Study by the Associate Professorship of Exercise Biology examines 40-year-old Mader model of energy metabolism

What happens in the muscles when we sprint, cycle, or run a marathon? Based on a 40-year-old mathematical model, new software has been developed and has now been used to simulate key processes of human energy metabolism during physical exercise. An international research team led by Prof. Dr. Henning Wackerhage, head of the Associate Professorship of Exercise Biology, has now systematically investigated the so-called Mader model using experimentally observed metabolic phenomena. At the same time, “MetaboliSim” is the first freely available software that allows researchers to test and further develop the model themselves. The study “The Mader model of muscle energy metabolism simulates key metabolic exercise phenomena” has been published in the journal “Proceedings of the National Academy of Sciences” (PNAS). The journal is one of the world’s most prestigious multidisciplinary scientific journals and has an impact factor of 9.5.

The foundations of the model date back to the 1980s. Sports medicine specialist Alois Mader developed a system of more than 30 mathematical equations describing how the body supplies energy during physical exercise. Among other things, the model links oxidative phosphorylation, glycolysis, and the phosphocreatine system. For a long time, this approach received little international attention.

“What makes the Mader model special is that it simulates human energy metabolism during physical exercise both mechanistically and quantitatively,” explains Prof. Wackerhage. Together with now emeritus Bochum sports medicine specialist Prof. Dr. Hermann Heck, he initiated the current study. To this end, the international research team also developed the software “MetaboliSim,” which is based on the Mader model. To test the model, the team investigated whether it could reproduce ten known phenomena of energy metabolism. For this purpose, the researchers replicated previously published exercise experiments without adapting the equations or their constants to the individual phenomena.

The phenomena investigated included, for example, the fact that ATP concentration in the muscle remains largely stable even during fatigue, while phosphocreatine stores decrease. The characteristic pattern of fat oxidation with increasing exercise intensity, different lactate responses, and faster lactate elimination during active recovery could also be simulated. Overall, the model qualitatively reproduced all ten phenomena investigated. However, the absolute values do not correspond exactly to the experimental measurements in every case and depend on the individual parameterization.

The model can also represent processes that cannot easily be measured continuously during exercise. “It calculates metabolic variables that are difficult or impossible to measure continuously in humans during exercise,” says Wackerhage. These include phosphocreatine, glycogen, and AMP, for example. The latter is also of interest because it regulates important adaptations to endurance training via the enzyme AMPK. The model can therefore initially provide predictions of how such metabolic variables change at different exercise intensities. These hypotheses can subsequently be tested experimentally in a targeted manner.

For Hermann Heck, the publication also brings full circle a process that began several decades ago. Together with Mader, he was already using the equations in computer simulations in the 1980s. “When I wrote my first program, computers were still very slow. You pressed the button in the evening and saw the calculated curves the next morning,” Heck recalls. Today, such calculations can be performed much more easily with “MetaboliSim” and reproduced by other research groups.

Looking ahead, the model could also complement data from sports watches and other wearables. These devices record variables such as power output, speed, or heart rate, but cannot directly measure many processes in the muscle. “If you combine a person’s individual metabolic profile with the exercise pattern, the model can simulate metabolic variables that a wearable itself cannot measure,” explains Wackerhage. Interval training is one possible example: The model simulates both the decrease in phosphocreatine during exercise and its replenishment during the recovery period. In the future, such information could help to individually plan the duration of exercise and recovery periods. The study also identifies the simulation of fat and carbohydrate oxidation as a potential application in combination with wearable data.
There are also potential applications for longer endurance exercise. “MetaboliSim” can simulate how the use of fat and carbohydrates changes under different exercise conditions. In the future, it could additionally take into account how many carbohydrates are consumed during a race. “I see the model less as a finished product and more as a foundation for future research and training applications,” says Wackerhage.

However, further research is necessary before precise individual predictions can be made. Among other things, parameters such as active muscle mass and glycolytic capacity need to be reliably determined. In addition, the model currently does not account for hormonal regulation or the differential recruitment of various muscle fiber types, for example. The study therefore emphasizes that individual calibration is necessary if quantitative predictions are to be made.

“MetaboliSim” has already found one concrete application: Wackerhage uses the software in his Master’s-level teaching. Students can change parameters and immediately observe the effects on lactate, phosphocreatine, and other metabolic variables during exercise. “Metabolism is often taught by looking at individual metabolites, enzymes, and reactions. It is much more difficult to convey to students how these processes interact as a dynamic system. The Mader model helps students understand how human energy metabolism works during physical exercise,” explains Prof. Wackerhage.

About the study: “The Mader model of muscle energy metabolism simulates key metabolic exercise phenomena”
To the homepage of the Associate Professorship of Exercise Biology

Contact

Prof. Dr. Henning Wackerhage
Associate Professorship of Exercise Biology
Am Olympiacampus 11
80809 Munich

Tel.: 089 289 24480
Email: Henning.Wackerhage(at)tum.de

Text: Bastian Daneyko
Photos: Private