An intense 60-day weight-loss course leads to an 18 kg body weight reduction and metabolic reprogramming of soldiers with obesity

Participants and lifestyle improvement and health center

Every three months throughout the year, groups of military personnel with obesity who need to reduce their body weight to continue their activities within the Mexican Army are recruited at the Lifestyle Improvement and Health Center of the Physical and Aquatic Skills Program.

The Center aims to help soldiers regain their physical condition to fulfill their duties. During the 60-day training program, the participants cannot leave the Center and may only consume the food provided.

In the present study, 12 soldiers of the Mexican military with obesity who entered the Lifestyle Improvement and Health Center of the Physical and Aquatic Skills Program in Mexico City during the last recruitment period were included, all of whom signed an informed consent form and completed the program until the last day.

The Body-Mass Index (BMI) was calculated according to the World Health Organization (WHO) definition. Soldiers with a BMI equal to or higher than 25 were classified as overweight, and those with a BMI equal to or above 30 were classified as obese (World Health Organization (WHO), 2021).

All participants of the study were obese with a BMI >30 kg/m², with no upper limit, age, or sex restriction. However, since the training program is intense, only healthy individuals with no other diseases, such as diabetes, were admitted.

Clinical data were obtained according to Standard Operating Procedures (SOPs) in a military hospital’s clinical laboratory. These methods include enzymatic, colorimetric, and photometric detection.

Nutritional treatment

The nutrition followed the recommendations of the European Society of Cardiology (ESC) (Visseren et al., 2021). Still, it incorporated elements of the local milpa diet, such as regional vegetables and fruits, monounsaturated fatty acids provided by avocados, and whole grains from tortillas.

During the 60-day training program, total energy intake was gradually reduced by 200 kcal every 20 days under the guidance of a nutrition professional. Meals were prepared by cooks supervised by this same professional at the Lifestyle Improvement and Health Center facilities of the Physical and Aquatic Skills Program.

They started from 1,800 kcal and ended the last twenty days with a total consumption of 1,400 kcal per day. Macronutrient distribution over the 60-day period consisted of 50.1% to 53.41% carbohydrates, 18.26% to 20% proteins, and 28.33% to 30.6% lipids, with saturated fatty acids comprising 4.3% to 5.4%, polyunsaturated fatty acids 6.6% to 7.5%, monounsaturated fatty acids 13.08% to 15.8%, trans fatty acids ranging from 0% to 0.6%, and other fatty acids ranging from 2.3% to 3.71%. Analysis of diets was conducted using the Food Processor software, Version 11.11, ESHA Research (Salem, Oregon, United States). Details of the food composition are presented in Table 1.

The total energy intake was reduced by adjusting the portions’ size and the food’s caloric value. The dietary equivalents were prescribed according to the Mexican System of Equivalent Foods (SMAE), with portion adjustments designed to gradually reduce 200 kilocalories every 20 days. Over the course of the dietary intervention, fruit equivalents were reduced from an initial average of 6.9 to 5.9 in the final 20 days. Concurrently, vegetable equivalents increased from 5.6 to 5.9, and cereal equivalents decreased from 9.3 to 6.5. Throughout the 60-day dietary protocol, dairy intake was consistently maintained at two equivalents per day. Fat intake, including those containing and devoid of protein, was systematically reduced from an average of 8.2 to 6.6 equivalents per day. Very low-fat food equivalents were decreased from 2.3 to one per day, while low-fat food equivalents were maintained between zero and one per day. Moderate-fat food equivalents were restricted to one per day, and moderate to high-fat food equivalents were limited to two per week.

During the intervention, the meal plan was divided into five sections, with ten menus for each 20 days. The diet was based on consuming vegetables such as nopal (cactus), mushrooms, and locally available fruits, including papaya, orange, and watermelon. Nuts or seeds such as peanuts or pistachios were included. Low-fat dairy products like yogurt, lactose-free milk, and panela cheese were also incorporated. Red meat intake was limited to less than 200 g per week. Fish and poultry, including skinless chicken, egg, tuna, and locally sourced white fish, were consumed in portions ranging from one to three servings per day, while turkey ham was limited to less than two servings per week. Sugary drinks were avoided and replaced with unsweetened tea, hibiscus, or lemonade with sucralose. For food preparation, one to three servings of cooking, olive, or margarine were used in each meal.

Physical activity program

The Physical Activity Guidelines for Americans suggest performing more than 150 to 300 min per week of moderate-intensity aerobic physical activity or more than 75 to 150 min per week of vigorous-intensity aerobic physical activity (Piercy et al., 2018). The purpose of the 60-day program was to progressively engage the participants in physical exercises to decrease their body mass index gradually.

The program was divided into nine weeks and distributed to progress the physical activity over 60 days. The timing and planned exercises were subject to and supervised at the sports instructor’s discretion based on the physical capabilities of the participating military personnel in the study while still following the established program at the Lifestyle Improvement and Health Center of the Physical and Aquatic Skills Program.

A fixed schedule was followed throughout the seven days of the week to carry out the planned activities. Exercises were conducted three times daily from Monday to Saturday. On the seventh day, participants only engaged in 60 min of mild to moderate aerobic activities, with the remainder of the day allocated for rest and family activities at the improvement center. Joint lubrication and muscle activation exercises were performed at the beginning of each session.

Additionally, aerobic and resistance activities were included, gradually increasing in intensity from mild to vigorous. The intensity of these exercises was assessed using the Metabolic Equivalent of Task (MET) from the Compendium of Physical Activities 2011. Relaxation and respiratory recovery exercises were conducted at the end of each session. Detailed descriptions of the physical activity program by weeks and hours are provided in Table 2.

Sample preparation

The urine samples were collected on the first day of admission to the Lifestyle Improvement and Health Center of the Physical and Aquatic Skills Program before starting the nutritional and physical activity intervention. The samples were obtained in the morning, as the first urine of the day and mid-stream, with participants in a fasting state and previously cleansed. Samples were collected in a sterile 120 ml container, with hands washed with soap and water previously to prevent contamination. The same procedure was repeated to collect the final urine sample, which was taken the morning after completing the 60-day intervention.

The urine samples were stored at −60 °C until processing. The processing of each urine sample was carried out following the URINE protocol (Alseekh et al., 2021). In a 1 ml microcentrifuge tube, 800 µl of cold methanol (MeOH) was added along with a 200 µl sample of urine from each participant. Similarly, a urine pool for each diet and a pool for each time group were prepared. Ten quality control (QC) tubes containing similar proportions of all the samples were also included. Each tube was then agitated for 30 s, and the samples were incubated at −20 °C for 24 h, followed by centrifugation at 16,000 g for 15 min at four °C. After centrifugation, 200 µl of the supernatant was transferred to 1 ml tubes and evaporated to dryness using a Speed Vacuum Eppendorf concentrator at 10 °C. The dried extract was reconstituted in 200 µl of MeOH/water (1:1, v/v) and agitated for 1 min. Subsequently, the samples were centrifuged at 16,000 g for 15 min at 4 °C. The supernatant of each sample was transferred to UPLC-MS vials (Ultra High-Performance Liquid Chromatography with Mass Spectrometry) for subsequent injection. Three blanks containing the solvent used in sample processing (MeOH/H2O, 1:1, v/v) were included.

High-performance liquid chromatography coupled to mass spectrometry

The samples were analyzed using a Waters XEVO G2-XS system with a quadrupole time-of-flight (QToF) mass analyzer. Modified methods described in 2022 (Albores-Mendez et al., 2022) were used to separate molecules. The Waters Corp (Milford, MA, USA) ACQUITY UPLC BEH C18, 130 Å, 1.7 µm, 2.1 mm × 100 mm BEH C18, 2.1 × 100 mm, 1.7 µm column was used. The mobile phases used solvent A, consisting of water with 0.1% formic acid, and solvent B, consisting of acetonitrile with 0.1% formic acid. All solvents used were LC-MS grade, acquired from Sigma Aldrich (St. Louis, MO, USA) and JT Baker (Phillipsburg, NJ, USA). The flow rate in the column was 0.1 mL/min until it was established at 0.4 mL/min, and the column temperature was maintained at 55 °C during the process. The gradient of injection in the column started at 1% of solvent B and increased to 99% at minute 10, remaining at this injection gradient until minute 13. At minute 14, the injection gradient decreased to 1% and remained at that level until minute 17. The Waters XEVO G2-XS was used in high-resolution positive mode. A cone voltage of 40V and a capillary voltage of 3 kV were applied. The desolvation temperature was set at 450 °C, while the source temperature was 120 °C. The cone gas flow was set to 50 l/h, and the desolvation gas flow was set to 1000 L/h. A fragmentation energy ramp from 10 to 40 eV was applied, and the scan time was set at 0.2 s with a 30-second interval.

Conversion of raw files to mzML

We used the ProteoWizard msconvert tool (https://proteowizard.sourceforge.io/) (Kessner et al., 2008) for converting the raw mass spectrometry data into the mzML community format.

Mass spectrometry data pre-processing

For pre-processing the mass spectrometry data and the following statistical analyses, we used the web-based version of MetaboAnalyst (https://www.metaboanalyst.ca/) (Xia et al., 2009; Pang et al., 2021; Wishart, 2020). To improve subsequent statistical analyses, we applied data filtering (Hackstadt & Hess, 2009). We used the functions provided by MetaboAnalyst: an Interquartile range (IQR) filter of 40%, normalization on the median, square root transformation, and auto-scaling (mean-centered and divided by the standard deviation of each variable). The uploaded data set contained 24 samples. The pre-processing resulted in 1,377 features, and the data matrix was organized into two groups: before and after the treatment.

Figure 1 demonstrates the effect of the mass spectrometry data normalization. The normalized data were used for further statistical and pathway analyses.

Global metabolomics and functional analysis

For global metabolomics and functional analysis, we defined a molecular weight tolerance of 10 ppm and used the Mummichog 1.0 (Li et al., 2013) and gene set enrichment analysis (GSEA) (Subramanian et al., 2005) algorithms with a p-value cutoff of 0.15. For the visual examination, we used a scatter plot, and for the peak annotation, the Homo sapiens (MFN) meta library, compiled from KEGG, BiGG, and the Edinburgh model (Li et al., 2013). We only considered pathways and metabolite sets with at least two entries and results with at least two hits.

Institutional review board statement

This study complies with the International Ethical Guidelines for health-related research with human beings, elaborated by the Council for International Science Organizations Doctors (CIOMS) in collaboration with the World Health Organization (WHO). The Research Committee and the Bioethics Committee of the Medical School approved this work. Escuela Militar de Medicina, Universidad del Ejército y Fuerza Aérea Mexicanos (reg. 0129012020.).

Clinical data of the participants

All twelve participants of the metabolic study, eleven men and one woman, were obese with a body mass index (BMI) above 30 kg/m². Notably, only soldiers without additional diseases were admitted to the course to ensure that they would resist the intense 60-day training program.

Table 3 summarizes the clinical data of the individuals who participated in the metabolic study before and after entering the 60-day training program.

Weight reduction during the training program

Figure 2 shows the weight of the twelve soldiers before and after passing the training program. The mean weight decreased from 94.2 kg to 74.2 kg. Applying a pairwise t-test reported a p-value of 4.082 ⋅ 10⁻⁹, with a mean difference of 18.075 kg for the participants’ weight before and after the training program. The 95% confidence interval for the weight loss was between 15.668 kg and 20.482 kg.

The participants’ average body mass index (BMI) was reduced from 32.8 kg/m² to 26.4 kg/m².

Consequently, the protocol is highly efficient for significantly reducing the participants’ body weight.

Classification by metabolic profiles

To test if the treatment impacts the metabolic profile of the participants, we subjected the data of all twelve datasets before and after the 60-day training to a random forest algorithm and evaluated the generated models.

The metabolic datasets before and after the treatment could be distinguished without error by the Random Forest model, indicating distinct metabolic profiles of the individuals before and after the treatment.

Figure 3 presents the classification of the sample groups according to their metabolic profiles. The predictive Random Forest model permits reliable discrimination between samples before and after the treatment. Figure 3A shows the correct clustering of the samples. The graphs B and C of the exact figure indicate that as few as five features of the metabolic profile are sufficient for identifying a sample with an error of about 1%. Figure 3D lists the important variables for one model.

The highly reliable predictive model demonstrates substantial differences in the metabolic state of the soldiers before and after the training program. In addition, suitable sets of a few (five to ten) compounds could be used for monitoring the training state of participants of a weight loss program.

Metabolic pathways

Table 4 lists the most prominent differential metabolic pathways identified by the combined Mummichog (Li et al., 2013) and GSEA (Subramanian et al., 2005) analyses.

As expected, pathways related to fat, sugar, and nutrition-related metabolism, such as saturated fatty acids beta-oxidation, bile acid biosynthesis, purine metabolism, and the TCA cycle, were modified.

Most differential features can be annotated to amino acid metabolism, including the routes of β-alanine, tryptophan, methionine and cysteine, lysine, tyrosine, and related pathways, such as urea/amino group metabolism.

Another large group is related to the metabolism of vitamins, such as C, D3, E, C21 steroid hormones, and C1 metabolism.

In addition, the caffeine and porphyrin metabolism were affected.