sponges and thus alleviates its biological function. Of note, miR‐16‐1‐3p represses Ucp1 expression by binding to the 3′ UTR of Ucp1. By this way, lnc266 unravels the repression of miR‐13‐1‐3p on Ucp1 expression and promotes beige adipocyte differentiation toward brown‐like cells and stimulates the thermogenic program.

According to the sequencing data, only 1.2% RNAs belong to protein‐coding genes; the most majority of transcripts are noncoding RNAs (ncRNAs) (Scheideler, 2019). LncRNAs are a newly identified class of ncRNAs with the length above 200 nucleotides. Subsequently, a growing number of evidences have shown that lncRNAs hold multiple biological functions in various cellular events. As for adipogenesis, lncRNAs also play a key role for determining adipocyte differentiation (Alvarez‐Dominguez et al, 2015; Ding et al, 2018; Sun & Lin, 2019; Squillaro et al, 2020). Recently, a specific group of adipocytes in white adipose tissue was identified, which could be transformed into brown‐like adipocytes upon various external stimuli (Guerra et al, 1998; Ishibashi & Seale, 2010; Petrovic et al, 2010; Wu et al, 2012). These unique adipocytes are termed as beige adipocytes. In mice, adipocytes in iWAT exhibit the most characteristics of beige adipocytes and for this reason, iWAT was widely used for developing novel agents against obesity and obesity‐related metabolic syndromes (Sakers et al, 2022). In the present study, we also performed Microarray by using iWAT and found a number of differentially expressed lncRNAs. Of them, lnc266 caused our attention due to its acute response to thermogenic adaption under cold exposure. Moreover, lnc266 might be involved in systemic energy homeostasis as it was markedly reduced in obese mouse iWAT, eWAT, and BAT.

Following studies revealed that lnc266 stimulates preadipocyte differentiation into brown‐like adipocytes. In line with our findings, several other lncRNAs, including Blnc1, AK079912, lincRNA H19, lncRNA Ctcflos, and lncRNA Dio3os, also hold capacity for driving adipocyte differentiation toward brown adipocytes (Zhao et al, 2014; Schmidt et al, 2018; Xiong et al, 2018; Bast‐Habersbrunner et al, 2021; Chen et al, 2021). Of note, lnc266 could be reversely aligned to the 3′‐UTR of Kcnb1 gene. Furthermore, as well as lnc266, Kcnb1 expression in iWAT was increased by cold exposure. These results strongly suggest that lnc266 is an overlapping‐antisense lncRNA, which is transcribed from the same gene of Kcnb1. KCNB1 is a potassium channel, it was found to be upregulated in BAT as compared to WAT in mice (Baboota et al, 2015). Interestingly, it has been shown that potassium channel plays a key role in white fat browning and thermogenesis. For example, KCNK3 (potassium channel K3) was identified as a molecular marker that was highly enriched in UCP1‐positive human adipocytes, and it is required for beige adipocyte differentiation and thermogenic function (Shinoda et al, 2015). Kcnk3‐deficient mice became overweight, mainly because of an increase in WAT mass and BAT whitening (Pisani et al, 2016). Therefore, we predict that KCNB1 might hold similar functions in white fat browning and thermogenesis.

Consequently, these lncRNAs may have benefits against obesity by inducing brown adipocyte formation to consume excessive energy. Indeed, lincRNA H19 transgenic mice show reduced body weight gain and improved insulin sensitivity when mice were fed with high‐fat diet, these benefits are likely due to promoted energy expenditure (Schmidt et al, 2018). Increased expression of lncRNA Dio3os in BAT stimulates brown adipogenesis and thermogenesis and thus improves energy expenditure and glucose tolerance (Chen et al, 2021). However, in the present study, overexpression of lnc266 in iWAT had no effect on high fat‐induced body weight gain. We noticed that there are no data were presented with the in vivo effects of Blnc1, AK079912, and lncRNA Ctcflos on metabolic dysfunction, although all these lncRNAs were identified as a positive regulator for brown adipocyte differentiation and thermogenesis (Zhao et al, 2014; Xiong et al, 2018; Bast‐Habersbrunner et al, 2021). By considering lnc266 is a cold‐inducible lncRNA, we then placed mice treated with AAV‐lnc266 in a cold environment and found that lnc266 attenuates body weight gain and stimulates thermogenic gene expression in iWAT. Therefore, we predict that the function of lnc266 on brown adipocyte differentiation and thermogenic function requires cold habitation.

UCP1 is a proton channel located in the inner membrane of mitochondria. Its activation provides a path for protons to return to the matrix, thus uncoupling oxidative phosphorylation with ATP synthesis and dissipating chemical energy as heat (Rui, 2017). Hence, UCP1 is a specific marker for brown adipocytes. In the present study, both in vitro and in vivo data showed that lnc266 is a potent inducer for UCP1. However, overexpression of lnc266 in iWAT had no effects on systemic metabolic parameters in mice under normal room temperature, indicating the browning status in iWAT have minimal effects on systemic metabolism. The reason for these phenomena is likely due to lower levels of UCP1 in iWAT as compared to BAT. In a cold environment, some unique white adipocytes in iWAT can switch to brown‐like adipocytes, in which UCP1 expression was robustly induced (Harms & Seale, 2013). Under this condition, browning iWAT may hold an ability to affect systemic metabolic parameters. In line with this notion, manipulations of lnc266 expression iWAT have impacts on body temperature in mice with cold exposure.

As for the underlying mechanism for lnc266 induces Ucp1 expression, our data showed that lnc266 has no direct regulation on Ucp1 expression. It has been shown that lncRNAs play their functions by interacting with miRNAs as sponges and thus sequester miRNAs away from target mRNA (Schmitt & Chang, 2016). To explore the underlying mechanism for lnc266 induced UCP1 expression; therefore, we predicted the potential miRNAs targeted by lnc266 by bioinformatics analysis; meanwhile, we also predicted miRNAs with potential regulation on Ucp1. Combined these two clusters of miRNAs, we found that there are five miRNAs targeted by lnc266 with interaction with Ucp1. Following functional studies showed that miR‐16‐1‐3p is a substrate of lnc266, and it interacts with the 3′‐UTR of Ucp1 and represses UCP1. In another words, lnc266 interacts with miR‐16‐1‐3p and abolishes its repression on Ucp1, thus leading to upregulation of Ucp1. As a result, lnc266 drives adipocyte differentiation toward brown adipocytes and stimulates the thermogenic program. In line with our data, several previous studies also have shown that lncRNA sequester specific miRNAs away from their target mRNAs (Zhang et al, 2019; Cai et al, 2021; Lu et al, 2021). It is worthy to note that, most recently, it has been shown that cytosolic mitochondrial RNA (mtRNA) stimulates mitobiogenesis and thermogenesis by activating mitochondrion‐to‐nucleus signaling in adipocytes (Hoang et al, 2022). At this regard, in addition to miRNAs, it is possible that lnc266 targets mtRNAs to induce thermogenic response in adipocytes.

In summary, lnc266 was identified as a cold‐inducible lncRNA in iWAT, which promotes preadipocyte differentiation into brown‐like adipocytes with increased expression of thermogenic genes. Under room temperature, lnc266 has no effect on white adipocyte browning and energy consumption. In a cold environment, however, lnc266 increases core body temperature and reduces body weight gain in mice fed with high‐fat diet. Loss‐of‐function experiments further confirmed that lnc266 plays a key role in white fat browning. Mechanistically, lnc266 sponges miR‐16‐1‐3p and dismantles its repression on Ucp1, leading to enhanced expression of Ucp1. By this way, lnc266 promotes white adipocyte browning and thermogenic gene expression. These results deepen our understanding the reprogram of white adipocyte browning with cold exposure. Moreover, cold‐inducible lncRNAs in iWAT such as lnc266 might be a therapeutic target for combating obesity and related metabolic syndromes.

Materials and Methods

Biochemical reagents

The antibodies anti‐UCP1 and anti‐Actin were from Cell Signaling Technology (Beverley, MA). Oil Red O, insulin, T3, rosiglitazone, IBMX, indomethacin, dexamethasone, glucose, PMSF, leupeptin, aprotinin, okadaic acid, NaF, Na₄P₂O₇, EDTA, NP‐40, EGTA, Tris base, Triton X‐100, Bodipy, and bovine serum albumin (BSA) were from Sigma‐Aldrich (St. Louis, MO). Total RNA Extraction Reagent and cDNA Synthesis Kit were from Vazyme (Nanjing, China). FASTSTART ESSENTIAL DNA GREEN MASTER was from Roche (Indianapolis, IN). BCA Protein Assay Kit and Chemiluminescence blotting substrate were from ThermoFisher Scientific (Waltham, MA, USA). MicroRNAs and inhibitors were synthesized by RiboBio (Guangzhou, China). All other chemical reagents are analytical grade.

Plasmid construction and virus packaging

Adeno‐associated virus expressing lnc266 (AAV‐lnc266) was generated at HANBIO (Shanghai, China). Briefly, the fragment of lnc266 was amplified using primers 5′‐TAT CGT ACA TAT GTC GGT ACC AAG CTT GAT GAG CCT TTA GCA TCC TAC‐3′ (forward) and 5′‐ATC CAG AGG TTG ATT ATC GAT AAG CTT TCT CTC ATG GCG TCA GCT GAG‐3′ (reverse). The resulting lnc266 fragment was incorporated into the pHBAAV‐CMV‐MCS‐3flag‐EF1‐ZsGreen vector at the site of Hind III to produce pHBAAV‐lnc266. This shuttle vector pHBAAV‐lnc266, together with pAAV‐RC and pHelper, was co‐transfected into AAV‐293 cells by using Lipofiter (HANBIO, Shanghai, China) to produce AAV‐lnc266, which was then subjected to purification and titration. Control virus was generated with the same procedures except pHBAAV‐GFP was used to instead of pHBAAV‐lnc266. Adenovirus expressing lnc266 shRNA (ADV‐lnc266 shRNA) was generated by using BLOCK‐IT Adenoviral RNAi Expression System (Invitrogen) according to the manual instructions (Sun et al, 2014). The lnc266 shRNA sequences are CAC CGA TCA GGA AGA CAT GCC CGC GAA CGG GCA TGT CTT CCT GAT. Control virus ADV‐scramble shRNA was produced with the same protocol. Adeno‐associated virus expression lnc266 shRNA (AAV‐lnc266 shRNA) and control virus AAV‐scramble shRNA were constructed and produced at OBiO technology company (Shanghai, China). The virus vector is H12663 pAAV‐U6‐shRNA/spgRNA v2.0‐CMV‐EGFP‐WPRE (OBiO, Shanghai, China). The fragment of lnc266 was obtained from the plasmid of pHBAAV‐lnc266 using Hind III digestion, which was then subcloned into pcDNA3.1(+) vector (Invitrogen) at the site of Hind III. The plasmid expressing Prdm16 pcDNA‐PRDM16 was a gift from Bruce Spiegelman (Addgene plasmid #15503 http://n2t.net/addgene:15503; RRID: Addgene_15503) (Seale et al, 2007). The reporter plasmid −5.5 kb‐UCP1‐GFP was a gift from Irfan Lodhi (Addgene plasmid #104585; http://n2t.net/addgene:104585; RRID: Addgene_104585) (Lodhi et al, 2017).

Cell culture and treatment

C3H10T1/2 cells (Cat. No. CRL‐226) and HEK 293T cells (Cat. No. CRL‐3216) were from ATCC (Manassas, VA, USA). Cells were cultured in DMEM (Hyclone) containing 10% fetal bovine serum (Gibco) and 1% penicillin–streptomycin (Gibco) and incubated at 37°C in a humidified atmosphere with 5% CO₂. Transfection in C3H10T1/2 cells was accomplished by electroporation with an apparatus (NEPA21, NEPA GENE, Japan). The procedures for adipocyte differentiation induction were described elsewhere (Wang et al, 2019). Briefly, confluent C3H10T1/2 cells were cultured in DMEM containing 20 nM insulin, 1 μM rosiglitazone, 1 nM 3,3′, 5‐triiodo‐L‐thyronine (T3), 0.5 mM 3‐isobutly‐1‐ethylxanthine (IBMX), 125 μM indomethacin and 1 μM dexamethasone. 2 days later, the medium was changed to DMEM containing 20 nM insulin, 1 nM T3 and 1 μM rosiglitazone for 8–10 days to accomplish adipocyte induction, during which culture medium was changed every 2 days. For white adipocyte differentiation, confluent cells were cultured in DMEM containing 10 μg/ml insulin, 1.0 μM rosiglitazone, 0.5 mM 3‐isobutly‐1‐ethylxanthine (IBMX), and 1.0 μM dexamethasone. 2 days later, and the medium was changed to DMEM containing 10 μg/ml insulin for 8–10 days to accomplish adipocyte induction, during which culture medium was changed every 2 days. 293T cells were transfected with plasmids using lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions.

Knockdown of lnc266 by siRNAs

To knockdown lnc266 expression, three pairs of siRNAs against lnc266 were synthesized at RiboBio (Guangzhou, China) with the order number PA20230320027. The targeting sequences are: CGC AGC TAT GGA AAC AAG A (siRNA‐1), CGC CGT AAC TAG AGC ATT T (siRNA‐2), and GAC AGC TTC TCA GAC TAA A (siRNA‐3). C3H10T1/2 cells were transfected with lnc266 siRNAs by electroporation with an apparatus (NEPA21, NEPA GENE, Japan). Control cells were transfected with negative control (NC) siRNA (RiboBio, Guangzhou, China).

Lipid droplet and mitochondrial staining

Lipid droplets in differentiated adipocytes were stained with Oil Red O or Bodipy. Mitochondrial staining was performed by using mitochondrial cytopainter (Abcam, ab112145). Cell morphology was observed, and images were taken with a fluorescent microscope (Olympus, IX73).

Animals

All experimental animals were purchased from Experimental Animal Center of Nantong University. Animals were housed in the poly‐acrylic cages with 22–24°C, 55–60% relative humidity, and 12 h light/dark cycle. Mice had free access to food and drinking water. 6‐week‐old male C57BL/6J mice (RRID: IMSR_JAX:000664) fed on a high‐fat diet (HFD; Research Diet, D12451) for consecutive 60 days and AAV‐con or AAV‐lnc266 was in situ injected into bilateral iWAT. Control mice were fed with standard diet (SD; SWC9101; Xietong Biotech, Nanjing, China). To knockdown lnc266 expression, AAV‐lnc266 shRNA was in situ injected into bilateral iWAT, control mice were received the same dose of AAV‐scramble shRNA. For virus injection, mice were anesthetized with isoflurane and the small area in the flanks, proximal of hip joints was shaved and cleaned with ethanol. A 0.5–0.8 cm incision in the skin was created with surgical scissors to expose iWAT. Virus (1.0 × 10¹¹ vg/mouse) was directly injected into five distinct spots in the fat pad using a syringe (Balkow et al, 2016). For cold exposure, mice were cold‐acclimated for 3 days, and then mice were placed in cold‐room (4°C) for 4, 8, and 16 days. Mice in control group were housed at 22–24°C. The animal protocols were approved by the Animal Care and Use Committee of Nantong University and the Jiangsu Province Animal Care Ethics Committee (Approval ID: SYKX (SU) 2017‐0046).

Body weight, food intake, and core body temperature assays

For mice housed at 22–24°C, body weight, and food intake were measured every three days. Core body temperature was measured rectally with a thermistor (Micro‐Therma 2T/ThermoWorks) during the light cycle once a week. For cold‐exposed mice, body weight, food intake, and core body temperature were measured every day.

Sample preparation

Blood was taken from eyehole using capillary tubes and plasma was divided by centrifugation at 2,400 g for 20 min at 4°C. iWAT, BAT, eWAT, and liver were removed and frozen in liquid nitrogen instantly. Samples were stored in −80°C freezer until use.

RNA extraction and quantitative real time PCR (qRT‐PCR)

Total RNA extraction reagent RNA isolator (Vazyme) was used to RNA extraction from cells or animal tissues and transcribed into cDNA using a cDNA Synthesis Kit (Vazyme). Cytoplasmic and nuclear RNA samples were prepared with a Nuclear or Cytoplasmic RNA Purification Kit (Fisher BioReagents). The resulting RNA was used as templates for cDNA synthesis. The levels of mRNA were analyzed with iQ5 Multicolor Real‐Time PCR Detection System (Bio‐Rad) with FASTSTART ESSENTIAL DNA GREEN MASTER (Roche). The mRNA levels were normalized to the expression of 18S rRNA. The primer sequences used in this study were listed in Appendix Tables S2 and S3. The cycle threshold values of 18S rRNA were presented in Appendix Table S4.

Tissue protein extraction

Tissues were homogenized with a homogenizer in ice‐cold lysis buffer, which was constituted by 25 mM Tris–HCl, pH 7.4; 50 mM Na₄P₂O₇; 100 mM NaF; 10 mM Na₃VO₄; 10 mM EDTA; 10 mM EGTA; 1% NP‐40; 10 μg/ml Leupeptin; 10 μg/ml Aprotinin; 20 nmol/l okadaic acid; and 2 mmol/l PMSF (phenylmethanesulfonyl fluoride). After homogenization, lysates were rotated for 20 min at 4°C and then subjected to centrifugation at 13,800 g for 20 min at 4°C. The lipid layer was removed, and the supernatant was transferred into Eppendorf tubes. Protein concentration was quantified by using a Protein Assay Kit (ThermoFisher Scientific). Equivalent protein concentration in each sample was prepared and boiled at 100°C for 5 min in 1× Loading buffer. Lysates were cooled to room temperature before western blot analysis.

Western blot analysis

Western blot analysis was described previously (Sun et al, 2014). Samples from tissue or cell lysates were resolved by SDS‐PAGE and then transferred to PVDF membrane. After 1 h blocking at room temperature using 10% blocking reagent (Roche), membrane was incubated overnight with primary antibodies including anti‐UCP1 (1:1,000; Abcam; ab23841; RRID: AB_2213764) and anti‐Actin (1:5,000; Abcam; ab8226; AB_306371) in Tris‐buffered saline solution/Tween (TBST) containing 10% blocking reagent at 4°C. After the incubation, membranes were washed three times in TBST and incubated with secondary antibody (1:10,000; Abcam) for 1 h at room temperature. After three‐time washing in TBST, membranes were developed using a chemiluminescence assay system (Roche) and exposed to Kodak film. Relative protein levels were quantified by Image J program, and the bands of interest were normalized to Actin.

Histochemistry and immunostaining

Adipose tissues were removed and washed with saline and then immediately placed in 4% paraformaldehyde and incubated for 24 h at 4°C. The fixed tissues were embedded in paraffin for preparing paraffin sections after dehydration and paraffin infiltration. 5 μm thickness sections were used for hematoxylin and eosin (H&E) staining and immunohistochemistry analysis. For immunohistochemistry, adipose tissue sections were antigenically repaired with sodium citrate buffer and incubated with 10% bovine serum at 37°C for 30 min. Then sections were incubated with anti‐UCP1 antibody (Abcam; ab10983; RRID: AB_2241462) overnight at 4°C. Next, sections were incubated with biotinylated antimouse IgG for 2 h at 37°C. Color reaction was initiated by adding 3,3′‐diaminobenzidine tetrahydrochloride substrate. Images were viewed under a microscope (Zeiss, Axio Image M2). Size of adipocyte was analyzed by the Adiposoft (Galarraga et al, 2012).

Triglyceride assay

Total lipid extraction from the liver was performed with a previous method (Bligh & Dyer, 1959). Triglyceride was analyzed by a Serum Triglyceride Determination Kit (Sigma‐Aldrich; TR0100) according to its manual instructions. Liver true glyceride was obtained by subtracting the initial liver glycerol from the glycerol that was created by the enzyme in the assay.

Microarray analysis

Affymetrix Whole Transcript microarray was conducted at Gminix (Shanghai, China). Briefly, total RNA of mouse iWAT was extracted and treated with RNase‐free DNase I to remove genomic DNA contamination. cDNA was then synthesized and amplified. The fragmented and biotin‐labeled cDNA was hybridized with the Affymetrix Mouse Transcriptome Array 1.0 (Mus musculus mRNA, lncRNA) and then washed, stained, and scanned according to the Affymetrix protocol. Hybridization data were obtained by processing with the Affymetrix Expression Console Software. The median of NUSE (normalized unscaled standard error) value and RLE (relative log expression) value of each chip were used as the evaluation criteria for the feasibility of scheme design and the reliability of results. MAS 5.0 was used to evaluate the probe detection rate to determine the gene expression. The robust multichip average (RMA) algorithm was used to calculate the expression value of the probe set. Significance analysis of microarrays (SAM) was used to screen genes with significant differences. The differentially expressed genes were ranked according to the misclassification rate q‐value and d‐score of the differences.

Luciferase reporter assay

To test the direct effect of lnc266 on Ucp1 expression, the plasmids of pGL4.10[luc2] and pGL4.74[hRluc/TK] (Promega) were used to construct the dual luciferase reporter gene system. Ucp1 promoter was inserted into pGL4.10 to construct pGL4.10‐Ucp1 plasmid. 293T cells were co‐transfected with 300 ng of plasmid carrying lnc266, 100 ng of pGL4.10‐Ucp1, and 25 ng of pGL4.74. 24 h post‐transfection, cells were lysed and subjected to luciferase activity assay using a Dual Luciferase Reporter Assay Kit (Vazyme). To detect interaction between lnc266 and microRNAs, pmirGLO Dual‐Luciferase miRNA Target Expression Vector (Promega) was used to construct the dual luciferase reporter gene system. Ucp1 3′UTR, Ucp1 3′UTR‐Mut, and lnc266 were inserted into pmirGLO to construct pmirGLO‐Ucp1 3′UTR, pmirGLO‐Ucp1 3′UTR‐Mut, and pmirGLO‐lnc266 plasmids, respectively. 293T cells were cultured in 24‐well plates. When cells at 70–80% confluence, 293T cells were co‐transfected with 50 nM of pmirGLO luciferase vector (pmirGLO‐Ucp1 3′UTR, pmirGLO‐Ucp1 3′UTR‐Mut, or pmirGLO‐lnc266) and 50 nM of miR‐16‐1‐3p or mutated miR‐16‐2‐3p. 48 h post‐transfection, cells were lysed and subjected to luciferase activity assay.

Fluorescence in situ hybridization (FISH)

Fresh iWAT was incubated in 10% calcium formaldehyde solution (R23047; Yuanye Biotech, Shanghai, China) for 24 h to prepare frozen sections. The prepared sections were then subjected to denaturation, hybridization, elution, and signal amplification. After sealing, signals for lnc266 were observed by a microscope (Zeiss, Axio Image M2). For fluorescence in situ hybridization in cells, C3H10T1/2 cells were planted on microscope cover glass (Fisher Scientific). FISH was performed with Fluorescent In Situ Hybridization Kit (RiboBio, Guangzhou, China). Lnc266 probe was generated using primers (CCC GTC ACT GTT CTG AGG ACA (forward); AGC CAC ACT CCT AGG GCA TA (reverse)) with a DIG‐RNA Labeling Kit (Roche). U6 and 18S rRNA probes were provide by the kit.

RNA pulldown

The procedures for RNA pulldown were described previously (Grelet et al, 2017). Briefly, wide type or mutated lnc266 was cloned into the pGEM‐T Easy Vector (Promega), which were used as templates in in vitro transcription. Biotin‐labeled wild‐type lnc266 (lnc266‐wt) and mutated lnc266 (lnc266‐mut) were generated by using a RNAmax‐T7 biotin‐labeled transcription kit (RiboBio) according to the manufacture's instructions. The biotin‐labeled RNA probes were then incubated with streptavidin‐Dyna beads (Invitrogen). Excess free probes were removed using wash buffer. The resulting biotin‐Dyna bead complex was incubated with total cell lysates prepared from C3H10T1/2 cells for 3–4 h at 4°C. RNA was released from beads by addition of RNA isolator (Vazyme) and subjected to miRNA‐specific qRT‐PCR analysis (RiboBio). Alternatively, released RNA samples were analyzed by PCR, and the resulting products were resolved by electrophoresis in agarose gel. The used primers for lnc266 are CCC GTC ACT GTT CTG AGG AC (forward) and TGG TCA GCC TGT CTC TGG TA (reverse).

Inhibition of miR‐16‐1‐3p

MiR‐16‐1‐3p inhibitor (UCA GCA GCA CAG UCA AUA CUG G) was synthesized at RiboBio (Guangzhou, China) with the order number PA20211114003. To inhibit miR‐16‐1‐3p, C3H10T1/2 cells were transfected with miR‐16‐1‐3p inhibitor by electroporation with an apparatus (NEPA21, NEPA GENE, Japan). Control cells were transfected with negative control (NC).

Experimental design and statistical analysis

Most experiments were performed at least twice, biological replicates were indicated in the relevant figure legends. No statistical methods were used for sample size estimate. No data were excluded, and no blinding mode was adopted for data collecting. Data are presented as means ± standard error of the mean (SEM). Comparisons between two groups were analyzed by two‐tailed Student's t test. Statistical significance for multiple groups was calculated with one‐ or two‐way ANOVA with Bonferroni's post hoc test. Statistical significance was accepted at P < 0.05.

Author contributions

Jingyu Ma: Data curation; formal analysis; investigation; methodology; writing – original draft. Yuting Wu: Data curation; formal analysis. Lixue Cen: Formal analysis; investigation. Zhe Wang: Investigation; methodology. Ketao Jiang: Investigation. Bolin Lian: Conceptualization; funding acquisition; project administration; writing – review and editing. Cheng Sun: Conceptualization; supervision; funding acquisition; project administration; writing – review and editing.

Disclosure and competing interests statement

The authors declare that they have no conflict of interest.

Supporting information

Appendix

Expanded View Figures PDF

PDF+

Source Data for Figure 1

Source Data for Figure 2

Source Data for Figure 3

Source Data for Figure 4

Source Data for Figure 5

Source Data for Figure 6

Source Data for Figure 7

Source Data for Figure 8

Source Data for Figure 9

EMBO reports (2023) 24: e55467

Data availability

Datasets generated for this study are deposited in ArrayExpress with the accession code E‐MTAB‐13074 (https://www.ebi.ac.uk/biostudies/arrayexpress/studies/E‐MTAB‐13074).