Introduction

Today the importance of inorganic entities of life is being more clearly understood through research in different areas of life sciences. Some elements such as sodium (Na), potassium (K), magnesium (Mg), and calcium (Ca) are the main cations acting as electrolytes and essential constituents for life. Trace elements such as iron (Fe), copper (Cu), zinc (Zn), selenium (Se), and chromium (Cr) perform unique roles and act in numerous metabolic tasks such as electron transport (e.g., Cu, Fe), catalysis (e.g., Cu, Zn, and Se), and maintenance of structural stability of proteins. Concentrations of trace elements and electrolytes in various biological samples (e.g., serum, tissues, urine) are of interest [1-4], since their levels should be maintained within physiological limits as well as at the proper location in a timely fashion. Homeostasis of trace elements has been associated with diverse factors such as aging [1], exposure to xenobiotics [2], and exercise [4, 5].

Regular and moderate physical activity has been widely recognized as an integral component of a healthy lifestyle. A recent study suggested that exercise is so effective that it may be considered as a drug and emphasized the importance of dosing and individual variations [6]. Moderate activity is generally considered better than none at all [7]; however, vigorous or intense physical activity may sometimes cause a disturbance in the homeostasis [7-9]. Elemental fluctuations in the body during physical performances have been a point of interest [10, 11]. In a recent report [12], plasma Zn concentrations have been measured below normal levels, suggesting a possible Zn deficiency in elite swimmers. It has been demonstrated that urinary excretion of Zn and Cu has been increased during physical exercise [13]. In another study conducted in rats, plasma Zn and Se levels have been found to decrease immediately after acute swimming [14]. It has been reported that decrease in Cu and Zn levels in kidney tissue of aged rats may be recovered by long-term swimming exercise [15].

This study was designed at two dimensions: First one probes the effect of training frequency on the serum concentrations of some trace elements and electrolytes in different-level swimmer groups (i.e., the elite swimmers, amateurs, and sedentary controls), while the second one focuses on the elemental alterations at three different sampling time points, i.e., pre-test, post-test, and 1 h later (within-group comparisons). To the best of our knowledge, this time-dependency, influence of training frequency, and different proficiency levels have not been examined previously in the same study with regard to serum element levels.

Materials and Methods

Study Design and Groups

A total of 35 male volunteers were assigned into three groups (Table 1) according to their swimming frequencies: elite swimmers (n = 14), amateur swimmers (n = 11), and sedentary individuals (n = 10). The elite and amateur swimmer groups were subjected to a controlled 3-week training program before blood collection, while sedentary individuals (control subjects) were not included in the training period. The schedule for elite swimmers was double training 6 days per week, while the amateur swimmers had a training frequency of 3 times a week. A typical training consisted of a free warming up process for 10 min, followed by 2,000–2,500 m (depending on the condition of the swimmer on the day) interval swimming (100 m loaded freestyle swimming and 20 s resting). The uninterrupted swimming capacities of swimmers were determined as 800 m for elite swimmers and amateurs, whereas 400 m for sedentary control group, and the test was performed accordingly. Volunteers in all groups were subjected to the test, and serum element levels were monitored at three different time points (i.e., pre-test, post-test, and 1 h later). The 1-h interval was chosen to ensure hemodynamic stabilization [16]. The dietary and supplementation information has been given by the volunteers, and all volunteers were asked to stop taking any supplementation at least 1 week before sampling. All participants were informed about the objectives and methods of study and were asked to give written voluntary informed consent. The study protocol was approved by the local ethics committee of Selcuk University Meram Medical School, Konya-Turkey (protocol number: 2009/032).

Table 1 Demographic characteristics of the study groups
Full size table

Sampling and Sample Preparation

Blood samples were collected from the antecubital vein of volunteers into venous blood vacuum serum collection tubes, and allowed to stay for 30 min at room temperature before serum separation. After centrifugation at 3,000 rpm for 10 min, serum samples were pipetted out into 2.0-mL volume clean storage tubes and stored at −80 °C until analysis. All glassware and plastic ware were cleaned with 10 % HNO3 at least overnight and gently rinsed with ultrapure water. Serum samples were allowed to thaw at room temperature and vortexed for 2 min. Samples were diluted 14-fold in polyethylene tubes for determination of trace elements, while 100-fold dilution was applied for determination of electrolytes prior to analysis. The tubes were vortexed and placed in the autosampler to be analyzed immediately.

Instruments and Working Conditions

An Agilent 7700x inductively coupled plasma mass spectrometry (Tokyo, Japan) with a collision/reaction cell (CRC-ICP-MS) was used throughout the study. The calibration standards were prepared by dilution of multi-element stock standard solution of trace elements (Agilent Technologies, Santa Clara, CA, USA) and electrolytes (Jenway, Essex, England). The internal standards containing 203Bi, 72Ge, 115In, 6Li, 175Lu, 103Rh, 45Sc, and 159Tb (Agilent Technologies) were on-line merged into the sample flow to negate the possible mixing problems. Nitric acid (69 % TraceSELECT®) was purchased from Sigma (St. Louis, MO, USA). Ultrapure water was produced from Milli-Q system (Millipore, Billerica, MA, USA). The instrumental conditions were set as follows: plasma power 1,600 W; plasma gas flow rate 15 L/min; carrier gas flow rate 1.15 L/min; and sampling depth 8 mm; isotopes monitored 52Cr, 57Fe, 63Cu, 66Zn, 78Se, 23Na, 24 Mg, 39 K, and 43Ca; and repetition three. Collision cell was pressured with helium at a flow rate of 5.0 mL/min. The octopole and quadrupole voltages were adjusted to −18 and −15 V, resulting in a kinetic energy discrimination of 3 V.

A graphite furnace atomic absorption spectrometry (GF-AAS) system (Perkin Elmer AAnalyst 100, Norwalk, CT, USA) with pyrolytically coated graphite tubes (Perkin Elmer Part no. B010-9322) was used as a comparative instrumental technique. The analysis of serum samples by GF-AAS were performed in the recommended conditions of the manufacturer. Samples were diluted 50 times for Cu, Zn, and Fe; and 20 times for Se determination using 1 % HNO3 containing 0.1 % Triton X-100. Serum Cr determination was performed as described [17] with a slight change. As diluent, 0.1 % Triton X-100 solution was used instead of 0.02 % CTAC, and the injection volume was applied to be 50 μL to increase absorbance signal [18], giving a limit of detection of 0.03 ng/mL.

Statistical Analysis

Statistical analyses were performed using GraphPad Prism 6.01 (La Jolla, CA, USA). Data were analyzed for significant differences between the three study groups using one-way ANOVA followed by Tukey’s post hoc test. Repeated measures ANOVA followed by post hoc test was performed for the assessment of intraindividual changes in elements at different time points (pre-test, post-test, 1 h later). Differences were accepted statistically significant when the p value was <0.05.

Results

In the present study, as shown in Table 1, there were no significant differences in elite swimmers, amateur swimmers, and controls relating to age and body mass index. However, as a consequence of the study design, the elite swimmers were engaged in regular swimming training for a longer period as compared to amateur swimmers. Serum electrolyte and trace element levels of study groups are presented in Figs. 1 and 2, respectively. The elements were within the reference values reported [19] with minor exceptions. Among all the elements analyzed, Na presented only one result close to upper limit of the reference range [19]. Three individuals in the elite swimmer group had K levels higher than the reported [19] reference values (136.5–198.9 μg/mL, corresponding to 3.5–5.1 mmol/L). Regarding the serum Cr concentrations, although all results were within the reported ranges [19], low (0.109 ng/mL) and high (0.276 ng/mL) values were found in pre-test data of elite swimmers resulting in a wider distribution in this group. Exclusion of these data revealed no change in mean values (0.187 ± 0.01 ng/mL, n = 12 vs 0.188 ± 0.03 ng/mL, n = 14) as well as in group comparisons. Besides, Zn levels were lower than the limits of the reference range [19] in the samples obtained 1 h after the test in all groups. It should be mentioned that there was neither Fe deficiency nor Fe overload in any of the volunteers included in the study throughout the period (Fig. 2).

Fig. 1
figure 1

Serum concentrations (mean ± standard deviation) of some electrolytes (Na, K, Mg, Ca) evaluated in study groups set as elite swimmers (n = 14), amateur swimmers (n = 11), and sedentary controls (n = 10). Three different sampling time points were selected: pre-test (before swimming), post-test (immediately after swimming), and 1 h after the activity. The p values were obtained with one-way ANOVA for inter-group comparisons and repeated measures ANOVA for within-group differences followed by post hoc test (Tukey’s), and differences between groups were judged to be statistically significant when the p value was <0.05 (GraphPad Prism 6.01)

Full size image
Fig. 2
figure 2

Serum concentrations (mean ± standard deviation) of some trace elements (Cr, Fe, Cu, Zn, and Se) evaluated in study groups set as elite swimmers (n = 14), amateur swimmers (n = 11), and sedentary controls (n = 10). Three different sampling time points were selected: pre-test (before swimming), post-test (immediately after swimming), and 1 h after the activity. The p values were obtained with one-way ANOVA for inter-group comparisons and repeated measures ANOVA for within-group differences followed by post hoc test (Tukey’s), and differences between groups were judged to be statistically significant when the p value was <0.05 (GraphPad Prism 6.01)

Full size image

Regarding the inter-group comparisons, serum post-test Na levels were significantly higher than the 1-h-later-sampling time point in the control group (p < 0.05); however, there was no statistical difference in other groups in this regard (Fig. 1). In all groups, serum Mg concentrations were insignificantly elevated after the swimming test (post-test mean values) as compared with the corresponding pre-test results, whereas the concentrations were significantly reduced (p < 0.05) at 1-h-later samples in elite and amateur swimmer groups. There were no significant differences in the Mg concentrations within control group results (Fig. 1). The pre-test serum K and Ca levels were significantly higher in the elite swimmer group than the amateur swimmers and sedentary controls (Fig. 1, p < 0.05). With regard to K levels, the post-test values in elite swimmers were significantly decreased as compared to the pre-test status (p < 0.05); however, the samples obtained 1 h after the test in this group showed a significant increase when compared to the post-test results (p < 0.05), exhibiting a rebalancing effect. On the other hand, amateur swimmers and control group showed elevated K levels 1 h after the test as compared to pre- and post-test results (p < 0.05). With respect to within-group differences for Ca, the post-test serum Ca levels of elite and amateur swimmers were significantly higher than 1-h-later-sampling results (p < 0.05); however, there was no statistical difference in the control group in this regard (Fig. 1).

As depicted in Fig. 2, post-test serum Cr concentrations were significantly lower in elite swimmers than in amateur swimmers and sedentary controls (p < 0.05). Similarly, the serum Cr levels of elite swimmers obtained 1 h later were significantly lower than the corresponding value of amateur swimmers (p < 0.05). Interestingly, serum Cr levels revealed no within-group differences. When the within-group Fe concentrations were compared, it was found that the levels were slightly higher in the post-test results than the pre-test and 1-h later values; however, there was no statistical difference in this regard (Fig. 2). It is of interest that the mean serum Cu concentrations of elite swimmers obtained 1 h after the test were significantly lower than the pre-test and post-test levels (p < 0.05), whereas in the other groups, pre-test and 1-h-after results did not significantly differ. The mean pre-test Cu concentration was lower in amateur swimmers as compared with elite swimmers (p < 0.05), while there was no statistical difference between elite swimmer and control groups (Fig. 2). In amateur swimmer and control groups, the mean post-test Cu levels were higher than pre-test and 1-h-after values (p < 0.05). Regarding Zn concentrations (Fig. 2), the post-test Zn levels were significantly higher than pre-test and 1-h-after results (p < 0.05), while the mean Zn concentrations obtained 1 h later were significantly lower than the pre-test levels (p < 0.05) in all groups. As for serum Se concentrations, in elite swimmers, 1-h-after levels were lower with regard to pre-test and post-test values (p < 0.05). On the other hand, in the amateur swimmers group, post-test values were higher than the pre-test and 1-h-later levels (p < 0.05); despite a lack of statistical significance, a similar profile was noted in the control group for Se concentrations.

In regard with analytical methodology, some figure of merits including limit of detection (LOD) values of the studied elements, together with the internal standards giving most accurate results, are shown in Table 2. The spike recovery experiments showed that the recovery values were acceptable. An excess value in Se recovery, probably due to the increased ionization of Se in carbon-containing samples [20], was simply corrected after data acquisition. In addition, comparative analysis of pooled sera was performed by ICP-MS and GF-AAS for trace elements. The results (mean ± confidence interval, expressed as nanogram per milliliter; n = 4) obtained by ICP-MS vs. GF-AAS were highly consistent with each other at 95 % confidence level: Cr (0.185 ± 0.006 vs. 0.182 ± 0.020), Fe (970.4 ± 8.1 vs. 975.3 ± 13.6), Cu (735.3 ± 5.7 vs. 736.8 ± 14.5), Zn (667.5 ± 6.4 vs. 669.0 ± 10.7), and Se (133.5 ± 1.0 vs. 133.8 ± 3.6). Recovery values of internal standards were very stable during the analysis of 136 samples for 5 h, underpinning the robust plasma conditions (data not shown). A remarkable point of the method is the direct “dilute and shoot” procedure, allowing negligible contamination and low limit of detection values.

Table 2 Some figure of merits for determination of the analytes in serum samples using internal standards
Full size table

Discussion

In this study, the differences in serum element levels between study groups could obviously be attributed to different training schedules and intensity followed. Between-group comparisons revealed higher pre-test levels of serum Ca and K in elite swimmers as compared to other groups (Fig. 1, p < 0.05) which may be related to adaptive mechanisms. Swimming has been shown to alter the expression of duodenal genes pertaining to transport of some ions including Ca and K as well as water homeostasis [21]. Therefore, higher pre-test serum Ca levels in elite swimmers (Fig. 1), at least in part, may be related with the enhanced intestinal Ca absorption via upregulation of Ca transporter genes as shown in rats [21]. In order to retain a long-term effect in Ca absorption, scheduled training program may be suggested; however, the stimulatory effect of swimming on duodenal calcium absorption in rats has been found to be hindered by forced training or prolonged stress [22]. Consequently, a key point with this regard may be to consider exercise as a drug and the importance of dosing and individual variations [6]. The Cr levels also exhibited inter-group variances, especially in post-test results of elite swimmers, with lower levels, as compared to amateur swimmers and sedentary controls (Fig. 2). A decrease in mean Cr concentrations in blood samples after the marathon has been reported in athletes [5]. An extensive review in this field [23] indicates that endurance activity may exert a negative influence on Cr status possibly through enhancing urinary Cr excretion.

On the other hand, the serum element levels were evaluated at three sampling time points with regard to test (i.e., pre-test, post-test, and 1 h later). The significant fall in post-test K levels of elite swimmers (Fig. 1, p < 0.05) and a very slight change in the other groups may be attributed to improved uptake of plasma K+ into muscle cells by training-induced upregulation of Na+,K+ pump [24-27]. Serum K levels have been shown to decrease immediately after training in male field hockey players, while higher concentrations have been observed 1 h later [16]. The significantly higher concentrations of K after 1 h as compared to post-test levels in all groups (Fig. 1) may be a consequence of test [28]. Since human skeletal muscles contain the largest single pool of K+ in the body, even modest relative release may cause marked changes in plasma K concentrations [24].

Another interesting point regarding within-group differences was the similar pattern in Mg, Ca, Fe, Cu, Zn, and Se levels, showing higher post-test values in all groups. This finding might be explained as a consequence of exhaustion and water loss. As shown in Fig. 1, elite swimmer and amateur swimmer groups exhibited increased post-test Mg values which are consistent with previous studies underlining higher Mg levels after exercise [4, 10]. Westmoreland et al. have reported that high-intensity exercise induces an increase in plasma Mg, possibly as a compensatory mechanism that involves rapid mobilization of Mg from apatite crystals in bone [10]. Our finding regarding the lower serum Mg concentrations at 1-h-later samples of elite and amateur swimmer groups (p < 0.05) was in line with a previous report in male field hockey players [16]. On the other hand, the only observed alteration in regard to Na in the control group 1 h later (p < 0.05) may be a consequence of its stringent control in healthy individuals.

Present results highlight lower serum Cu, Zn, and Se concentrations in elite swimmers at 1 h after the test as compared with the pre-test and post-test levels (Fig. 2). Synergic role of Cu, Zn, and Se has been well known, due to their essential functions in antioxidant enzymes. Supplementation of Zn at physiologic doses to athletes, particularly in periods of intense exercise, has been suggested to be beneficial to antioxidant system and to physical performance [29]. Plasma Zn concentrations below normal levels in elite swimmers have been previously reported [12]. It is of note that in all groups, Zn concentrations 1 h after the test (Fig. 2) were found below reference range [19], pointing out a generalized potential Zn deficiency. In young amateur boxers, decline in plasma Cu and Zn levels after acute exercise and decrease in Zn concentrations after 4 weeks of regular boxing training have been shown [30]. This effect may be related to element loss since an increased urinary excretion of Cu and Zn during exercise has been demonstrated [13]. Other mechanisms such as unbalanced diet or Zn redistribution between plasma and erythrocytes may also contribute to this alteration. With regard to redistribution, significant decreases in the erythrocyte content of Zn as a consequence of acute effects of prolonged exercise and increase in hepatic Zn levels after daily physical training have been reported [31]. Moreover, the change is more elevated at the end of a competitive season [31].

It can be concluded that, apart from the swimming test applied, observed inter-group variations between pre-test levels of study groups may be a consequence of training frequency. On the other hand, within-group variations might be related to dehydration and compartmental element shifts due to swimming test. Although the elemental homeostatic alteration might be usually transient, it could be sufficient to predispose the individual to become more susceptible to potential health concerns. Therefore, regular control of elemental status to be aware of the potential impacts of strenuous exercise can be suggested to individuals especially engaged professionally in sports. Integration of the present methodology with approaches such as metallomics [32, 33] to probe such stimuli is among our future work.