The effect of aerobic vs. resistance training on plasma homocysteine in individuals with type 2 diabetes

Abstract

Elevated plasma homocysteine concentration is a risk factor for cardiovascular disease, which seems to be the main cause of increased mortality in patients with type 2 diabetes. Previous studies have demonstrated the effect of exercise on homocysteine levels and the magnitude of these benefits seems to depend on the type, mode and frequency of training. The present study aimed to compare the effects of aerobic and resistance training on plasma homocysteine in individuals with type 2 diabetes. The study included 15 individuals undergoing aerobic training, 14 subjects undergoing resistance training, and 18 individuals in the control group. Homocysteine, total cholesterol and fractions, glucose, and anthropometric measurements were conducted. The training program lasted 16 weeks. Aerobic training was performed twice a week and lasted 75 min, and resistance training was performed twice a week and lasted 75 min. Homocysteine levels were not significantly different between before and after training. High-density lipoprotein levels increased in both training groups and decreased in the control group. Glucose levels decreased after aerobic and resistance training. Body fat mass (percentage and total) decreased in both training group, but with more expression in the aerobic group. We conclude that 16-week aerobic and resistance training programs did not significantly affect plasma homocysteine levels in patients with type 2 diabetes. Nevertheless, these training programs yielded positive results in HDL control, plasma glucose, and body composition.

Introduction

Diabetes is characterized by high blood glucose levels, which can be associated with reduced sensitivity of glucose receptors to insulin or changes in insulin production. In addition, diabetes favors the occurrence of other diseases such as atherosclerosis [1, 2]. Atherosclerosis involves the formation and accumulation of fatty plaques in blood vessels [3, 4]. The mortality associated with these diseases is high, and several studies have indicated that exercise-training programs can contribute to the control of risk variables [5], reducing the risk for atherosclerotic vascular disease in patients with non-insulin-dependent (type II) diabetes and in the general population [6]. In fact, it has been proposed that exercise both prevents and helps treat many atherosclerotic risk factors, including elevated blood pressure, insulin resistance and glucose intolerance, elevated triglyceride concentrations, low high-density lipoprotein cholesterol (HDL-C) concentrations, obesity [7], oxidative stress, and homocysteine [8]. Elevated plasma homocysteine concentration is a risk factor for cardiovascular disease, which seems to be the main cause of increased mortality in patients with type 2 diabetes [9]. Homocysteine is a thiol amino acid synthesized during the metabolic conversion of methionine to cysteine in the liver and is a risk factor for the development of fatty plaques [10]. It has been proposed that elevated homocysteine levels cause endothelial cell dysfunction and induces apoptotic cell death in cell types relevant to atherothrombotic disease, including endothelial cells and smooth muscle cells [11]. Studies on the effects of aerobic and/or resistance training programs on plasma homocysteine levels have yielded divergent results [12–14]. Predominantly aerobic training programs can either maintain [15] or increase homocysteine levels when comparing results before and after training [16–20]. According to Herrmann et al. [21], aerobic activities can increase homocysteine levels, and this increase may be associated with the duration and intensity of physical activity. Resistance training programs influence the metabolism of amino acids and proteins. Although the effect of resistance training on homocysteine levels remains unknown, increased energy demand can lead to decreased homocysteine levels [22–24]. At present, few studies have compared the effects of aerobic and resistance training programs on homocysteine levels. Therefore, the present study aimed to compare the effects of aerobic and resistance training programs on plasma homocysteine levels in individuals with type 2 diabetes.

Methodology

Participants characterization

This quasi-experimental and quantitative study involved the random recruitment of 47 female subjects diagnosed with diabetes type 2 and followed by the University Center of Itajubá. If patients were considered eligible for inclusion and accepted participation in the study, they were referred for randomization. These were randomized following a 1:1:1 scheme to one of three groups: aerobic training, resistance training and control. Randomization was implemented using the lottery method sampling with sequentially numbered opaque, sealed envelopes picked by each volunteer. Group A comprised 15 individuals undergoing aerobic training, group B comprised 14 individuals undergoing resistance training, and control group comprised 18 individuals. Table 1 shows the characteristics of the study groups.

Table 1.

Characteristics of the samples in mean and standard deviation

Variables	Aerobic	Resistance	Control	p value
Age (years)	69.68 ± 11.3	68.00 ± 6.3	67.61 ± 6.0	0.932
Weight (kg)	63.00 ± 12.3	72.60 ± 13.6	67.55 ± 9.6	0.224
Height (m)	1.52 ± 0.06	1.53 ± 0.04	1.53 ± 0,0	0.808
BMI	27.21 ± 4.32	30.28 ± 5.25	28.74 ± 4.4	0.358
Systolic Blood Pressure (mmHg)	134.37 ± 22.50	137.13 ± 19.79	128.88 ± 12.0	0.409
Diastolic Blood Pressure (mmHg)	81.25 ± 17.84	80.00 ± 6.5	77.50 ± 6.6	0.636

Inclusion and exclusion criteria

The exclusion criteria were smoking history, absence of established hypertension, and changes in medication throughout the study for the patients with type 2 diabetes. The inclusion criteria were the willingness to participate in the study and a sedentary lifestyle for at least 12 months. All study participants signed an informed consent form, stating their willingness to participate and their awareness of all the risks and benefits of participating in the study. The study was approved by the Research Ethics Committee of the University Center of Itajubá (Protocol 139), in accordance with Resolution of the National Health Counsel and Declaration of Helsinki (1975).

Instruments

Homocysteine analysis was performed using high-performance liquid chromatography (HPLC), with a confidence interval of 95% [25–27]. Glucose, total cholesterol, very-low-density lipoprotein (VLDL), low-density lipoprotein (LDL), and high-density lipoprotein (HDL) levels were measured using a spectrophotometer model SP-22 (Benfer, Brazil) (reading wavelength of 505 nm), in addition to the cholesterol PP kit (Labtest, Brazil) and the glucose PAP liquiform kit (Labtest, Brazil), using a confidence interval of 95% [3, 4]. Height was measured with a rigid PVC Seca® brand stadiometer with a retractable metal tape measure (220 cm in length). Body weight was measured on a scale (Filizola, Brazil) with a maximum capacity of 180 kg and 100 g fractions, with a base plate and rubber mat [28]. The percentage of body fat was determined with a Cescorf® scientific compass, with a constant pressure at 10 g/mm² and reading accuracy of 0.1 mm [28]. Blood pressure was measured using a HICO aneroid sphygmomanometer model HM-1001 and a HICO stethoscope model HM-3005 [3]. For the performance and assessment of resistance training, we used the following exercises and equipment: table bench, bars and dumbbells for upright row exercise, low rowing machine for leg curl, leg extension and leg press, pulley for triceps, and barbells and dumbbells for biceps [22, 29]. The instruments used for the performance and assessment of aerobic activity comprised a Movement® exercise mat and a Polar® heart rate monitor, respectively [30].

Procedures

Biochemical analysis

For the measurement of plasma homocysteine levels, 12-h fasting blood samples (10 mL) were collected before and after 16 weeks of the study by venipuncture of the antecubital vein. The blood samples were transferred to test tubes containing ethylenediaminetetraacetic acid (EDTA) and analyzed by HPLC. Immediately after collection, the plasma was separated by centrifugation and frozen at −20 °C until analysis. Quantification was performed using a diagnostic reagent (DPC Medlab, Brazil). Normal values ranged between 5.0 and 15.0 μM/L and the analytical sensitivity was 0.5 μM/L. Quantitation was performed by calculating the peak areas from the chromatograms [25–27]. The method of enzymatic reaction endpoint was used for the quantification of cholesterol, HDL, LDL, VLDL, and glucose levels. After collection, the blood samples was separated from the serum and plasma samples. The sera were used to measure the concentrations of total cholesterol, HDL, LDL, VLDL, triglyceride, and plasma glucose fluoride [3, 4].

Blood pressure and anthropometric measurements

The auscultatory method was used to measure blood pressure. Before the measurements, the subjects remained at rest for 5 min in the seated position [3]. Body mass index (BMI) was calculated using the following formula: weight/height². Subjects were required to wear light clothing and no shoes during measurements [28, 29]. The percentage of body fat was determined with Durnin and Womersley specific prediction equations using 4 skinfold measures (biceps, triceps, subscapular, and suprailiac) [31].

Diet prescription

During the study both groups (aerobic, resistance and control) were instructed to diet by a nutrition professional. The diet was prescribed according to the recommended daily intake. To meet the dietary patterns of survey participants, the method of Food Recall 24 h (R24h) was adopted. Upon becoming aware of the consumption pattern of the population to be studied, possible corrections were made by a specialized nutritionist. The eating habits and purchasing power of each participant were respected in order not to compromise health or interfere in the daily life of the groups studied [32, 33].

As a way of standardizing consumption, it was briefly assumed that all research subjects should eat at least four meals throughout the day, distributed as follows: breakfast, lunch, snack, and dinner (20%, 35%, 15% e 30%), respectively, of the caloric value ingested throughout the day and that included all food groups as recommended in the food pyramid. Meetings were established with a minimum of 15 days’ regularity to clarify doubts and correct any deviations in the pattern of oriented consumption [32, 33].

Evaluation of the training programs

Individuals completed 2 sessions of training for familiarization with the resistance exercise prior to the session test 1 Maximum repetition (1-RM). After the familiarization was performed the 1-RM test for quantification the intensity training [22, 29]. Bruce protocol was used to estimate VO₂max relative to body weight. The test was initially performed with a belt speed of 2.74 km/h, and a gradient of 10% in the first 3 min. Subsequently, the belt angle was tilted 2% every 3 min and the speed increased gradually for a total 10 steps. The following formula was used to calculate the relative VO₂max in women: VO₂max = 4.38 × T – 3.9 [30].

Training programs

The resistance training program began with a warm-up and elongation. The training program comprised 75-min sessions performed twice a week for 16 weeks, with an initial load of 60% and 1-RM. This program consisted of 3 sets of 8 to 12 repetitions. The increased load of 5 kg was established when subjects were able to complete 12 repetitions. A rest period of 1.5 min was established between each set. The session was completed with a 10-min relaxation and stretching. Nine exercises were performed: bench press, upright row, low row, leg curl, leg extension, leg press, and biceps and triceps curl [22, 29, 34]. The aerobic training program comprised 75-min sessions performed twice a week for 16 weeks. Training began with a 10-min warm-up and a 10-min elongation. Subsequently, subjects walked for 45 min, and the session was completed with a 10-min relaxation and stretching. Training intensity was set at 60%–70% maximum heart rate established for the 45 min of aerobic training. The intensity of aerobic training sessions was monitored by a heart rate monitor [15, 29].

Statistical analysis

The study design compared the plasma homocysteine levels during the aerobic and resistance training programs. The data were analyzed quantitatively using descriptive statistics (mean and standard deviation) and Shapiro-Wilk test was used to test data normality. To compare de interaction between the effect of training (before and after training) and groups, a General Linear Model for the repeated measures was used with Sidak adjustments for multiple comparisons. Effect sizes were interpreted according to partial Eta-Squared statistic. Statistical analysis was performed using SPSS Statistics software version 20.0, and the level of significance was p < 0.05.

Results

The present study analyzed the effects of exercise training on homocysteine levels. No significant differences in the variation of homocysteine levels was observed over time (before and after the 16 weeks of intervention) in the three groups (Table 2). No interaction effect of time*groups was observed plasma LDL, VLDL, triglycerides, total cholesterol, and lean body mass. The variation with time*groups was significant in plasma HDL (F = 3.916, p = 0.02), glycaemia (F = 4.991, p = 0.01), body fat percentage (F = 13.830, p = 0.00), and total body fat (F = 9.92, p = 0.00). While plasma HDL increased in both aerobic and resistance training groups, in the control group a decrease was observed. Plasma glycaemia, body fat percentage, and total body fat mass decreased in both aerobic and resistance training groups whereas in the control group have observed. Multiple comparisons using Sidak correction revealed differences in variation in body fat percentage both between aerobic group and resistance group (p = 0.031) and between aerobic group and control group (p = 0.021), and in body fat mass variation between aerobic group and resistance group (p = 0.029).

Table 2.

Body composition and metabolic related atherosclerosis in type 2 diabetic subjects before and after 16 weeks and effects between the groups program aerobic training, resistance and control

Variable	Group	Pre	Post	Partial Eta Square	Effect of Time*Group (p)
Homocysteine (mmol/L)	Aerobic	13.69 ± 6.1	12.31 ± 4.0
	Resistance	13.46 ± 2.9	12.87 ± 3.3	0.048	0.325
	Control	13.53 ± 6.6	13.87 ± 5.6
LDL (mg/dL)	Aerobic	108.43 ± 43.3	90.62 ± 30.9
	Resistance	129.14 ± 61.1	95.78 ± 51.3	0.079	0.151
	Control	121.27 ± 48.1	120.81 ± 52.0
HDL (mg/dL)	Aerobic	47.56 ± 14.5	56.12 ± 15.0
	Resistance	43.26 ± 11.5	58.28 ± 15.1	0.145	0.027*
	Control	46.38 ± 11.0	45.83 ± 10.4
VLDL (mg/dL)	Aerobic	33.12 ± 15.0	31.18 ± 12.7
	Resistance	41.92 ± 16.4	36.21 ± 11.3	0.031	0.490
	Control	33.11 ± 15.6	33.57 ± 12.6
Triglycerides (mg/dL)	Aerobic	164.68 ± 75.2	155.81 ± 63.6
	Resistance	209.33 ± 82.3	181.50 ± 57.0	0.029	0.511
	Control	164.72 ± 77.7	166.83 ± 63.4
Total cholesterol (mg/dL)	Aerobic	194.93 ± 40.3	177.93 ± 36.5
	Resistance	214.40 ± 58.4	190.28 ± 60.1	0.067	0.202
	Control	199.66 ± 44.2	200.22 ± 49.9
Glycemia (mg/dL)	Aerobic	118.93 ± 39.7	94.68 ± 27.8
	Resistance	123.46 ± 29.3	110.14 ± 23.8	0.178	0.011*
	Control	103.38 ± 26.7	114.33 ± 31.9
% Fat	Aerobic	38.22 ± 6.2	35.80 ± 5.7
	Resistance	42.67 ± 3.8	40.32 ± 4.6	0.376	0.000*
	Control	41.40 ± 3.8	41.63 ± 3.7
Lean mass (kg)	Aerobic	38.29 ± 5.0	38.62 ± 5.7‡
	Resistance	41.19 ± 5.5	42.85 ± 5.2	0.050	0.305
	Control	39.28 ± 3.8	39.79 ± 3.6
Fat mass (kg)	Aerobic	24.69 ± 8.0	22.30 ± 6.9
	Resistance	31.44 ± 8.4	29.70 ± 8.2	0.301	0.000*
	Control	28.25 ± 6.3	28.82 ± 6.3

*Significant difference (p < 0.05)

Discussion

The present study aimed to analyze the effects of aerobic and resistance exercise training on plasma homocysteine concentration after 16 weeks. No significant interaction effect on the variation of both types of exercise with time was observed. In general, our sample showed an elevated plasma homocysteine concentration which has been claimed to contribute to the development of several cardiovascular diseases, one of the main causes of increased mortality in patients with type 2 diabetes [9]. Average plasma homocysteine concentration of our sample is above normal values for women of this age [35]. Chen et al. [36] found that with age the expression of glutamate cysteine ligase in skeletal muscle under stress is impaired which could explain the elevated homocysteine of our sample, once at least one of the metabolic pathways of homocysteine is compromised. Altogether with some other studied variables such as, systolic blood pressure, triglycerides and total cholesterol, we may say that several women of this sample displayed an elevated risk for cardiovascular disease. Considering the main benefits of physical exercise described in literature, we would expect to observe a reduction of some of the studied risk factors. Concerning plasma homocysteine, literature is not consistent in relation to the effect of different types of exercise. Previous studies reported that aerobic training programs do not affect homocysteine levels [15–20], whereas other studies indicate that homocysteine levels increase after aerobic exercise [16, 17]. According to Herrmann et al. [21], aerobic activities can increase homocysteine levels, and this increase may be associated with the duration and intensity of training. These changes may be confusing apparently, because it has been described that elevated homocysteine may be a risk factor for cardiovascular disease, and if so, how can aerobic exercise exert a protective effect for cardiovascular disease if it increases plasma homocysteine? In fact, homocysteine plays an important metabolic role, but too much is toxic and too little results in metabolic problems [35]. Homocysteine is a non-protein α-amino acid, which is biosynthesized from methionine by the removal of a terminal methyl group and is a storage molecule for cysteine, the rate limiting amino acid for glutathione production, an important antioxidant molecule [37]. Therefore, it is known that aerobic exercise induces an increase in antioxidant protection, including, glutathione stores, which results from the trans-sulfuration pathway of homocysteine [38]. The magnitude of these changes is, however, heavily dependent on the metabolic cell imbalance induced by the exercise intensity and its frequency. It has been described that after strenuous exercise, reactive oxygen species increase, and glutathione expression increase possibly due to an increase of glutamate-cysteine ligase that catalyzes the conversion of glutamate and cysteine into γ-glutamyl cysteine [37]. However, in this study no significant variation in plasma homocysteine occur with aerobic exercise. Once exercise intensity was controlled, the main explanation for this fact is the possible age-related down-regulation of glutamate cysteine ligase and the low exercise frequency. In fact, it is possible that more than two sessions per week could induce significant changes in plasma homocysteine levels.

Considering resistance training, no significant variation was observed either. It has been suggested that resistance training can decrease homocysteine levels [22–24]. Resistance training programs have been reported to influence the metabolism of amino acids and proteins. Although the influence of resistance training in homocysteine levels remains poorly understood, increased energy demand can lead to decreased homocysteine levels [23, 24, 39]. Overload induced by resistance training stimulate different cell pathways and hormonal response comparatively to aerobic training [40]. Besides the metabolic role of cysteine, it is also used in the synthesis of contractile skeletal muscle protein such as myosin and actin [41]. It is known that resistance training induces mechanical damage to these and other proteins [42], increasing the use of cysteine that results from the trans-sulfuration pathway of homocysteine. Consequently, this might be one of the possible explanations for homocysteine reduction with resistance training described in literature. Nevertheless, it was not possible to observe significant variation of plasma homocysteine with resistance training in our sample. Once again, we may expect that significant variations may occur with a higher week frequency.

Of all the analyzed lipoproteins, HDL levels variance increased in both training groups and decreased in the control group. Partial Eta Square revealed that 14.5% of the variation could be explained by exercise. By contrast, VLDL, triglyceride, and total cholesterol levels variance were not significantly different between groups. These results are in line with others studies that found significant changes in HDL with aerobic [43] and with resistance training [44, 45], which suggest that these changes are the result of metabolic improvement with exercise. However, other studies did not find changes in plasma levels of LDL and HDL [23, 46–51]. These differences may be associated with the period, volume and intensity of training, and study sample characteristics. Considering that, although the differences were not significant for some variables, the training program showed a good metabolic control.

Both types of exercise were also significant in lowering glucose levels, which contrasts with the control group, in which it increased. Training explained 17.8% of the observed. It is known that metabolic changes induced by exercise (increased sarcoplasmic calcium and excitation-contraction coupling) stimulate vesicles containing glucose transporter 4 (GLUT4) from intracellular stores to the sarcolemma and T-tubules, allowing for an increase in facilitated glucose uptake [52]. In this manner, the translocation of the transporter and glucose uptake occurs. In addition to stimulating the signaling pathway, there is an improvement in insulin sensitivity with its receptor, leading to a decrease in plasma glucose [46, 53]. Therefore, aerobic and resistance training programs can help to control plasma glucose.

The body fat percentage and body fat mass decreased in both training groups in comparison with the control group. Moreover, 37.6% of changes in body fat percentage are explained by exercise. Multiple comparison also revealed that changes were more pronounced when the aerobic exercise was performed. In obese individuals, glucose uptake is compromised. With the increase in plasma free fatty acids, there is an inactivation of the cells’ receptors for insulin, hence less stimulus of the insulin signaling pathway occur. Therefore, the increase in body fat can lead to insulin resistance [54, 55]. As exercise induces an increase in metabolism which depends on, both during and after exercise, fats, it is probable that changes in body fat mass (total and percentage) have also contributed to improve cells glucose uptake. As aerobic training favors fat burn, the effects of this type of training were greater.

We conclude that 16-week aerobic and resistance training programs did not significantly affect plasma homocysteine levels in patients with type 2 diabetes. Nevertheless, these training programs yielded positive results in the control HDL, plasma glucose, and body composition.

Abbreviations

HPLC

High-performance liquid chromatography

VLDL

Very-low-density lipoprotein

LDL

Low-density lipoprotein

HDL

High-density lipoprotein

EDTA

Ethylenediaminetetraacetic acid

BMI

Body mass index

1-RM

1 Maximum repetition

Funding information

This work was financed by National Funds through FCT - Foundation for Science and Technology under the project UID/DTP/04045/2019.

Compliance with ethical standards

Conflict of interest

The authors declare that there is no conflict of interest associated with this manuscript.