Morning Training Improves Endurance Adaptation in Mice

The timing of exercise is a central variable in circadian physiology, yet many training studies treat it as a scheduling detail rather than an experimental factor. The reference study by Hesketh and colleagues directly tested whether endurance-training time changes long-term adaptation in mice. Its central contribution is the separation of acute time-of-day effects on performance from the chronic effect of performing repeated training sessions at a particular circadian phase.

Study Background and Research Question

Endurance performance varies across the day in both humans and rodents, with capacity often higher during the later active phase. In mice, this variation depends on a functional circadian clock; disruption of core clock genes can eliminate time-of-day differences in endurance capacity. Earlier work also connected daily performance variation with liver glycogen availability and feeding-related rhythms. However, a short training period did not establish whether the time of repeated exercise determines the magnitude of adaptation.

The reference study therefore asked a more specific question: when mice train at the same relative intensity, does early-active-phase or late-active-phase exercise produce superior endurance adaptation? This distinction matters because a mouse may perform better at one time of day before training begins, while still adapting more strongly when trained at another time. The authors focused on six weeks of repeated treadmill exercise rather than the approximately two-week intervention used in earlier time-restricted exercise work, as described in the reference study.

Key Innovation from the Reference Study

The study’s main innovation was an experimentally controlled comparison of training time using individualized exercise intensity. Female mice trained during either the early active phase, designated ZT13, or the late active phase, designated ZT22. Importantly, treadmill workload was set relative to each animal’s maximal running capacity rather than imposed as one identical absolute speed. This design reduced a common confound: a fixed workload can represent very different physiological stress levels for animals that begin with different endurance capacities.

The design also measured performance repeatedly during the intervention. As a result, the investigators could compare baseline capacity, the trajectory of improvement, and the final adapted state. That longitudinal perspective revealed a finding that would be missed by an endpoint-only experiment: early-active-phase training generated a steeper improvement curve, even though final performance was not significantly different between training groups. The work consequently frames exercise timing as a determinant of training efficiency and response variability, not simply as a determinant of daily test performance.

Methods and Experimental Design Insights

Animals were assigned to endurance training at ZT13 or ZT22 and exercised on a treadmill five days per week for six weeks. Training was performed at 70% of each animal’s maximal capacity. Endurance performance was assessed before training and again at week three and week six. Control animals were included for comparisons of body composition and other physiological outcomes. The study also integrated systemic, tissue-level, and muscle-biochemical measurements, allowing performance changes to be interpreted alongside metabolic and contractile phenotypes.

Protocol Parameters

• Training windows: Compare the early active phase at ZT13 with the late active phase at ZT22, as specified in the published study.
• Training frequency and duration: Perform treadmill endurance training five days per week for six weeks; these are literature-reported parameters rather than universal recommendations.
• Relative exercise intensity: Set treadmill work at 70% of each mouse’s maximal capacity to account for individual baseline performance.
• Performance monitoring: Measure endurance at baseline, week three, and week six to distinguish early adaptation from the final trained state.
• Secondary phenotyping: Include blood glucose and lactate, cage activity, body composition, liver and skeletal-muscle glycogen, and muscle mitochondrial and contractile protein measurements.

This structure offers a useful template for circadian exercise experiments: standardize the light-dark schedule, report zeitgeber time, test performance at defined phases, and distinguish relative workload from absolute training volume. The individualized intensity is particularly important because baseline performance differed between the time-of-day groups.

Core Findings and Why They Matter

Before training, mice assessed during the late active phase at ZT22 had significantly greater endurance capacity than mice assessed at ZT13. This result is consistent with the established acute time-of-day pattern in nocturnal rodents. It also confirms that the two groups did not begin with equivalent performance, making the subsequent training response more informative than a simple comparison of final running capacity.

After training, the early-active-phase group showed a 132% increase in endurance performance, whereas the late-active-phase group improved by 45%. Despite this difference in the magnitude of adaptation, performance at week six was similar between groups. The early-active-phase group reached that comparable endpoint with lower absolute training volumes, supporting the authors’ interpretation that ZT13 training was more efficient. The result does not mean that morning training is universally superior under all conditions; rather, it demonstrates that training time can alter the amount of adaptation produced by a given relative stimulus in this mouse model.

Body-composition changes were broadly shared. Fat mass declined by 31% in the ZT13-trained group and 32% in the ZT22-trained group compared with controls, while lean mass and food intake did not differ between training conditions. Liver and skeletal-muscle glycogen content also showed no significant difference between the training groups. This negative result is biologically useful: the superior performance trajectory after ZT13 training was not accompanied by a detectable difference in bulk glycogen stores at the measured endpoint.

The most distinctive molecular findings occurred in skeletal muscle. ZT13 training was associated with higher COXIV protein expression, increased citrate synthase activity, and shifts in myosin heavy-chain isoform expression. These changes indicate remodeling of oxidative and contractile muscle characteristics. At the same time, overall mitochondrial content did not change significantly. Thus, the data favor qualitative or functional remodeling of muscle oxidative machinery over a simple increase in mitochondrial abundance. Because the study was not designed as a mechanistic intervention on one molecular pathway, these associations should not be interpreted as proof that any single protein caused the performance advantage.

For metabolic researchers, the glycogen result is equally important as the muscle-protein result. A glycogen quantification assay can help test whether altered carbohydrate storage accompanies circadian training responses, but unchanged total glycogen does not exclude differences in turnover, mobilization, or the timing of substrate use. The reference study supports measuring glycogen alongside performance and muscle adaptation rather than treating glycogen content as a stand-alone explanation.

Comparison with Existing Internal Articles

The internal article Morning Training Drives Superior Endurance Adaptations in Mice provides a concise interpretation of the same central result: early-active-phase training improves endurance more efficiently than later training. The primary paper adds the quantitative training trajectory, the lower absolute workload in the ZT13 group, and the detailed distinction between unchanged mitochondrial content and altered oxidative or contractile markers.

A second overview, Morning Training Enhances Endurance Adaptation in Mice, emphasizes implications for experimental design in circadian and metabolic research. Read alongside the primary report, it is most useful as a conceptual entry point; the reference paper remains necessary for evaluating the female-mouse model, zeitgeber-time assignments, repeated performance testing, and tissue-level evidence.

Limitations and Transferability

Several limitations constrain the interpretation. The study used female mice, so the findings cannot automatically be generalized to males, other strains, different ages, or disease models. Mouse activity is nocturnal, and ZT13 and ZT22 represent phases of the mouse active period rather than direct equivalents of human morning and afternoon exercise. Human circadian phase, sleep timing, meal timing, and habitual training status may produce different relationships between exercise time and adaptation.

The individualized 70% workload is a strength, but it also means that absolute treadmill work differed between groups. Because the ZT13 group began with lower endurance, its absolute training volume was lower even though relative intensity was matched. This complicates attempts to identify the precise stimulus responsible for adaptation. Treadmill running also introduces handling and motivational factors that may interact with circadian arousal.

In addition, the findings are primarily comparative and associative. The study measured glycogen, citrate synthase activity, COXIV, and myosin heavy-chain expression, but it did not establish that any one of these variables mediated the timing effect. The absence of a difference in bulk glycogen content is informative but does not resolve dynamic substrate flux. Future studies should preserve the separation between training time and testing time, include both sexes where appropriate, and use carefully timed sampling to determine how performance, carbohydrate handling, and muscle remodeling evolve across the intervention.

For glycogen storage disease research, diabetes-related exercise studies, and broader energy-metabolism models, the practical lesson is methodological rather than clinical: record circadian timing explicitly, match exercise intensity appropriately, and measure substrate stores together with functional outcomes. The mouse results justify such designs but do not by themselves establish a human exercise prescription.

Research Support Resources

Researchers extending this type of work can use the Glycogen Colorimetric Assay Kit II (SKU K2144), a glycogen assay kit intended to support glycogen measurement in biological samples, including circadian metabolism and glycogen storage disease research. Its workflow combines a glycogen hydrolysis assay with glucose oxidation colorimetric assay chemistry and is described for high-throughput glycogen assay applications. The product information reports absorbance at 450 nm, detection of glycogen concentrations as low as 4 µg/mL, and storage at -20°C; these specifications should be validated against the sample matrix and study design. This resource supports measurement workflows but was not identified as a reagent used in the reference study.