Depot-Specific White Adipose Tissue Remodeling Supports Non-Thermogenic Metabolic Homeostasis During Shallow Hibernation in Raccoon Dogs

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The study investigates how white adipose tissue (WAT) in raccoon dogs supports non-thermogenic metabolic homeostasis during shallow hibernation, specifically examining depot-specific remodeling (back-fat vs. tail-fat) in the absence of significant brown adipose tissue (BAT) activity.

The first detailed transcriptomic and histological characterization of depot-specific WAT remodeling in a shallow-hibernating canid.

A clear demonstration that WAT can assume a central metabolic role in the absence of BAT activation, offering a valuable comparative model for non-thermogenic hibernation.

Overall methodology is sound, but the following improvements are suggested:

1.The methods section does not explain why only male animals were selected. Please provide a justification for the use of a single sex.

2.The ethics approval number for animal experiments is not provided.

The figures and tables are generally of high quality. However, the following minor improvements are suggested:

Figure 3D: The HSP90 loading control in the Western blot shows some variability between samples. Please provide quantitative bar graphs along with statistical comparisons.

Minor revision

1.The methods section does not explain why only male animals were selected. Please provide a justification for the use of a single sex.

2.The ethics approval number for animal experiments is not provided.

3.Figure 3D: The HSP90 loading control in the Western blot shows some variability between samples. Please provide quantitative bar graphs along with statistical comparisons.

4.Please standardize the use of “back- and tail-fat” versus “back-fat and tail-fat” throughout the manuscript.

Author Response

Point-by-point response to reviewers

Reviewer #1

Comments to Author

The study investigates how white adipose tissue (WAT) in raccoon dogs supports non-thermogenic metabolic homeostasis during shallow hibernation, specifically examining depot-specific remodeling (back-fat vs. tail-fat) in the absence of significant brown adipose tissue (BAT) activity. The first detailed transcriptomic and histological characterization of depot-specific WAT remodeling in a shallow-hibernating canid. A clear demonstration that WAT can assume a central metabolic role in the absence of BAT activation, offering a valuable comparative model for non-thermogenic hibernation. Overall methodology is sound, but the following improvements are suggested: The methods section does not explain why only male animals were selected. Please provide a justification for the use of a single sex. The ethics approval number for animal experiments is not provided. The figures and tables are generally of high quality. However, the following minor improvements are suggested: Figure 3D: The HSP90 loading control in the Western blot shows some variability between samples. Please provide quantitative bar graphs along with statistical comparisons.

Minor revision

1.The methods section does not explain why only male animals were selected. Please provide a justification for the use of a single sex.

Response 1: We thank the reviewer for this important comment. We agree that the rationale for using a single sex should be clearly stated. In the revised manuscript, we have added an explanation in the Materials and Methods section. Only male raccoon dogs were used to minimize potential confounding effects associated with seasonal reproductive cycles and sex hormone fluctuations, which may influence adipose tissue metabolism, immune status, and hibernation-related physiological responses. We have revised the Materials and Methods section accordingly (Page 11, Lines 351–353).

2.The ethics approval number for animal experiments is not provided.

Response 2: We thank the reviewer for pointing this out. In the revised manuscript, we have provided the full animal ethics approval information, including the approving committee, protocol code, and approval date. The animal study protocol was approved by the Ethics Committee of Jilin Agriculture University (protocol code 20250319001 and date of approval 19 March 2025). We have revised the Materials and Methods section and the Institutional Review Board Statement accordingly (Page 11, Lines 361–363; Page 14, Lines 473–475).

 

3.Figure 3D: The HSP90 loading control in the Western blot shows some variability between samples. Please provide quantitative bar graphs along with statistical comparisons.

Response 3: We thank the reviewer for this helpful suggestion. We have now added densitometric quantification and statistical comparisons for the Western blot data in Figure 3D. For each sample, the band intensity of mitochondrial respiratory chain complex proteins was normalized to the corresponding HSP90 loading control. The revised Figure 3D now includes quantitative bar graphs showing the relative protein abundance of CV-ATP5A, CIII-UQCRC2, and CI-NDUFB8 in back-fat and tail-fat during autumn and winter. These quantifications support the downregulation of oxidative phosphorylation-related proteins during winter sleep. We have also revised the Results and figure legend accordingly (Page 5-6, Lines 180–185, Lines 189–190; Figure 3D).

4.Please standardize the use of “back- and tail-fat” versus “back-fat and tail-fat” throughout the manuscript.

Response 4: We thank the reviewer for pointing this out. We have carefully checked the entire manuscript and standardized the terminology throughout the text, figure legends, and abstract. We now consistently use “back-fat and tail-fat” throughout the manuscript, including the revised abstract, Results, figure legends, Discussion, and Materials and Methods.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The background section suggests coordinated roles of BAT and WAT during hibernation. Based on this, one might expect winter beiging in raccoon dogs, especially they have low BAT activity. However, although the morphology appears beiging-like, the molecular data do not support beiging. The data are interesting, but I have several questions. I study BAT in mice and humans, but not raccoon dogs. Since the readership of IJMS is not limited to raccoon dog specialists, I would appreciate clarification of the following points.

 

Line 121: WAT from the “back” and “tail” regions
Please clarify why these fat regions were selected, around line 121 and/or in the Methods section.

 

Fig. 2A–D:
The morphology appears consistent with typical beiging, such as smaller multilocular lipid droplets. Therefore, I do not understand why beiging markers, including UCP1, were downregulated by approximately 80% compared with autumn. Is this commonly observed in other animals or previous studies? If not, immunohistology would ideally be needed. If such data are not available, please discuss this discrepancy in more depth.

 

Fig1 Line 82

I wonder whether the timing of sample collection is important for interpreting the data, especially gene expression data. For example, samples were collected after approximately 50% body weight loss. If this is important, please clarify it in Fig. 1 or around line 82.

 

Fig. 1E:
To me, the interscapular region appears to contain activated BAT. Could you clarify whether this conflicts with the statement in the Introduction?

 

Fig. 2H:
Does this Venn diagram show genes regulated in the same direction? If not, please replace it with a Venn diagram showing only commonly regulated genes in the same direction. Separating the data into two figures, one for downregulated genes and one for upregulated genes, would also help. Even if this disrupts the connection to Fig. 3, please provide the numbers in the text.

 

Lines 165–168:
These statements are somewhat broad and may be overinterpreted.

 

Lines 330–337:
Was the order of tissue collection randomized? If not, please state this in the limitations section.

Author Response

Reviewer #2

Comments to Author

The background section suggests coordinated roles of BAT and WAT during hibernation. Based on this, one might expect winter beiging in raccoon dogs, especially they have low BAT activity. However, although the morphology appears beiging-like, the molecular data do not support beiging. The data are interesting, but I have several questions. I study BAT in mice and humans, but not raccoon dogs. Since the readership of IJMS is not limited to raccoon dog specialists, I would appreciate clarification of the following points.

1. Line 121: WAT from the “back” and “tail” regions
   Please clarify why these fat regions were selected, around line 121 and/or in the Methods section.

Response 1: We thank the reviewer for this helpful suggestion. We agree that the basis for selecting these two WAT depots should be clearly explained. In the revised manuscript, we have added a brief explanation before presenting the histological and transcriptomic analyses. Back-fat was selected because it represents a classical subcutaneous WAT depot, whereas tail-fat represents a distinct caudal fat depot. Since tail-associated fat storage has been reported in several mammals as an adaptive energy reserve under harsh environmental or nutritional stress (Xu et al., 2023), we compared these two anatomically and functionally distinct depots to investigate depot-specific adipose adaptations during winter sleep. We have revised the Results section accordingly (Page 4, Lines 124–129).

Reference:

Xu, Y. X., Wang, B., Jing, J.-N., Ma, R., Luo, Y.-H., Li, X., Yan, Z., Liu, Y.-J., Gao, L., Ren, Y.-L., et al. (2023). Whole-body adipose tissue multi-omic analyses in sheep reveal molecular mechanisms underlying local adaptation to extreme environments. Communications Biology 6, 159.

2. Fig. 2A–D:
   The morphology appears consistent with typical beiging, such as smaller multilocular lipid droplets. Therefore, I do not understand why beiging markers, including UCP1, were downregulated by approximately 80% compared with autumn. Is this commonly observed in other animals or previous studies? If not, immunohistology would ideally be needed. If such data are not available, please discuss this discrepancy in more depth.

Response 2: We thank the reviewer for this insightful comment. We agree that the reduced adipocyte size observed in winter WAT could superficially resemble WAT beiging. However, in our samples, the smaller adipocytes were not accompanied by clear multilocular lipid droplet structures, which are characteristic of classical beige adipocytes. Moreover, the expression of thermogenic markers, including UCP1, PGC1a, CIDEA, and ELOVL3, was markedly downregulated rather than induced during winter sleep. Therefore, this pattern is not consistent with typical cold-induced WAT beiging.

In classical deep-hibernating rodents, hibernation is often associated with BAT activation and WAT browning, including multilocular adipocyte morphology and upregulation of thermogenic genes (Chayama et al., 2018; Hampton et al., 2013). In contrast, suppression of thermogenic programs in WAT has also been reported in non-classical hibernation strategies, such as brown bears, in which WAT remodeling during hibernation is characterized by broad metabolic suppression rather than robust browning activation(Jansen et al., 2019). This is consistent with the physiological pattern observed in raccoon dogs, which show diminished BAT activity during winter and limited beige adipocyte recruitment or UCP1 induction in previous studies. Thus, we interpret the reduced adipocyte size in our study as lipid mobilization-associated adipocyte atrophy rather than classical WAT beiging.

To clarify this point, we have revised the Results section to emphasize that no typical multilocular lipid droplets were observed in either depot. We have also expanded the Discussion to explain that adipocyte shrinkage during winter sleep may superficially resemble beiging, but the lack of multilocular morphology and thermogenic gene activation supports adipocyte atrophy rather than beige adipocyte formation. We have revised the manuscript accordingly (Page 4, Lines 129–132; Page 10, Lines 287–293).

Reference:

Chayama, Y., Ando, L., Sato, Y., Shigenobu, S., Anegawa, D., Fujimoto, T., Taii, H., Tamura, Y., Miura, M., and Yamaguchi, Y. (2018). Molecular Basis of White Adipose Tissue Remodeling That Precedes and Coincides with Hibernation in the Syrian Hamster, a Food-Storing Hibernator. Front Physiol 9, 1973.

Hampton, M., Melvin, R.G., and Andrews, M.T. (2013). Transcriptomic analysis of brown adipose tissue across the physiological extremes of natural hibernation. PLoS One 8, e85157.

Jansen, H.T., Trojahn, S., Saxton, M.W., Quackenbush, C.R., Evans Hutzenbiler, B.D., Nelson, O.L., Cornejo, O.E., Robbins, C.T., and Kelley, J.L. (2019). Hibernation induces widespread transcriptional remodeling in metabolic tissues of the grizzly bear. Commun Biol 2, 336.

3. Fig1 Line 82

I wonder whether the timing of sample collection is important for interpreting the data, especially gene expression data. For example, samples were collected after approximately 50% body weight loss. If this is important, please clarify it in Fig. 1 or around line 82.

Response 3: We thank the reviewer for this helpful comment. We agree that the timing of sample collection is important for interpreting the physiological and transcriptomic data. In the revised manuscript, we have clarified the sampling design in both Figure 1A and the Results section. Samples collected in October were used to represent the autumn fat-accumulation phase, when raccoon dogs actively accumulate WAT reserves. Samples collected in January were used to represent the stable winter sleep phase, after seasonal reductions in food intake and body weight had occurred. This clarification helps define the biological context of the gene expression analyses as a comparison between the autumn fat-accumulation phase and the winter sleep phase, rather than an acute response to body weight loss alone. We have revised the manuscript accordingly (Page 2, Lines 81–89).

4. Fig. 1E:
   To me, the interscapular region appears to contain activated BAT. Could you clarify whether this conflicts with the statement in the Introduction?

 Response 4: We thank the reviewer for raising this important point. We agree that the localized warm signal near the interscapular region in Figure 1E may visually appear similar to a BAT-associated thermal signal. However, Figure 1E shows infrared thermography, which reflects fur-covered body surface temperature distribution rather than BAT activity. Therefore, the localized surface temperature pattern should not be interpreted as direct evidence of activated BAT.

The statement in the Introduction is based on previous PET/CT and histological analyses in adult raccoon dogs, in which BAT was not detected during the winter season (Niiranen et al., 2021). In that study, metabolically active signals in the scapular region were further examined and were attributed to lymph nodes rather than BAT. Thus, the infrared signal observed in our Figure 1E does not conflict with the previous conclusion that adult raccoon dogs show diminished or undetectable BAT activity during winter.

Reference:

Niiranen, L., Makela, K.A., Mutt, S.J., Viitanen, R., Kaisanlahti, A., Vicente, D., Noponen, T., Autio, A., Roivainen, A., Nuutila, P., et al. (2021). Role of Brown and Beige Adipose Tissues in Seasonal Adaptation in the Raccoon Dog (Nyctereutes procyonoides). Int J Mol Sci 22.

5. Fig. 2H:
   Does this Venn diagram show genes regulated in the same direction? If not, please replace it with a Venn diagram showing only commonly regulated genes in the same direction. Separating the data into two figures, one for downregulated genes and one for upregulated genes, would also help. Even if this disrupts the connection to Fig. 3, please provide the numbers in the text.

Response 5: We thank the reviewer for this helpful suggestion. In the original version, the Venn diagram showed the overlap of all DEGs between back-fat and tail-fat without separating genes according to their regulatory direction. We agree that this presentation could obscure whether the shared genes were regulated in the same direction.

In the revised manuscript, we have reanalyzed the DEGs by separating upregulated and downregulated gene sets. Figure 2H has been revised to show the overlap of commonly upregulated and commonly downregulated genes separately between back-fat and tail-fat during winter sleep. We have also provided the corresponding numbers in the Results section, including 399 commonly upregulated genes and 379 commonly downregulated genes. This revision clarifies the directionality of shared transcriptional responses while preserving the connection between the shared DEGs in Figure 2 and the downstream pathway analysis in Figure 3. We have revised the Results section and figure legend accordingly (Page 5, Lines 142-146, 162–171; Figure 2H).

6. Lines 165–168:
   These statements are somewhat broad and may be overinterpreted.

Response 6: We thank the reviewer for this important comment. We agree that the original statement was too broad. We have therefore revised this sentence to avoid overinterpretation and now describe the upregulation of G6PC1, PCK1, and PSAT1 as selective regulation of specific metabolic nodes rather than definitive evidence of strategic metabolic redirection. We have revised the Results section accordingly (Page 7, Lines 203–205).

7. Lines 330–337:
   Was the order of tissue collection randomized? If not, please state this in the limitations section.

Response 7: We thank the reviewer for this helpful comment. The order of tissue collection was randomized across animals. We have clarified this in the Materials and Methods section and further stated that back-fat and tail-fat samples were collected within a short time window during the same surgical procedure to minimize potential variation caused by sampling time or handling. We have revised the Methods section accordingly (Page 11, Lines 359–361).

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

In the current study, the authors investigated the WAT remodeling during shallow hibernation in raccoon dogs. The study is very interesting and I would like to recommend its acceptance though several minor issues the authors may consider:

1. In infrared thermography images, the highest temperature in the figures is only around 17 C. However, the normal body weight of Raccoon dog is around 38-39C. I accept that the infrared thermography images does not equal to the body temperature, however, this difference seems to be too big.
2. Using only males limits generalizability. Please provide a reason for this choice or acknowledge it as a limitation
3. The authors extracted RNA from adipose tissue using TRIzol reagent, however, based on my experience, due to high content of lipid, the TRIzol often could not achieve desirable extraction from adipose tissue. Please clearly state are there any extra steps  used to remove lipid; or I would like to recommend QIAzol lysis reagent, which is specifically designed for RNA extraction from adipose tissue.
4. For CIBERSORT analysis, two key questions should be considered. 1) this tool seems to be designed for tumor analysis, does this also work for adipose tissue;
5. 2) LM22.txt is for human immune cells, does this also work for raccoon dogs.

Author Response

Reviewer #3

Comments to Author

In the current study, the authors investigated the WAT remodeling during shallow hibernation in raccoon dogs. The study is very interesting, and I would like to recommend its acceptance though several minor issues the authors may consider:

1. In infrared thermography images, the highest temperature in the figures is only around 17 C. However, the normal body weight of Raccoon dog is around 38-39C. I accept that the infrared thermography images does not equal to the body temperature, however, this difference seems to be too big.

Response 1: We thank the reviewer for this important comment. We agree that the apparent temperature values in the infrared thermography images are much lower than the rectal body temperature of raccoon dogs. This difference is expected because infrared thermography measures the external body surface temperature, particularly the fur-covered surface, rather than core body temperature. Raccoon dogs have a dense winter fur coat and substantial subcutaneous fat insulation, which markedly reduce heat transfer from the body core to the external surface. Under low ambient temperatures, the outer fur surface can therefore appear much colder than the internal body temperature measured by a rectal probe.

In our study, rectal temperature was measured separately and remained around 38–39°C, confirming that the animals maintained normal core body temperature. The infrared thermography images were used to compare relative changes in surface heat distribution and posture-related heat conservation between seasons, not to estimate absolute body temperature. Importantly, all infrared images were acquired using the same camera and standardized imaging procedures, allowing reliable relative comparisons between autumn and winter. To avoid misunderstanding, we have clarified in the revised manuscript that infrared thermography reflects fur-covered body surface temperature rather than core body temperature, and that the low apparent temperature values likely result from the combined effects of dense fur insulation, low ambient temperature, and imaging geometry. We have revised the Methods section accordingly (Page 12, Lines 378-382).

2. Using only males limits generalizability. Please provide a reason for this choice or acknowledge it as a limitation

Response 2: We thank the reviewer for this important comment. We agree that the rationale for using a single sex should be clearly stated. In the revised manuscript, we have added an explanation in the Materials and Methods section. Only male raccoon dogs were used to minimize potential confounding effects associated with seasonal reproductive cycles and sex hormone fluctuations, which may influence adipose tissue metabolism, immune status, and hibernation-related physiological responses. We have revised the Materials and Methods section accordingly (Page 11, Lines 351–353).

3. The authors extracted RNA from adipose tissue using TRIzol reagent, however, based on my experience, due to high content of lipid, the TRIzol often could not achieve desirable extraction from adipose tissue. Please clearly state are there any extra steps used to remove lipid; or I would like to recommend QIAzol lysis reagent, which is specifically designed for RNA extraction from adipose tissue.

Response 3: We thank the reviewer for this helpful comment. We agree that RNA extraction from adipose tissue can be technically challenging because of its high lipid content. In this study, total RNA was extracted from adipose tissue samples using TRIzol reagent and further purified with RNeasy Mini spin columns according to the manufacturers’ instructions.

Although QIAzol is well suited for lipid-rich tissues, TRIzol-based RNA extraction, either alone or combined with column-based purification, has also been widely used in adipose tissue studies. For example, previous studies on adipose thermogenesis have extracted RNA from adipose tissues and adipocytes using TRIzol-based methods (Ikeda et al., 2017; Rahbani et al., 2024). In our study, to ensure that the extracted RNA was suitable for downstream analyses, RNA quality and purity were assessed using a NanoDrop 2000 spectrophotometer. All samples used for subsequent qPCR and RNA-seq analyses showed acceptable purity, with A260/A280 ratios ranging from 1.80 to 2.00. These quality-control results indicate that our RNA extraction and purification procedure provided RNA of sufficient purity for the analyses performed in this study.

We have clarified the RNA extraction and quality-control information in the Materials and Methods section (Page 12, Lines 394–397). We also appreciate the reviewer’s recommendation regarding QIAzol and will consider this reagent in future adipose tissue RNA extraction experiments.

Reference:

Ikeda, K., Kang, Q., Yoneshiro, T., Camporez, J.P., Maki, H., Homma, M., Shinoda, K., Chen, Y., Lu, X., Maretich, P., et al. (2017). UCP1-independent signaling involving SERCA2b-mediated calcium cycling regulates beige fat thermogenesis and systemic glucose homeostasis. Nat Med 23, 1454-1465.

Rahbani, J.F., Bunk, J., Lagarde, D., Samborska, B., Roesler, A., Xiao, H., Shaw, A., Kaiser, Z., Braun, J.L., Geromella, M.S., et al. (2024). Parallel control of cold-triggered adipocyte thermogenesis by UCP1 and CKB. Cell Metab 36, 526-540 e527.

4. For CIBERSORT analysis, two key questions should be considered. 1) this tool seems to be designed for tumor analysis, does this also work for adipose tissue;

2) LM22.txt is for human immune cells, does this also work for raccoon dogs.

Response 4: We thank the reviewer for raising these important questions. We agree that the CIBERSORTx results should be interpreted carefully. Although CIBERSORTx has been frequently used in tumor immunology, the method itself is a general deconvolution approach for estimating relative immune cell composition from bulk transcriptomic data and is not restricted to tumor samples. Therefore, it can provide useful exploratory information for non-tumor tissues when immune-related transcriptional signatures are present.

We also agree that the LM22 reference matrix was originally developed from human immune cell expression profiles. Its application to raccoon dog adipose tissue therefore has an inherent cross-species limitation. To avoid overinterpretation, we have revised the manuscript to state explicitly that CIBERSORTx was used only as an exploratory approach and that the inferred immune cell proportions should be interpreted with caution.

Importantly, our conclusion was not based solely on CIBERSORTx. To further evaluate the predicted macrophage-associated remodeling, we performed immunofluorescence staining for CD163, a commonly used marker of M2-like macrophages. The increased CD163-positive macrophage density in winter tail-fat supports the CIBERSORTx-inferred increase in M2 macrophage-associated signatures. We have revised the Results sections accordingly (Page 8, Line 252-257).

Author Response File: Author Response.pdf