Analysis of energy expenditure and behavioral characteristics in different mouse strains under normal and disease conditions Authors: Jiacheng Zeng, Yuanjun Ji, Yan Zou, Jing Tan, Teng Wu, Conghui Shen, Tongsheng Huang, Shijie Xiong, Mengying Liu, Maoquan Yang, Jingwei Lai, Jiangwei Ma, Junhong Wan, Xinlu Fu, Jiang Qian, Hui Li, Ludong Yuan, Weibin Cai Published: 2026 ABSTRACT AbstractBackgroundMetabolic and behavioral traits in mice exhibit substantial variability across strains, ages, and disease states, which can confound preclinical study outcomes. However, systematic characterization of these parameters (e.g., energy metabolism, activity, and feeding behavior) across diverse experimental conditions remains limited.MethodsCore metabolic and behavioral parameters were assessed via metabolic cages in five 8‐week‐old mouse strains (ICR, C57BL/6J, BALB/c, BALB/c‐nude, NOD‐SCID), six C57BL/6J age groups (3 weeks to 18 months), and four mouse disease models (LPS‐induced pneumonia, chronic kidney disease, acute myocardial infarction, and type 1 diabetes). Data were stratified by light/dark cycles and statistically compared.ResultsHerein, metabolic and behavioral phenotypes of mice were systematically characterized across different strains, ages, and four disease models (three organ‐specific, one systemic). Distinct strain‐ and age‐related differences in basal metabolism, activity, and energy expenditure were observed. Specifically, ICR mice displayed higher basal aerobic metabolism, whereas C57BL/6J mice exhibited greater locomotor activity; metabolism and energy balance also underwent marked shifts during development, pregnancy, and aging. Furthermore, all disease models presented unique metabolic rearrangements: LPS‐induced acute pneumonia reduced aerobic metabolism and locomotor activity; CKD caused hypometabolism, polydipsia, and a substrate shift to carbohydrate oxidation; AMI increased respiratory exchange ratio (RER) and water intake; and T1DM induced polyphagia and polydipsia, with normal metabolism but impaired energy utilization.ConclusionsMouse metabolic and behavioral phenotypes are highly dependent on strain, age, and disease state. These findings highlight the need to standardize experimental conditions (e.g., age, strain, disease state) and account for baseline variability when designing preclinical studies. INTRODUCTION Mice are widely used as model organisms in biomedical research due to their genetic tractability, short life cycle, and the ability to closely mimic human physiology and disease states. 1 , 2 Different mouse strains exhibit varying degrees of susceptibility to diseases and display distinct physiological and behavioral characteristics, which can significantly influence the outcome of experimental studies. 3 Understanding the baseline differences in energy expenditure and behavioral patterns among different mouse strains is crucial for the accurate interpretation of experimental data and for the development of strain‐specific research models. At present, metabolic diseases, such as diabetes and obesity, are spreading globally. The World Health Organization's 2016 report indicates that diabetes alone resulted in 1.5 million deaths worldwide and affected 422 million individuals, 4 compelling us to pay close attention to these conditions. Understanding the relationship between metabolism and disease is more critical than ever, as it aids in identifying potential therapeutic targets and disease management strategies. 5 Among these, energy expenditure is a key parameter in metabolic research and is closely associated with metabolic disorders. Furthermore, energy balance, which is the difference between energy intake and expenditure, plays a central role in maintaining metabolic homeostasis and is a critical factor in the development of metabolic diseases. 6 There is ample evidence that factors such as age, diet, physical activity, and genetic background can influence energy expenditure. 7 However, the extent of differences in energy expenditure, energy balance, and behavior among various mouse strains under normal conditions and in disease models is not fully understood. This knowledge gap hinders the development of personalized treatment approaches and the design of effective preventive strategies. In addition to direct measures of energy balance and expenditure, the respiratory exchange ratio (RER)—the ratio of carbon dioxide production to oxygen consumption—offers critical insights into substrate utilization, with values of 1 and 0.7 indicating pure carbohydrate and lipid oxidation, respectively. 8 , 9 , 10 Influenced by diet, activity, and metabolic state, 11 RER alterations in disease models reflect shifts in energy demand and substrate preference; for example, elevated RER in diabetes may denote impaired glucose utilization and increased fat reliance, 12 while reduced RER could signal compensatory carbohydrate metabolism, thereby informing strain‐specific metabolic flexibility and adaptability in health and disease. Locomotor and ambulatory activity, key behavioral metrics from metabolic cage data, also serve as vital indicators of overall health and treatment efficacy, reflecting physical condition, neural function, and metabolic capacity. 13 , 14 , 15 Locomotor activity refers to the total activity count, including both ambulatory and non‐ambulatory movements (e.g., rearing, grooming). Ambulatory activity specifically describes horizontal movements that break consecutive infrared beams, reflecting the actual distance traveled by the animal. Furthermore, dynamic changes in activity patterns across different strains and disease models are essential for interpreting experimental behavioral outcomes. Metabolic cages enable the simultaneous determination of respiratory exchange ratio (RER), energy expenditure (EE), and locomotor activity in mice by integrating a closed/flow‐through gas exchange monitoring system and infrared/piezoelectric sensors. Specifically, the gas sensors measure the oxygen consumption (VO2) and carbon dioxide production (VCO2) of mice in a sealed chamber with a constant air flow, from which RER is calculated as the ratio of VCO2 to VO2 and EE is derived using the Weir equation; meanwhile, infrared beam grids or piezoelectric transducers record the frequency and duration of mouse movements to quantify horizontal and vertical locomotor activity. 16 Several studies have employed this technique to detect alterations in metabolism‐related indicators in mice. 17 , 18 , 19 However, amid the burgeoning interest in metabolism‐related research, a striking paucity of systematic reports persists regarding basal metabolic cage‐derived data across multiple mouse strains and distinct developmental stages. Inappropriate selection of experimental strains not only impairs the reproducibility and translational potential of scientific findings but also constitutes a violation of laboratory animal ethics by compromising the 3Rs principle (Replacement, Reduction, Refinement). 20 Thus, comprehensive comparative studies investigating basal metabolic variations across strains and developmental stages are urgently warranted. This study utilizes metabolic cage bulk data to comprehensively analyze energy expenditure and behavioral characteristics across diverse mouse strains under both physiological and pathological conditions, aiming to provide a reference for strain selection in metabolic disease research and fill a critical knowledge gap. By examining strains with distinct genetic backgrounds and differential disease susceptibilities, we identify strain‐specific responses to environmental challenges and disease states. Employing well‐established disease models—including lipopolysaccharide (LPS)‐induced pneumonia, chronic kidney disease, type 1 diabetes, and acute myocardial infarction—we further characterize how pathological states modify energy metabolism and behavior, insights critical for dissecting human disease pathophysiology and developing therapeutic strategies. Elucidating these dynamics will guide the rational selection of mouse strains and disease models, enabling more reliable interpretation of metabolic and behavioral outcomes in genetic and pharmacological studies. METHODS Animals and experimental groups Wild‐type mice of different strains (including C57BL/6J, ICR, BALB/c, BALB/c‐nude, and NOD‐SCID) and wild‐type C57BL/6J mice of different ages were all purchased from the Laboratory Animal Center of Sun Yat‐sen University. All mice were maintained under standard laboratory conditions: 12‐h light/dark cycle, ambient temperature 22–25°C, and relative humidity 40%–70%. Animals received ad libitum access to water and standard chow diet (Guangdong Medical Laboratory Animal Center) containing 4.8% fat, 18.6% protein, and 61% carbohydrate, unless stated otherwise. All experimental mice were male, with the exception of female mice at gestational day 12, which were only used for the comparison between C57BL/6J (inbred strain) and ICR (outbred closed‐colony strain). For the construction of all disease models, male C57BL/6J mice aged 8–10 weeks were used. All mice were acclimated for 1 week prior to the initiation of any experimental procedures. Within 24 h after confirming successful model establishment, mice were placed into metabolic cages for continuous monitoring. All animal experiments were approved by the Animal Care and Ethics Committee of Zhongshan School of Medicine, Sun Yat‐sen University, and followed the National Institutes of Health Guidelines on the Care and Use of Animals (the protocol number is SYSU‐IACUC‐2025‐001183). Specifically, all surgical operations were performed via isoflurane inhalation anesthesia using a dedicated small‐animal anesthesia machine, with an induction concentration of 2% and a maintenance concentration of 1.5% to sustain stable anesthetic depth throughout the procedure. Prior to surgical incision, local infiltration anesthesia was administered at the operative site using 0.5% lidocaine hydrochloride to further reduce intraoperative pain. Immediately after surgery, postoperative analgesia was administered via subcutaneous injection of buprenorphine at a dose of 0.1 mg/kg body weight. Additional booster injections were given every 8 h for 24 h postoperatively to maintain adequate analgesic efficacy. After raw data collection, all mice were euthanized by cervical dislocation under deep anesthesia. All anesthetic and analgesic procedures were strictly conducted in accordance with the NIH Guidelines for the Care and Use of Laboratory Animals, aiming to minimize surgical pain, distress, and secondary inflammatory responses in mice. LPS‐induced acute lung injury model Lipopolysaccharide (L2880, Sigma‐Aldrich, USA) was stored at −20°C and diluted in sterile phosphate‐buffered saline (PBS) prior to the procedure. Adult C57BL/6J male mice (10–12 weeks old) were anesthetized via isoflurane inhalation. Under deep anesthesia, the neck was hyperextended to expose the upper airway, and intratracheal instillation was performed using an insulin needle. 21 Mice in the ALI group received LPS (3 mg/kg) dissolved in 50 μL sterile PBS, while sham‐operated controls received an equal volume of PBS alone. Following instillation, the neck incision was sutured routinely, and mice were held in an upright position for 30 s to ensure uniform distribution of the solution within the lungs. Afterwards, the animals were placed on a warming pad and monitored until full recovery from anesthesia. 22 All surgical procedures were performed under strict aseptic conditions with standardized anesthesia and postoperative analgesia to minimize surgical trauma and background inflammation. Sham mice received identical surgical manipulation without LPS instillation. Twenty‐four hours after LPS administration, approximately 150 μL of anticoagulated whole blood was collected from the orbital venous plexus of mice. Routine blood parameters, including total white blood cell count and neutrophil percentage, were measured using an automated hematology analyzer to verify the successful establishment of the model. Subsequently, metabolic parameters were continuously monitored via a metabolic cage system for 120 h. Streptozotocin‐induced type 1 diabetes mellitus animal model Eight‐week‐old male C57BL/6J mice were used to establish a diabetic model with streptozotocin (STZ; Cat. No. 60256ES80, Yeasen, Shanghai, China). STZ was freshly prepared and dissolved in citrate buffer at pH 4.5 before use. Mice received intraperitoneal injection of STZ at a dose of 60 mg/kg once daily for 5 consecutive days. Animals were fasted for 12 h with free access to water prior to injection, and then provided with 10% sucrose water for 10 h after injection. Three days after the last STZ administration, blood glucose levels were measured in mice fasted for 6 h; a random blood glucose level >16.7 mmol/L was considered indicative of successful diabetic model establishment. 23 Acute myocardial infarction model Acute myocardial infarction (AMI) was induced in mice as previously described. 24 Briefly, adult C57BL/6J male mice (10 weeks of age) were anesthetized with isoflurane, and chest hair was removed using depilatory cream. A left lateral thoracotomy was performed between the 4th and 5th intercostal spaces, and gentle pressure was applied to extrude the heart from the thoracic cavity. The left anterior descending coronary artery (LAD) was rapidly ligated using an 8‐0 needle suture. The heart was then gently returned to the thoracic cavity, and excess air was expelled to prevent pneumothorax. The skin incision was subsequently closed with a purse‐string suture using 3‐0 silk suture, and mice were placed on a warming pad for several minutes until they regained consciousness. Echocardiography Echocardiography was performed using the Vevo 3100 system (FUJIFILM VisualSonics) according to the following protocol. Briefly, mice were anesthetized with 2% isoflurane, and heart rate was maintained within the range of 400–500 beats per minute (bpm) throughout the procedure. Following acquisition of B‐mode and M‐mode echocardiographic images, cardiac functional parameters (including ejection fraction and fractional shortening) were calculated using the system's built‐in analysis software. Chronic kidney disease model A mouse model of chronic kidney disease (CKD) was induced by administration of an adenine diet. Briefly, 8‐week‐old male C57BL/6J mice were randomly divided into two groups. The CKD model group was fed a diet containing 0.25% adenine (Jiangsu Collaborative Biotechnology Co., Ltd) for 4 weeks to induce renal injury. Mice in the control group were fed a normal control diet without adenine. 25 Experimental animal monitoring system (metabolic cage) Mouse metabolic monitoring was performed using the TSE PhenoMaster automated metabolic cage system. The supporting metabolic cage rack was designed with 12 cage positions, and the overall dimension (including casters) was 2000 mm (length) × 650 mm (width) × 2000 mm (height). The entire rack was made of stainless steel (Figure S1). After instrument calibration, mice of different strains and ages were housed in a one‐mouse‐per‐cage manner and subjected to uninterrupted metabolic monitoring for 120 consecutive hours. The internal dimension of each metabolic cage was 372 mm × 175 mm × 135 mm. All mice were housed at an ambient temperature of 23 ± 1°C and relative humidity of 50%–60%, with a 12 h light–dark cycle (lights on at 9:00 a.m. and lights off at 9:00 p.m.). SPF‐grade breeding rodent feed supplied by Guangdong Medical Laboratory Animal Center with an energy density of 3.56 kcal/g. No environmental enrichment devices were placed in the cages to avoid interference with locomotor activity recording. Corncob bedding was laid to a thickness of 8 mm. The monitored parameters included food intake, water intake, oxygen consumption, carbon dioxide production, respiratory exchange ratio, energy expenditure, body weight, oxygen concentration, carbon dioxide concentration, and locomotor activity. Food and water consumption were recorded continuously via gravity sensors. All metabolic and behavioral data were acquired at a time interval of 39 min. Statistical analysis Metabolic cage raw data were uploaded to CalR 2.0 (https://calrapp.org/) for preliminary online processing. Since body weight acts as a confounding variable that may bias group comparisons, one‐way analysis of covariance (ANCOVA) was performed using SPSS 25.0. In this model, group assignment was included as the fixed factor, metabolic cage‐derived parameters as dependent variables, and body weight as a continuous covariate. Prior to the main ANCOVA, the assumptions of linearity, normality of residuals, homogeneity of variances, and homogeneity of regression slopes (parallelism) were verified (Figures S3–, S9). ANCOVA was then applied to adjust and compare group differences in metabolic outcomes while controlling for the effect of body weight. 26 , 27 Statistical visualization and plotting were performed using GraphPad Prism version 10.1.2. For parameters not confounded by body weight, an unpaired Student's t‐test was used for two‐group comparisons, and two‐way analysis of variance (two‐way ANOVA). ANOVA was performed to assess the effects of time period (3 levels: total, dark, light) and group factor (5 levels, corresponding to different age stages, mouse strains, or experimental groups as specified in each figure legend) on the metabolic parameters, followed by Tukey's multiple comparisons test for post hoc analysis. We removed data from mice that became moribund or died during the experiment, animals with evident abnormalities in food or water intake, and outlier values resulting from technical artifacts such as blocked food hoppers or water bottles. RESULTS Basal metabolic and behavioral phenotypes in 8‐week‐old mice of different strains Following metabolic cage testing of five common laboratory mouse strains, we observed that both ICR and BALB/C‐nude mice exhibited markedly higher oxygen consumption, while NOD‐SCID mice demonstrated the lowest (Figures 1A and S3). Correspondingly, ICR and BALB/C‐nude mice also produced the greatest amount of carbon dioxide, whereas NOD‐SCID mice produced the least (Figures 1B and S3). Further comparative analysis of respiratory exchange ratios (RER) showed that all mouse strains exhibited RER values ranging from 0.85 to 0.95, indicating that their energy metabolism relied on a mixed substrate utilization of glucose and lipids. Nevertheless, C57BL/6J mice displayed a lower RER relative to the other strains, implying a greater contribution of lipids to their overall energy substrate pool (Figure 1C, two‐way ANOVA: interaction F(8,60) = 1.459, p = 0.1916; time main effect F(2,60) = 13.59, p < 0.0001; strain main effect F(4,60) = 13.62, p < 0.0001). Regarding diet and water intake, no significant differences were observed among the mouse strains, suggesting minimal inter‐strain variation in overall energy intake. Notably, BALB/C‐nude mice displayed higher food consumption, which may be attributed to their increased energy demand for thermoregulation due to the lack of body hair (Figures 1D,E and S3). In addition, C57BL/6J mice exhibited significantly higher ambulatory activity and locomotor activity compared with other strains, demonstrating a hyperactive phenotype overall (Figure 1F,G, Locomotor activity [LA]: interaction F(8,60) = 0.5282, p = 0.8307; time main effect F(2,60) = 3.585, p = 0.0338; strain main effect F(4,60) = 8.847, p < 0.0001; Ambulatory activity [AA]: interaction F(8,60) = 0.4024, p = 0.9149; time main effect F(2,60) = 2.808, p = 0.0683; strain main effect F(4,60) = 9.099, p < 0.0001). Energy expenditure is frequently associated with basal metabolic status, ambulatory and locomotor activity, as well as thermogenic demand. Among the mouse strains examined, ICR and BALB/C‐nude mice exhibited the highest energy expenditure. This could be attributed to the higher basal metabolic rate in ICR mice, whereas BALB/C‐nude mice displayed elevated energy expenditure due to increased thermogenic demands for body temperature maintenance (Figure 1H). Finally, no significant inter‐strain differences in energy balance were detected, and all strains maintained a positive energy balance (Figure 1I, Energy Balance [EB]: interaction F(8,60) = 1.050, p = 0.4100; time main effect F(2,60) = 35.95, p < 0.0001; strain main effect F(4,60) = 2.652, p = 0.0417). Collectively, our data comprehensively characterize the metabolic cage parameters of 8‐week‐old mice under normal physiological conditions, thereby establishing a reference baseline for related research. Basal metabolic and behavioral phenotypes. Basal metabolic and behavioral phenotypes were measured in five mouse strains, age 8 weeks (ICR, C57BL/6J, BALB/c, BALB/c‐nude, NOD‐SCID) across total, dark, and light phases. (A) Oxygen consumption (mL/h); (B) carbon dioxide production (mL/h) (C); respiratory exchange ratio (RER); (D) food consumption (kcal/h); (E) water consumption (mL/h); (F) locomotor activity (beam breaks/h); (G) ambulatory activity (beam breaks/h); (H) energy expenditure (kcal/h); (I) energy balance (kcal/h). Data were presented as the mean ± standard deviation (n = 6). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Age‐dependent metabolic and behavioral phenotypes in C57BL/6J mice Given that C57BL/6J is the most widely used inbred mouse strain, it was chosen as a representative model to investigate the metabolic characteristics across distinct physiological stages within the same strain. Among different physiological stages, three‐week‐old pups showed the lowest oxygen consumption and carbon dioxide production, reflecting their adaptive adjustment during the transition from a relatively anaerobic intrauterine environment to an aerobic extrauterine environment after birth. In contrast, 12‐month‐old mice exhibited relatively higher oxygen consumption and carbon dioxide production (Figures 2A,B and S4). Consistent with Figure 1C, the respiratory exchange ratio (RER) of C57BL/6J mice centered at approximately 0.85, suggesting a mixed pattern of carbohydrate and lipid oxidation with a relatively increased contribution from lipids. Overall, no significant changes in total RER were observed across different physiological stages, indicating that this metabolic feature was consistently maintained. However, during the dark phase, the RER values of pups were higher than those at other physiological stages, indicating a relatively higher proportion of carbohydrate utilization in their energy substrate metabolism (Figure 2C, two‐way ANOVA: interaction F(8,60) = 3.018, p = 0.0065; time main effect F(2,60) = 35.29, p < 0.0001; age main effect F(4,60) = 9.305, p < 0.0001). In measurements of food consumed and water consumed across different physiological stages, there were no significant differences among most stages; however, aged mice showed relatively lower food consumption (Figures 2D,E and S4). With regard to ambulatory activity and locomotor activity, 3‐week‐old pups exhibited lower activity levels (Figure 2F,G, Locomotor activity [LA]: interaction F(8,60) = 0.4203, p = 0.9044; time main effect F(2,60) = 18.15, p < 0.0001; age main effect F(4,60) = 2.573, p = 0.0466; Ambulatory activity [AA]: interaction F(8,60) = 0.4030, p = 0.9145; time main effect F(2,60) = 14.42, p < 0.0001; age main effect F(4,60) = 2.695, p = 0.0392). In the comparative analysis of energy expenditure, 3‐week‐old pups showed the lowest levels, due to reduced locomotor activity and lower basal metabolic consumption (Figure 2H). In the final energy balance assessment, 12‐month‐old mice exhibited the lowest energy balance, even presenting a state of negative energy balance. Conversely, three‐week‐old pups maintained a markedly positive energy balance, in line with their physiological state of rapid growth and development. (Figure 2I, Energy Balance [EB] interaction F(8,60) = 3.836, p = 0.0011; time main effect F(2,60) = 57.94, p < 0.0001; age main effect F(4,60) = 7.255, p < 0.0001). Age‐dependent metabolic and behavioral phenotypes Age‐dependent metabolic and behavioral phenotypes were measured in C57BL/6J mice at five different ages (3 weeks, 3 months, 6 months, 12 months, 18 months) across total, dark, and light phases. (A) Oxygen consumption (mL/h); (B) carbon dioxide production (mL/h); (C) respiratory exchange ratio (RER); (D) food consumption (kcal/h); (E) water consumption (mL/h); (F) locomotor activity (beam breaks/h); (G) ambulatory activity (beam breaks/h); (H) energy expenditure (kcal/h); (I) energy balance (kcal/h). Data were presented as the mean ± standard deviation (n = 6). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Total‐phase metabolic and behavioral phenotypes in ICR and C57BL/6J mice across distinct ages Among commonly used laboratory mouse strains, C57BL/6J, as a representative inbred strain, is widely employed in biomedical research. In parallel, the outbred (closed colony) ICR mouse is extensively utilized in pharmacological, toxicological, and routine screening studies due to its favorable characteristics. We therefore conducted a comparative analysis of these two commonly used strains across multiple age groups to observe alterations in metabolic monitoring. At all assessed time points (3 weeks, Pre 12 days, 3 months, and 12 months), ICR mice exhibited significantly higher oxygen consumption and CO2 production compared with C57BL/6J mice (Figures 3A,B and S5). These findings demonstrate that ICR mice possess a higher basal aerobic metabolic rate overall, which is consistent with the total energy expenditure results presented in Figure 1H. Respiratory exchange ratio (RER) reflects the main energy substrate used, with 1.0 and 0.7 indicating carbohydrate and fat oxidation, respectively. ICR mice relied more on carbohydrates at 3 months of age, while the two strains showed similar substrate utilization patterns at other time points (Figure 3C, two‐way ANOVA: interaction F(3,31) = 8.830, p = 0.0002; strain main effect F(1,31) = 10.61, p < 0.0001; age main effect F(3,31) = 0.1510, p = 0.7002). Regarding total food and water consumption, ICR mice exhibited significantly higher energy intake and water consumption than C57BL/6J mice at the juvenile stage (3 weeks); although differences still existed at other age points, the magnitude of these changes was relatively small (Figures 3D,E and S5). In contrast, for total locomotor activity and total ambulatory activity, the locomotor capacity of C57BL/6J mice gradually exceeded that of ICR mice with advancing age (Figure 3F,G), Locomotor activity [LA]: interaction F(3,31) = 5.391, p = 0.0042; strain main effect F(1,31) = 4.311, p = 0.0119; age main effect F(3,31) = 51.49, p < 0.0001; Ambulatory activity [AA]: interaction AA (F(3,31) = 5.359, p = 0.0043; strain main effect F(1,31) = 5.219, p = 0.0049; age main effect F(3,31) = 50.46, p < 0.0001). Collectively, analysis of the above basic indicators revealed that energy expenditure was higher in ICR mice than in C57BL/6J mice at all time points, further confirming the higher basal metabolic rate (BMR) characteristic of ICR mice (Figure 3H). Finally, with respect to energy balance, ICR mice, particularly during old age, exhibited a trend toward negative energy balance, which may be associated with their intrinsically high energy expenditure and aging‐related metabolic alterations. In contrast, energy balance remained relatively stable throughout the life cycle in C57BL/6J mice (Figure 3I, Energy Balance [EB] interaction F(3,31) = 7.605, p = 0.0006; strain main effect F(1,31) = 17.88, p < 0.0001; age main effect F(3,31) = 0.01055, p = 0.9188). This study provides a reference framework to assist researchers in selecting appropriate experimental animal models for studies on metabolism, aging, and behavioral phenotypes. Total‐phase metabolic and behavioral phenotypes. Total‐phase metabolic and behavioral phenotypes were measured in two mouse strains (ICR, C57BL/6J), four age groups (3 weeks, Pre 12 days, 3 months, 12 months). (A) Total oxygen consumption (mL/h); (B) Total carbon dioxide production (mL/h); (C) Total respiratory exchange ratio (RER); (D) Total food consumption (kcal/h); (E) Total water consumption (mL/h); (F) Total locomotor activity (beam breaks/h); (G) Total ambulatory activity (beam breaks/h); (H) Total energy expenditure (kcal/h); (I) Total energy balance (kcal/h). Data were presented as the mean ± standard deviation (n = 6). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Metabolic and behavioral phenotypes are altered in a mouse model of lipopolysaccharide (LPS)‐induced acute pneumonia Inflammation is the most common basic pathological process, and LPS‐induced acute lung injury serves not only as a typical example thereof but also as a representative and common pulmonary disease. We established a mouse model of acute inflammation via intratracheal instillation of lipopolysaccharide (LPS). Peripheral blood analysis revealed that LPS treatment significantly increased the total white blood cell count (WBC), absolute neutrophil count (NEUT), and neutrophil percentage (NEUT%), while markedly reducing the lymphocyte percentage (LYMPH%). These findings confirm the successful induction of a systemic acute inflammatory response in mice (Figure 4A). Notably, there was no significant difference in body weight between the Sham and LPS groups (Figure 4B). This eliminates the confounding effect of body weight variation on subsequent metabolic and behavioral assessments. Analysis of core respiratory metabolic parameters revealed that the change trend of carbon dioxide production in mice was consistent with that of oxygen consumption, and the LPS group was significantly lower than the Sham group in all detected time periods. However, there was no significant overall difference in the respiratory exchange ratio (RER), which was consistent with the change trends of carbon dioxide production and oxygen consumption (Figures 4C and S6, two‐way ANOVA: interaction F(2,24) = 1.226, p = 0.3113; group main effect F(1,24) = 9.800, p = 0.0045; time main effect F(2,24) = 4.298, p = 0.0254). These findings indicate that LPS can significantly inhibit the overall aerobic metabolic capacity of mice. Regarding feeding, drinking, and behavioral parameters, no significant differences in total food or water consumption were observed between the two groups at any time point. However, locomotor and ambulatory activities were significantly reduced in the LPS group, specifically during the dark phase. These results indicate that LPS did not alter basal feeding or drinking behavior but selectively impaired spontaneous and ambulatory activity during the active (dark) phase, demonstrating the selective impact of inflammation on the mice's behavioral activity. The significantly lower energy expenditure in the LPS group was likely attributable to the reduced basal metabolic rate and diminished locomotor activity in LPS‐treated mice. Meanwhile, the energy balance level in the LPS group was comparable to that in the Sham group, with no obvious differences observed at all time points (Figures 4D and S6, Locomotor activity [LA]: interaction F(2,24) = 8.867, p = 0.0013; group main effect F(1,24) = 24.90, p < 0.0001; time main effect F(2,24) = 22.79, p < 0.0001; Ambulatory activity [AA]: interaction F(2,24) = 7.887, p = 0.0023; group main effect F(1,24) = 26.39, p < 0.0001; time main effect F(2,24) = 21.35, p < 0.0001; Energy Balance [EB]: interaction F(2,24) = 0.391, p = 0.6759; group main effect F(1,24) = 0.041, p = 0.8275; time main effect F(2,24) = 8.241, p = 0.0019). Collectively, although LPS did not affect food or water consumption in mice, it reduced oxygen consumption and CO2 production, leading to a marked decrease in overall energy expenditure across all periods. Moreover, LPS impaired behavioral activity during the active dark phase, ultimately contributing to disturbances in energy metabolism and balance (Figure S2). Metabolic and behavioral phenotypes are altered in a mouse model of lipopolysaccharide (LPS)‐induced acute pneumonia. (A) Analysis of peripheral immune cell counts and proportions. Mice were administered lipopolysaccharide (LPS, red) or saline control (Sham, blue). Quantification of white blood cell count (WBC), neutrophil count (NEUT), neutrophil percentage (NEUT%), and lymphocyte percentage (LYMPH%) is shown. (B) Assessment of body weight (Weight) in Sham and LPS‐treated mice. (C) Oxygen consumption, carbon dioxide production, and respiratory exchange ratio (RER) were measured in Sham and LPS‐treated mice over a 24‐h cycle (total), as well as during the dark (dark) and light (light) phases. (D) Food consumption, water consumption, locomotor activity, ambulatory activity, energy expenditure, and energy balance were evaluated in Sham and LPS‐treated mice during the total 24‐h, dark, and light phases. Data were presented as the mean ± standard deviation (n = 5). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Metabolic and behavioral phenotypes are altered in a mouse model of chronic kidney disease (CKD) The common renal disease chronic kidney disease (CKD) was investigated in the present study using a mouse model, through which we compared the alterations in a panel of metabolic monitoring parameters after the induction of CKD. Renal histopathological examination was performed via HE staining. In the control group, renal tubules were structurally intact, featuring clear lumens, polarized epithelial cells, and uninterrupted basement membranes. Conversely, CKD mice manifested marked tubule dilatation, intraluminal protein casts, vacuolar degeneration, and sloughing of epithelial cells, alongside interstitial expansion (Figure 5A). These findings corroborated tubular injury, interstitial fibrosis, and inflammatory infiltrates, hallmarking the pathological features of CKD. Furthermore, comparable body weights between groups eliminated the confounding effects of severe cachexia, ensuring a reliable background for subsequent metabolic characterization (Figure 5B). Analysis of oxygen consumption and carbon dioxide production revealed that CKD mice displayed diminished baseline oxygen uptake coupled with reduced metabolic CO2 emission. These observations suggest impairments in mitochondrial respiratory capacity or a reduction in the whole‐body metabolic rate, aligning closely with the phenotype of decreased systemic bioenergetic stores characteristic of CKD. Nevertheless, CKD mice exhibited a significantly elevated RER compared with control mice, indicating a metabolic switch toward enhanced carbohydrate oxidation and reduced fatty acid oxidation, which may result from the suppression of fatty acid oxidation by metabolic acidosis and uremic toxins (Figures 5C and S7, two‐way ANOVA: interaction F(2,24) = 0.7894, p = 0.4655; group main effect F(1,24) = 13.65, p = 0.0011; time main effect F(2,24) = 7.816, p = 0.0024). Although food consumption was comparable between groups, CKD mice exhibited significantly elevated water consumption at all time points, indicative of typical polydipsia, likely resulting from osmotic diuresis secondary to renal dysfunction or thirst triggered by accumulated metabolic wastes such as urea. Concurrently, CKD mice displayed markedly reduced locomotor and ambulatory activities. Such hypoactivity, together with diminished basal metabolic turnover, accounted for the significantly lower overall energy expenditure in CKD mice relative to control mice. Despite divergent patterns of energy intake and expenditure, energy balance remained similar between groups, suggesting that CKD mice preserve relative energy homeostasis under a hypoactive and hypometabolic state (Figures 5D and S7, Locomotor activity [LA]: interaction F(2,24) = 5.352, p = 0.0120; group main effect F(1,24) = 89.03, p < 0.0001; time main effect F(2,24) = 38.35, p < 0.0001; Ambulatory activity [AA]: interaction F(2,24) = 4.807, p = 0.0176; group main effect F(1,24) = 80.95, p < 0.0001; time main effect F(2,24) = 32.91, p < 0.0001; Energy Balance [EB]: interaction F(2,24) = 0.02934, p = 0.9711; group main effect F(1,24) = 3.191, p = 0.0867; time main effect F(2,24) = 49.67, p < 0.0001). Metabolic and behavioral phenotypes are altered in a mouse model of chronic kidney disease (CKD). (A) Representative histological images of renal tissue from Control and CKD mice (scale bar = 50 μm). (B) Assessment of body weight (Weight) in control and CKD mice. (C) Oxygen consumption, carbon dioxide production, and respiratory exchange ratio (RER) were measured in control and CKD mice over a 24‐h cycle (total), as well as during the dark (dark) and light (light) phases. (D) Food consumption, water consumption, locomotor activity, ambulatory activity, energy expenditure, and energy balance were evaluated in control and CKD mice during the total 24‐h, dark, and light phases. Data were presented as the mean ± standard deviation (n = 5). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Metabolic and behavioral phenotypes are altered in a mouse model of acute myocardial infarction (AMI) We employed myocardial infarction as a representative cardiac pathological alteration and surgically established a mouse model of acute myocardial infarction (AMI). No significant difference in body weight was observed between Sham and AMI groups (Figure 6A). Subsequently, cardiac function was assessed by echocardiography. Notably, AMI mice exhibited significantly reduced ejection fraction (EF%) and fractional shortening (FS%) compared to Sham mice, confirming the successful establishment of the AMI model and demonstrating severe impairment in myocardial contractility (Figure 6B). Metabolic parameters were then measured using metabolic cages as described above. Monitoring of oxygen consumption, carbon dioxide production, and respiratory exchange ratio (RER) showed that overall oxygen consumption was comparable between AMI and control mice. However, carbon dioxide production was increased in the AMI group during the dark phase (the active period of mice), which may be associated with enhanced metabolic activity during this interval. Meanwhile, an elevation in RER was observed in the AMI group, indicating a preferential shift toward carbohydrate oxidation over fatty acid oxidation. This likely reflects a global metabolic switch toward efficient glucose utilization to meet the increased energy demand following myocardial infarction, at the expense of less efficient fatty acid oxidation. (Figures 6C and S8, two‐way ANOVA: interaction F(2,18) = 2.825, p = 0.0857; group main effect F(1,18) = 188.7, p < 0.0001; time main effect F(2,18) = 22.81, p < 0.0001). Furthermore, analysis of food intake, water consumption, and activity‐related energy balance revealed that food intake was significantly higher in the AMI group than in the control group. Correspondingly, energy balance was also elevated in the AMI group as a result of increased food intake, which may reflect an important compensatory strategy adopted by mice to maintain energy homeostasis via upregulating food consumption. In addition, water consumption was markedly increased in the AMI group, which may serve as a compensatory response to preserve circulating blood volume under conditions of impaired cardiac function. For other parameters, including locomotor activity, ambulatory activity, and energy expenditure, although quantitative differences were observed between the two groups, none reached statistical significance (Figures 6D and S8, Locomotor activity [LA]: interaction F(2,18) = 0.01550, p = 0.9846; group main effect F(1,18) = 4.346, p = 0.0516; time main effect F(2,18) = 14.79, p = 0.0002; Ambulatory activity [AA]: interaction F(2,18) = 0.01001, p = 0.9900; group main effect F(1,18) = 5.575, p = 0.0297; time main effect F(2,18) = 12.79, p = 0.0003; Energy Balance [EB]: interaction F(2,18) = 2.310, p = 0.1280; group main effect F(1,18) = 32.02, p < 0.0001; time main effect F(2,18) = 33.29, p < 0.0001). Metabolic and behavioral phenotypes are altered in a mouse model of acute myocardial infarction (AMI). (A) Assessment of body weight in Sham and AMI mice. (B) Cardiac functional parameters in Sham and AMI mice. Ejection fraction (EF%) and fractional shortening (FS%) were quantified, with numerical p‐values for intergroup comparisons indicated above each plot. (C) Oxygen consumption, carbon dioxide production, and respiratory exchange ratio (RER) were measured in Sham and AMI mice over a 24‐h cycle (total), as well as during the dark (dark) and light (light) phases. (D) Food consumption, water consumption, locomotor activity, ambulatory activity, energy expenditure, and energy balance were evaluated in Sham and AMI mice during the total 24‐h, dark, and light phases. Data were presented as the mean ± standard deviation (n = 4). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Metabolic and behavioral phenotypes are altered in a mouse model of streptozotocin‐induced type 1 diabetes mellitus (T1DM) Finally, in addition to characterizing the metabolic phenotypes of these organ‐specific diseases, we established a systemic disease model—mice with type 1 diabetes (T1DM). We subsequently performed comprehensive metabolic monitoring to investigate this multi‐organ and multi‐system disorder. Regarding the baseline parameters of body weight and blood glucose in type 1 diabetes mellitus (T1DM) mice, the T1DM group exhibited significantly lower body weight and markedly elevated blood glucose levels compared with the control group (Figure 7A,B). These findings align with the core features of T1DM, including absolute insulin deficiency and weight loss (one of the cardinal manifestations of “polyphagia, polydipsia, polyuria, and weight loss”), and further indicate that insulin deficiency leads to a shift toward catabolism over anabolism. Following the successful establishment of the mouse model, we conducted a systematic comparison. The results demonstrated that oxygen consumption, carbon dioxide production, and respiratory exchange ratio (RER) were all decreased in the T1DM group relative to the control group. These findings reflect a reduced basal metabolic rate in T1DM mice; meanwhile, absolute insulin deficiency reduces the efficiency of carbohydrate utilization, driving a greater reliance on lipolysis for energy supply, which may eventually lead to the occurrence of lipotoxicity (Figures 7C and S9, two‐way ANOVA: interaction F(2,18) = 2.287, p = 0.1303; group main effect F(1,18) = 47.80, p < 0.0001; time main effect F(2,18) = 3.610, p = 0.0481). Similarly, both locomotor activity and ambulatory activity declined in T1DM mice, which may be attributed to impaired glucose utilization and a state of functional starvation in the body. However, obvious increases in food consumption and water consumption were observed in T1DM mice, a typical feature of polydipsia. This phenomenon is mainly due to hyperglycemia‐induced osmotic diuresis, which causes substantial fluid loss and consequently stimulates thirst. In terms of energy expenditure, T1DM mice exhibited a lower level than control mice, which was consistent with their decreased basal metabolism and reduced locomotor activity. Although no significant intergroup difference in energy balance was reached, T1DM mice showed a trend toward negative energy balance accompanied by body weight loss. The increased food intake and reduced physical activity in T1DM mice could be regarded as a compensatory response to energy loss, yet such compensation is insufficient to completely reverse the metabolic disorders caused by insulin deficiency (Figures 7D and S9, Locomotor activity [LA]: interaction F(2,18) = 0.3434, p = 0.7139; group main effect F(1,18) = 38.58, p < 0.0001; time main effect F(2,18) = 25.12, p < 0.0001; Ambulatory activity [AA]: interaction F(2,18) = 0.3531, p = 0.7072; group main effect F(1,18) = 46.07, p < 0.0001; time main effect F(2,18) = 23.66, p < 0.0001; Energy Balance [EB]: interaction F(2,18) = 0.9866, p = 0.3921; group main effect F(1,18) = 3.634, p = 0.0727; time main effect F(2,18) = 7.814, p = 0.0036). Metabolic and behavioral phenotypes are altered in a mouse model of streptozotocin (STZ)‐induced type 1 diabetes mellitus (T1DM). (A) Assessment of body weight (Weight) in control and T1DM mice. (B) Blood glucose levels in control and T1DM mice. (C) Oxygen consumption, carbon dioxide production, and respiratory exchange ratio (RER) were measured in control and T1DM mice over a 24‐h cycle (total), as well as during the dark (dark) and light (light) phases. (D) Food consumption, water consumption, locomotor activity, ambulatory activity, energy expenditure, and energy balance were evaluated in control and T1DM mice during the total 24‐h, dark, and light phases. Data were presented as the mean ± standard deviation (n = 4). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. DISCUSSION This study systematically integrated multidimensional metabolic cage‐derived data to delineate the basal energy metabolism profiles of five widely used mouse strains, namely C57BL/6J, ICR, BALB/c, BALB/c‐nude, and NOD‐SCID, within a unified experimental framework. Furthermore, it characterized the dynamic shifts in energy metabolism profiles across distinct age stages using C57BL/6J mice as a representative strain. Subsequently, a comprehensive metabolic profiling comparison was performed between a representative inbred strain (C57BL/6J) and a typical outbred strain (ICR). Last but not least, four distinct disease animal models were established using the corresponding strains, enabling a direct comparison of metabolic pattern alterations between the pathological and basal physiological states. Collectively, these findings not only corroborate the pivotal role of genetic background in shaping metabolic phenotypes but also provide empirical criteria to guide strain selection for disease model construction. Under normal physiological conditions, distinct metabolic phenotypes are observed across different mouse strains. For instance, although multiple strains exhibit a consistent pattern of higher respiratory exchange ratio (RER) during the dark phase than the light phase—likely attributable to the nocturnal nature of mice, which require rapid carbohydrate‐derived energy (evidenced by elevated RER) to sustain high‐intensity activities during their active dark period 28 —ICR mice display a more stable RER compared with other strains. Furthermore, ICR mice show higher basal energy expenditure (BEE), which may be associated with their greater genetic diversity as an outbred closed colony. 29 Notably, the immunodeficient BALB/c‐nude strain is characterized by markedly decreased spontaneous locomotor activity, drastically increased energy expenditure and a negative shift in whole‐body energy balance; these metabolic phenotypes differ substantially from another immunodeficient strain, NOD‐SCID. Such discrepancy is presumably driven by the hairless phenotype of BALB/c‐nude mice, which accelerates body heat dissipation and consequently raises energy consumption. Such strain‐specific differences emphasize that when comparing genetically modified models, metabolic “baseline drift” of the background strain must be considered and corrected as a critical confounding variable. Longitudinal metabolic tracking of C57BL/6J mice uncovered distinct developmental transition phases: At 3 weeks of age, juvenile mice exhibited an elevated respiratory exchange ratio (RER), indicative of predominant carbohydrate oxidation—a pattern that reflects nutritional adaptation during the suckling‐to‐weaning transition. Adult mice (12 weeks, 3 months of age) reached a metabolic steady state, with optimal circadian coupling between energy expenditure and locomotor activity. In contrast, aged mice (18 months) displayed a biased negative energy balance, driven by reduced food and water intake; this phenotype recapitulates the metabolic dysregulation associated with human aging‐related metabolic syndrome. 30 , 31 , 32 In addition, consistent with our present findings, numerous previous studies have explored the effects of aging on metabolic parameters in C57BL/6J mice. For instance, distinct from C57BL/6N mice, male C57BL/6J mice exhibit a significant age‐related decline in food intake. 33 Another study similarly reported that aged mice display lower food and water consumption compared with young counterparts. Further comparison of metabolic phenotypes between middle‐aged and young male mice has revealed that multiple aging‐related metabolic abnormalities already emerge at the middle‐aged stage. 34 Beyond the alterations in food and water intake observed in our study, aging‐induced metabolic changes also include increased fat mass, elevated body fat percentage, and reduced insulin sensitivity. 35 , 36 The aforementioned time points indicate that age represents a critical confounding variable in the selection of mouse strains. We recommend prioritizing the 12‐week‐old age cohort for studies investigating adult metabolic phenotypes; conversely, aging research mandates rigorous control over intergroup age distribution matching within individual strains. Furthermore, a cross‐age comparison of two widely used laboratory mouse strains—the inbred C57BL/6J and outbred ICR—revealed that ICR mice exhibited significantly higher oxygen consumption and carbon dioxide production than C57BL/6J mice across all age stages. In contrast, C57BL/6J mice showed consistently elevated locomotor activity relative to ICR mice at every age point. Notably, ICR mice also had persistently greater total energy expenditure (TEE) across the entire age spectrum. This suggests that the basal metabolism (non‐activity‐related metabolism) of ICR mice contributes substantially to energy expenditure. Consequently, despite C57BL/6J mice exhibiting greater activity levels, the ICR mice still exhibit higher total energy expenditure, whereas in C57BL/6J mice, energy expenditure is more strongly correlated with activity itself (relatively lower basal metabolism). In other words, C57BL/6J mice more precisely reflect the independent role of “exercise”, reducing interference from basal metabolism. This also explains why C57BL/6J mice have become the classic experimental strain for exercise‐related research. 37 , 38 , 39 Following the acquisition of the baseline data, we established three organ‐specific and one systemic multi‐system mouse disease models, and monitored their metabolic and behavioral alterations using metabolic cages. In consideration of the previously acquired metabolic and behavioral data of C57BL/6J mice under normal physiological conditions, we uniformly adopted the C57BL/6J strain to establish all four disease models. 40 , 41 , 42 , 43 This design allows us to characterize the metabolic and behavioral alterations in C57BL/6J mice induced by either organ‐specific pathological conditions or systemic multi‐system diseases. As expected, these models generally recapitulated metabolic and behavioral phenotypes similar to those of clinical patients. Specifically, in the LPS‐induced acute pulmonary inflammation model, mice displayed suppressed systemic aerobic metabolism (reduced VO2 and VCO2), diminished locomotor activity during the active phase, and decreased energy expenditure. This hypometabolic and hypoactive phenotype aligns with the clinical manifestations of fatigue, reduced physical activity, and metabolic suppression in critically ill patients with acute pulmonary inflammation. 44 In the establishment of the LPS‐induced acute pulmonary inflammation model, surgical intratracheal instillation was adopted to ensure the precision and homogeneity of LPS administration. 21 Admittedly, this method may introduce mild background inflammatory interference compared with non‐surgical intratracheal instillation, which represents a limitation of our experimental approach. Nevertheless, the sham group received identical surgical manipulation without LPS treatment. Therefore, we believe that the obtained data reliably reflect the metabolic and behavioral alterations in mice under acute lung injury conditions. In the chronic kidney disease (CKD) mouse model, reduced whole‐body metabolic rate, elevated respiratory exchange ratio (RER, indicative of a metabolic shift from fatty acid oxidation to carbohydrate oxidation), characteristic polydipsia secondary to renal dysfunction, and impaired locomotor activity were all consistent with the fatigue, reduced mobility, and uremic metabolic dysregulation seen in human CKD patients—particularly the core feature of suppressed fatty acid oxidation driven by metabolic acidosis and uremic toxins. 45 The acute myocardial infarction (AMI) mouse model recapitulated the acute metabolic adaptations seen in human patients, with increased water intake reflecting the compensatory fluid intake that maintains circulatory homeostasis in patients with impaired myocardial contractility. 46 The streptozotocin (STZ)‐induced type 1 diabetes mellitus (T1DM) mouse model faithfully replicated the classic human phenotype: hyperglycemia, weight loss, polyphagia, and polydipsia, fully recapitulating the core “three polys and one loss” manifestations caused by absolute insulin deficiency. 47 , 48 Notably, however, the energy balance outcomes were not fully consistent with the changes in body weight in this T1DM model. We speculate that this discrepancy stems from inherent limitations of metabolic cage measurements. The energy balance assessed by metabolic cages is calculated solely based on food intake and whole‐body energy expenditure, without taking into account the substantial energy loss derived from urinary glucose excretion. Meanwhile, absolute insulin deficiency impairs the pathways responsible for energy storage and utilization, thereby preventing ingested nutrients from being efficiently converted into body mass. Furthermore, T1DM mice are inherently characterized by a systemic hypercatabolic state. Even without an obvious deficit in external energy balance, the body continuously breaks down endogenous adipose and muscle tissues, ultimately leading to significant body weight loss. In summary, all four murine models showed high translational consistency with human diseases in metabolic, behavioral, and physiological phenotypes. These models are reliable for dissecting disease‐specific metabolic mechanisms. Distinct pathological states drive unique, clinically relevant metabolic signatures, providing a basis for translating preclinical metabolic research into human disease management. This study utilized metabolic cages for dynamic monitoring, enabling the capture of metabolic variations across multiple time periods and mouse strains via long‐term continuous data acquisition. However, the following limitations should be acknowledged: First, the inherent limitations of the metabolic cage environment. To monitor the physiological status of individual mice, strict adherence to single‐housing (one mouse per cage) is mandatory during experimentation. Mice are inherently gregarious social rodents, and single‐housing inevitably induces social isolation stress. 17 , 49 , 50 This may lead to underestimation of the actual activity levels of gregarious strains, thereby introducing confounding effects on metabolic measurements. Second, the current study only compared and analyzed parameters detectable by metabolic cages; the molecular mechanisms underlying the observed behavioral and metabolic alterations remain to be fully elucidated. Third, although multiple mouse strains were included in the analysis, other commonly used strains (e.g., C57BL/6N) 51 , 52 were not investigated due to instrumental constraints. Meanwhile, all four disease models in the present study were established exclusively using C57BL/6J mice, without the inclusion of other mouse strains for parallel comparison. In addition, except for the comparison between ICR and C57BL/6J mice, all other experimental animals used in this study were male. Given that sex is a critical confounding factor affecting metabolic and behavioral phenotypes, the lack of female mice represents an obvious limitation of our work. Related female cohorts will be supplemented in future investigations to further validate our findings. Moving forward, we will build on the findings of this study by adopting multi‐omics approaches, including serum metabolomics, to construct an association network linking “metabolic phenotypes to molecular drivers.” This multi‐dimensional analysis will clarify the molecular mechanisms underlying mouse metabolic variations, with the aim of providing references and guidance for metabolism‐related research and studies on animal welfare ethics. CONCLUSION In summary, genetic background, age, and disease state are key determinants of behavioral and metabolic alterations in mice. The systematic metabolic phenotypic data generated in this study not only provide evidence‐based guidance for the selection of mice of different ages and strains, but also establish metabolic baselines under normal physiological conditions and in various disease animal models. These refined metabolic profiling systems contribute to optimizing experimental design and reducing animal usage, in compliance with the 3R principles and animal welfare ethics. More importantly, our work provides new avenues for the development of precise interventions targeting metabolic adaptability. AUTHOR CONTRIBUTIONS Yuanjun Ji: Methodology; software; resources; supervision; formal analysis; data curation; investigation. Jing Tan: Funding acquisition; software; data curation; supervision; resources. Tongsheng Huang: Software; resources; investigation; formal analysis. Teng Wu: Funding acquisition; software; data curation; supervision; resources. Mengying Liu: Software; resources; investigation. Conghui Shen: Resources; data curation; supervision; software. Shijie Xiong: Software; resources; investigation. Junhong Wan: Software; resources. Xinlu Fu: Software; resources. Hui Li: Software; resources. Jingwei Lai: Software; resources. Maoquan Yang: Software; resources; investigation. Jiacheng Zeng: Methodology; data curation; investigation; validation; formal analysis; supervision; project administration. Jiang Qian: Software; resources. Weibin Cai: Conceptualization; project administration; resources; funding acquisition; writing – review and editing; validation; supervision; software; data curation. Yan Zou: Resources; software; project administration; formal analysis; supervision; data curation. Ludong Yuan: Writing – original draft; writing – review and editing; visualization; formal analysis; data curation; methodology; conceptualization; project administration; resources. Jiangwei Ma: Software; resources. FUNDING INFORMATION W.C. is supported by the National Key Research and Development Program of China (Grant No. 2023YFF0724900), the National Nature Science Foundation of China (Grant Nos. 82570338, 82370283, 82170261, 81970219) and the Guangdong Special Support Program for Health and Medical Talents (Grant No. 0820250101); J.T. is supported by the National Nature Science Foundation of China (Grant No. 82200289), the Guangdong Basic and Applied Basic Research Foundation (Grant No. 2025A1515010845); T.W. is supported by the National Nature Science Foundation of China (Grant No. 82500352). CONFLICT OF INTEREST STATEMENT The authors have no competing interests to declare that are relevant to the content of this article. Supporting information Figure S1: Schematic overview of systemic metabolic and behavioral phenotyping in mice using TSE PhenoMaster automated metabolic cage system. Figure S2: Time‐course analysis of metabolic and behavioral parameters in a mouse model of LPS‐induced acute pneumonia. Figure S3: Regression‐based analysis of metabolic indicators against body mass across different mouse strains. Panels (A–E) represent total/dark/light phase metrics of oxygen consumption, carbon dioxide production, food intake, water intake, and energy expenditure, respectively. ANCOVA was performed with body mass as the covariate to detect strain differences independent of body weight, with significant pairwise comparisons (p < 0.05) annotated in each subpanel. Figure S4: Regression‐based analysis of metabolic indicators against body mass across different ages of C57BL/6J mice. Figure S5: Regression‐based analysis of metabolic indicators against body mass in ICR and C57BL/6J mice at different ages. Panels (A–E) display total/dark/light phase measurements of the same metabolic parameters as described in previous figures (Figure S3). ANCOVA was performed with body mass as the covariate to detect strain‐specific differences independent of body weight at each time point (3 weeks, pre‐12 days, 3 months, and 12 months), with significant comparisons (p < 0.05) annotated in each subpanel. Figure S6: Regression‐based analysis of metabolic indicators against body mass in a mouse model of lipopolysaccharide (LPS)‐induced acute pneumonia and sham controls. Figure S7: Regression‐based analysis of metabolic indicators against body mass in a mouse model of chronic kidney disease (CKD) and controls. Figure S8: Regression‐based analysis of metabolic indicators against body mass in a mouse model of acute myocardial infarction (AMI) and sham controls. Figure S9: Regression‐based analysis of metabolic indicators against body mass in a mouse model of streptozotocin (STZ)‐Induced Type 1 diabetes mellitus (T1DM) and controls.