C57BL/6N mice were obtained from Jackson laboratory (United States). Male mice at 2–3 months of age were maintained at 4°C for 24 h (acute cold exposure, AC) or 10 days (chronic cold exposure, CC). After cold exposure, samples were obtained or animals were returned to room temperature for the deacclimation condition. Deacclimation groups were maintained at room temperature (22°C) for 24 h (acute deacclimation, AD) or 1 week (chronic deacclimation, CD) after previous exposure to 4°C for 10 days. When indicated, deacclimating mice were subjected to daily intraperitoneal (i.p.) injections with hydroxychloroquine [Chlor; 50 mg/kg body weight (b.w.)] to inhibit autophagy or PBS as a control. When indicated, leupeptin (40 mg/kg b.w.) was administered i.p. 1 h before sacrifice to block autophagy. The control group (CT) are animals in standard housing conditions, at 22°C. Since the difference in the age of the animals at the time of sacrifice was only ±7 days, a single control group was used for comparison with the distinct temperature environment conditions. CT animals were sacrificed at the same time than the CD group, and the weight and food intake of the animals were monitored for the duration of the experiments. Animals were housed two per cage, and b.w. and food intake were monitored during all procedures. Mice were sacrificed by cervical dislocation. Tissues were dissected and frozen for further analysis.

Animals were anesthetized with 1.5% isoflurane and cardiac parameters were assessed by echocardiography with a VividQ instrument (GE Healthcare, Piscataway, NJ, United States) equipped with a 12-MHz microprobe. Ventricular measurements in M-mode and Doppler were made the day before sacrifice. Three different cardiac cycles were measured for each assessment, and average values were obtained. Analyses of echocardiographic images were performed by two different observers in a blinded manner.

After extraction, each heart was cut transversely at mid-height and the superior half was fixed in 4% formaldehyde, embedded in paraffin, and sectioned. The sections were deparaffinized and stained with hematoxylin and eosin (H&E), and posteriorly applied for the determination of cardiomyocyte size. Quantification of the areas was assessed in perpendicular cuts of the posterior left ventricular wall. Only cells in perpendicular plane were considered for the analysis. The quantification of the areas was performed with the Open Source software ImageJ, using the ROI manager tool.

Fibrosis was determined by Masson’s trichrome staining (Panreac, Barcelona, Spain). Images were obtained under an Olympus BX61 microscope.

RNA extraction from tissue was performed using TriPure reagent (Roche, Indianapolis, IN, United States). Reverse transcription (RT) was performed with 0.5 μg of total RNA in a reaction volume of 20 μl, using a High Capacity RNA-to-cDNA kit (Applied Biosystems, Foster City, CA, United States). mRNA expression was determined by real-time PCR using TaqMan Gene Expression Assays (ThermoFisher Scientific, Waltham, MA, United States). The final reaction mix (25 μl) contained 1 μl cDNA, 12.5 μl TaqMan Universal PCR Master Mix (ThermoFisher Scientific), 250 nM probes, and 900 nM primers from the Assays-on-Demand Gene Expression Assay Mix or the Assays-by-Design Gene Expression Assay Mix (ThermoFisher Scientific). Each sample analysis was conducted in duplicate for increased accuracy, and the mean value was used to calculate the expression of the gene of interest and the reference gene (Cyclophilin A, Ppia). The mRNA levels of each gene of interest were normalized to Ppia using the comparative (2^−ΔCT) method.

Whole-cell lysates or tissue homogenates were obtained in ice-cold RIPA buffer containing a protease inhibitor cocktail (Rupérez et al., 2018). Proteins were resolved by 12% SDS-PAGE and transferred to Immobilon-P membranes (Millipore). Protein detection was performed using antibodies against LC3B (Cell Signaling, Beverly, MA, United States; #2775), p62 (Cell Signaling; #5114), Atg7 (Cell Signaling; #2631) and Parkin (Cell Signaling; #2132). Ponceau staining and GAPDH (G9545, Sigma-Aldrich) were used as a loading control.

Cardiac samples were fixed in 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4), and postfixed in 1% osmium tetroxide and 0.8% FeCNK in phosphate buffer. After dehydration in a graded acetone series, tissue samples were embedded in Spurr resin. Ultrathin sections were stained with uranyl acetate and lead citrate, and examined with a Jeol 1010 transmission electron microscope (Izasa Scientific, Barcelona, Spain).

Cell culture experiments were conducted in triplicate on at least three independent cardiomyocyte isolations. Groups of five mice were used for the in vivo experiments. Results are presented as means ± SEM. Statistical analysis was done with one-way ANOVA or two-way ANOVA followed by post-hoc tests as appropriate, using the GraphPad Prism software (GraphPad Software Inc., San Diego, CA, United States). A p-value less than 0.05 was considered statistically significant. One, two, and three symbols denote p < 0.05, p < 0.01, and p < 0.001, respectively.

All animal procedures performed conform to the guidelines from Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes and approved by the Institutional Animal Care and Use Committee of the University of Barcelona.

Three-month-old mice maintained at 21°C were subjected to acute cold (AC; 4°C, 24 h), chronic cold (CC; 4°C, 10 days), or 10 days of cold exposure followed by acute deacclimation (AD; 21°C, 24 h) or chronic deacclimation (CD; 21°C, 7 days) at room temperature. Thereafter, hearts were analyzed (Figure 1A). We found that chronic cold exposure significantly increased the heart weight/tibia length (HW/TL) ratio (a measure of cardiac hypertrophy), whereas mice subjected to cold followed by chronic deacclimation showed a HW/TL ratio similar to that of non-cold-challenged mice (Figure 1B), indicating reversion. No changes in HW/TL ratio were observed in acute conditions (AC or AD). Histological examination of hematoxylin and eosin (H&E)-stained left ventricle tissue sections revealed that the cardiomyocyte cross-sectional area (CSA) was significantly increased in mice after cold exposure, indicating cellular cardiac hypertrophy (Figure 1C). In contrast, the CSA of mice subjected to cold exposure followed by deacclimation was significantly decreased from the post-cold-exposure level to that seen in non-cold-exposed mice.

Next, we assessed the mRNA expression levels of the hypertrophy marker genes encoding atrial natriuretic factor (Nppa) and α-actinin (Acta1) in the hearts of mice subjected to AC, CC, AD, and CD (Figure 1D). We found that CC mice exhibited significantly increased Nppa expression levels, and deacclimation caused Nppa expression to progressively normalize to levels similar to those seen in non-cold-exposed mice. Both AC and CC increased the transcript levels of Acta1, although the increase was statistically significant only for AC. Deacclimation significantly decreased these upregulations of Acta1 mRNA expression.

Finally, we performed echocardiographic measurements to determine cardiac function in these animals (Table 1). We found that the interventricular septum (IVS) and left ventricular internal diameter (LVID) in both systole and diastole were significantly increased in the CC group, indicating the development of cardiac hypertrophy. No difference was found in the left ventricular posterior wall dimensions (LVPW). During deacclimation, IVSd and LVPWd in diastole, and LVPWs in systole were significantly reduced compared with those seen in cold-exposed mice, indicating reversion of the cold-induced hypertrophic process. Finally, the measures of cardiac function, LV ejection fraction (EF) and fractional shortening (FS), were significantly decreased by cold exposure and significantly recovered after deacclimation.

Based on these findings, we concluded that chronic cold exposure for 10 days leads to development of cardiac hypertrophy, and that this process is reversed when mice are returned to room temperature.

Given that there is a close association between impaired fatty acid oxidation (FAO) and cardiac hypertrophy (Planavila et al., 2011; Rupérez et al., 2021), we next examined the expression levels of FAO-related genes and that encoding the glucose transporter, Glut1 (Figure 2A). We found that the transcript levels of the FAO-related genes encoding pyruvate dehydrogenase kinase-4 (Pdk4) and medium-chain acyl-CoA dehydrogenase (Acadm) were significantly decreased in mice subjected to chronic cold and reversed after chronic deacclimation back to room temperature. We did not find any significant difference for the FAO-related gene encoding carnitine palmitoyltransferase (Cpt1b). Conversely, Glut1 was significantly increased in mice after prolonged cold exposure, and this alteration was reverted after deacclimation. These data indicate that during chronic cold, the cardiac FAO machinery is reduced but glucose uptake is increased in the myocardium, and these changes are partly reversed after deacclimation from a cold environment to a warmer one.

Next, we assessed cardiac fibrosis by performing Masson’s trichrome staining and evaluating the transcript expression levels of fibrosis-related genes encoding collagen 3 (Col3a1) and transforming growth factor-β (Tgfb1) (Figures 2B,C). Quantification of fibrotic areas showed that there was less fibrosis during deacclimation than under chronic cold, although the difference was not statistically significant (Figure 2B). However, the expression levels of fibrotic genes Col3a1 and Tgfb1 were increased after chronic exposure to cold and significantly reduced back to pre-cold levels after deacclimation (Figure 2C).

To determine the molecular mechanism responsible for the remarkable cardiac plasticity seen in response to variation in environmental temperature, we next analyzed the autophagic process. Autophagy, which is a molecular mechanism responsible for the degradation of intracellular components, has been related to cardiac hypertrophy (Kaludercic et al., 2020). First, to determine autophagic activity, we examined the protein levels of phosphatidylethanolamine-conjugated microtubule-associated protein one light chain three beta (LC3bI and LC3bII) and the substrate of autophagic degradation, p62 (Figure 3A). Immunoblotting analyses showed that cold exposure (AC and CC) significantly reduced the level of LC3bII and CC increased the level of p62, suggesting that autophagy is reduced under cold exposure. During deacclimation (AD and CD), the level of LC3bII was unchanged, but that of p62 was dramatically reduced after AD and fully recovered after CD. Collectively, these data suggest that sustained cold exposure blocks autophagy and deacclimation appears to reactivate this process fairly quickly.

To confirm these findings, we treated mice with the autophagy inhibitor, leupeptin (Leup) just before sacrifice, and analyzed the autophagic flux (Figure 3B). We found that the LC3bII protein level was not altered by leupeptin treatment of cold-exposed mice, indicating that cold exposure did not trigger autophagic activation. However, LC3bII was accumulated in leupeptin-treated mice subjected to acute deacclimation, compared to leupeptin-untreated mice. This indicates that the autophagic flux was activated after 24 h of deacclimation from cold back to room temperature.

Finally, we used transmission electron microscopy to examine the subcellular structures of heart sections from mice subjected to cold and deacclimation, with or without leupeptin treatment (Figure 3C). In mice subjected to chronic cold, we observed very few degradation vesicles whose structures were consistent with that of an autophagosome. In contrast, we observed numerous autophagosome-like degradation vesicles in mice subjected to acute deacclimation, especially in those treated with leupeptin. This is consistent with activation of the autophagic flux.

Collectively, these results indicate that autophagy is blocked after prolonged cold exposure, but reactivated at 24 h after the mice are returned to room temperature.

In order to determine whether activation of autophagy during deacclimation from cold is responsible for the reversion of the hypertrophic process, we subjected 2-month-old mice to a daily intraperitoneal (i.p.) injection with the autophagy inhibitor hydroxychloroquine (Chlor) for 1-week in order to fully inhibit autophagy at the time of deacclimation (Figure 4). To confirm the effect of Chlor, we first analyzed the protein levels of LC3b and p62 (Figure 4A). In animals subjected to deacclimation and treated with Chlor, the levels of both LC3bII and p62 were higher (statistically significant in the case of p62) than in deacclimated mice treated with PBS, indicating that the autophagic flux was effectively inhibited by Chlor. The protein levels of Parkin were significantly reduced in deacclimated mice treated with PBS and the protein levels of Atg7 were unchanged. To further confirm inhibition of the autophagic process, we also analyzed the expression levels of genes involved in this process, such as Atg7, Ulk1, Lc3b, and Parkin (Figure 4B). We found that the mRNA levels of Atg7, Lc3b, and Parkin tended to decrease during deacclimation from cold, compared to those seen in mice under cold exposure. In deacclimated mice treated with Chlor, in contrast, the expression levels of Atg7, Lc3b, and Parkin were increased compared to those in Chlor-untreated deacclimated mice (statistically significant for Atg7 and Parkin). No change was observed for Ulk1. The increased levels of autophagic genes in Chlor-treated mice indicates a compensatory mechanism of the cell in order to respond to the autophagy blockade. These data further confirm that Chlor treatment successfully blocked autophagy in our system.

Next, we analyzed hypertrophy in the hearts of mice treated as described above (Figure 5). We found that the HW/TL ratio tended to be higher in deacclimated mice treated with Chlor than in non-Chlor-treated deacclimated mice, although this difference was not statistically significant (Figure 5A). When we analyzed the area of cardiomyocytes in these hearts, we found that mice subjected to deacclimation after previous cold exposure had significantly larger cardiomyocyte areas following Chlor treatment, compared to those in non-Chlor-treated mice (Figure 5B). We also assessed the transcript levels for markers of hypertrophy and fibrosis (Figure 5C). As expected, the mRNA level of the hypertrophy marker, Nppa, was significantly reduced in non-Chlor-treated deacclimated mice but not in Chlor-treated deacclimated mice, as compared to mice under cold exposure. No change was observed in the fibrotic markers, Col3a1 and Tgfb1, between Chlor-treated versus non-Chlor-treated CD mice. Echocardiographic measurements showed that IVSd, LVPWd, LVPWs, EF, and FS were significantly reduced during deacclimation from cold in non-Chlor-treated deacclimated mice but not in Chlor-treated deacclimated mice, compared to cold-exposed mice (Table 1). Collectively, these data indicate that autophagy is involved in the deacclimation-induced reversion of the cold-induced hypertrophic phenotype.