When performed at a sufficient intensity, resistance exercise is known to induce skeletal muscle damage (1) and an elevation in markers of oxidative stress (2). As a result, antioxidant supplementation has been used to potentially reduce this exercise-induced stress (3). Polyphenols are micronutrients with antioxidant properties that are common in many foods including tea (4) and have also been examined for their potential benefits associated with recovery from exercise-induced muscle damage (3,5,6). Previous studies have shown favorable results with regard to improved maintenance of muscular force and reduced markers of muscle damage after resistance exercise with polyphenol supplementation (3). However, recovery from exercise-induced muscle damage, as well as the adaptation to exercise training, is more complex than simply limiting the initial insult to skeletal muscle tissue.
An adequate resistance exercise initiates an inflammatory cascade that results in the secretion of various chemotactic factors (7), which in turn recruit specific immune cells (8,9). The two primary immune cells of interest are the neutrophil and monocyte/macrophage, which are recruited during three primary phases of recovery: preliminary, early, and late (10). While neutrophils are the predominant cell involved with the preliminary phase (10), the monocyte/macrophage is the primary cell of interest for both the early and late phases of recovery (10).
Macrophages, which represent a developmental transition of circulating monocytes within the tissue, can be subdivided into two primary subsets, M1 and M2 (11). M1 macrophages are most prominent during the early phase of recovery, while M2 macrophages are prominent during the late phase (10). Given that both subsets produce insulin-like growth factor 1 (12), macrophages play a prominent role in skeletal muscle recovery (10). Moreover, considering that M1 macrophages promote a cytokine milieu conducive to the proliferation of myogenic precursor cells, and M2 macrophages promote myogenic precursor cell differentiation (13), the timely transition between these phases may be of utmost importance during muscle adaptation resulting from exercise stress.
Although disputed (13), the development of macrophage subsets may be directly linked to the circulating monocyte subset of origin (14). Recently, monocytes have been formally organized into a three-subset paradigm (15), replacing the two-subset model of classical and nonclassical monocytes (16). Phenotypically, monocytes are identified based on their expression of the lipopolysaccharide receptor, cluster of differentiation (CD) 14, and the FcγRIIIa receptor, CD16. Though both classical (CD14++/CD16−) and nonclassical (CD14+/CD16++) monocytes have been suggested to give rise to M1 and M2 macrophages, respectively (13,14), a formal role of intermediate (CD14++/CD16+) monocytes within the resolution of skeletal muscle damage have yet to be identified.
Typically, the total monocyte population, which accounts for approximately 5% to 10% of the total leukocyte population (17), consists of 80% to 90% classical, 5% to 10% intermediate, and 5% to 10% nonclassical monocytes (15,16). The proportions of these subsets are altered immediately after exercise; characterized by reduced proportions of classical monocytes, with a concomitant increase in the proportion of nonclassical monocytes (17,18). At 1 h after exercise, this modulation of monocyte subset proportions has been shown to return to baseline (17), or super-compensate (increased classical monocyte proportion, and reduced nonclassical monocyte proportion) (18). To date, only one investigation has examined the change in classical monocytes after resistance exercise, indicating increased proportions of classical monocytes from 1 to 5 h into recovery (19). However, this prior investigation (19) did not examine the intermediate or nonclassical response to resistance exercise. We expect the response of intermediate and nonclassical monocytes to be different after resistance exercise compared to aerobic exercise due to increased muscle damage (6,20). To our knowledge, no other investigations have characterized the intermediate or nonclassical monocyte response to resistance exercise.
Recruitment of classical and nonclassical monocytes is accomplished primarily by monocyte chemoattractant protein-1 (MCP-1) (11,16,21), and fractalkine (CX3CL1) (22), respectively. Although existing data examining the postresistance exercise response of these specific cytokines is limited, elevations of both MCP-1 and CX3CL1 within skeletal muscle have been reported after resistance exercise for up to 4 and 2 h, respectively (8). Furthermore, data involving MCP-1 in circulation have been conflicting, with both a postexercise increase (23) and decrease (24) demonstrated previously, while no investigation has examined circulating CX3CL1 after resistance exercise.
In addition to changes in monocyte subset proportions and recruitment after resistance exercise, changes in cellular activation of these cells have also been documented (23,25). Complement receptor 3 (CR3) is a β2 integrin that is composed of CD11b and CD18, and involved in the late phases of transendothelial migration to damaged tissue (26). Therefore, CD11b plays a crucial role in the transmigration of monocytes into damaged tissue. Previously, polyphenol supplementation has resulted in decreases in CD11b expression at rest in individuals at high risk for cardiovascular or metabolic disease (27). Furthermore, in vitro incubation of monocytes and neutrophils in polyphenols has resulted in reduced expression of adhesion molecules and reduced chemotaxis (28). Consequently, it is possible that polyphenols may limit the expression of CD11b on monocytes during and after exercise.
To date, few investigations have examined CD11b or CR3 expression on monocytes after resistance exercise or muscle damage, yielding little consensus of a time course of elevated expression. Briefly, dynamic resistance exercise has been demonstrated to result in increased expression of CD11b during the first hour of recovery (23,25), though after exercise designed to elicit muscle damage, CD11b expression has been reported to both increase (29) or remain unchanged (30). To our knowledge, no investigation has examined how polyphenol supplementation may modify the postexercise response.
Therefore, the primary purpose of this investigation was to examine the impact of polyphenol supplementation on the overall recruitment of monocytes after resistance exercise, in addition to the overall expression of CD11b on monocytes. Furthermore, we aimed to detail the postresistance exercise recruitment, modulation and activation of monocyte subsets. We hypothesized that resistance exercise would result in a significant increase in the proportion of nonclassical monocytes immediately after resistance exercise, and that classical monocyte proportions would display a super-compensation at 1 h that remains elevated for up to 5 h into recovery. Furthermore, we anticipated polyphenol supplementation would reduce the CD11b expression on monocytes after resistance exercise.
Recreationally active, but nonresistance trained men (n = 38; 22.1 ± 3.1 y; 173.9 ± 7.9 cm; 77.8 ± 14.5 kg) volunteered to participate in this study. Sample size was determined based on creatine kinase (CK) concentrations from a previous investigation using the same supplement (5). After the explanation of all procedures, benefits, and risks, participants gave their informed written consent before participation, and all procedures were approved by the New England Institutional Review Board. For inclusion in this investigation, participants had to engage in less than 3 h·wk−1 of planned exercise, have a body mass index of between 18.5 and 34.9 kg·m−2, be free from physical limitations, and willing to maintain a normal diet while abstaining from tea, alcohols and additional dietary supplements. No other restrictions were placed on participants’ diets, though daily consumption was monitored and confirmed similar between groups via daily food logs (6).
For this randomized, placebo (PL)-controlled, between-subjects investigation, participants were assigned to one of three groups: proprietary polyphenol blend (PPB), PL, and control (CON). Participants reported to the Human Performance Laboratory for 4 d of testing (Fig. 1). Before the first day of testing, PPB and PL completed 28 d of supplementation. During day 1, participants completed one-repetition maximum (1-RM) testing of the squat, leg press and leg extension exercises. At least 72 h later, participants returned to the laboratory after a 12-h fast and provided a resting blood sample and skeletal muscle biopsy (PRE). Participants were then provided a small breakfast bar (Cal, 190; CHO, 19 g; protein, 7 g; fat, 13 g) with limited polyphenols (4), followed by the acute exercise protocol (PPB and PL) or rest for 1 h (CON). Participants provided blood samples immediately (IP), 1 h (1H) and 5 h (5H) postexercise, and skeletal muscle biopsies at 1H and 5H. After the 1H samples were obtained, participants were provided a small, standardized meal with limited polyphenols (4) (Cal, 250; CHO, 34 g; protein, 14 g; fat, 6 g). Participants returned to the laboratory in a fasted state, and provided resting blood samples at 24 h (24H) and 48 h (48H) postexercise, in addition to a skeletal muscle biopsy at 48H.
Both groups were supplemented daily for 28 d with either a proprietary PPB or PL (Kemin Foods, L.C., Des Moines, IA). Both participants and investigators were blinded to the actual group assignments. The PPB group consumed a blend of water-extracted green and black tea (Camellia sinensis) containing at minimum 40% total polyphenols, 1.3% theaflavins, 5% to 8% epigallocatechin-3gallate, 7% to 13% caffeine, 600 ppm manganese. The PL group consumed microcrystalline cellulose in capsules of similar shape and size. All products were tested for toxins including heavy metals and pesticides by an independent third party.
During the supplementation period, participants reported to the laboratory between 3 and 5 d·wk−1 to receive their supplement. Participants took one dose (1000 mg PPB, or PL) under the supervision of a member of the research team, and then were provided their remaining doses in individual containers (1000 mg PPB or PL for each additional time point). Participants consumed two doses daily, for a total of 2000 mg of either PPB or PL, and were asked to return all empty containers to the laboratory on their next visits to monitor compliance. A dose was considered to be administered if the participant consumed the dose in front of the study staff, or returned their empty container and verbally confirmed they had taken the dose. Supplementation continued throughout the acute exercise protocol, and during the 4 d of recovery. Participants that did not maintain 80% compliance for each phase of the study (the 28 d of supplementation, during the acute exercise protocol, and the recovery period from the acute exercise) were removed from the analysis.
Acute exercise protocol
The acute exercise session was completed by PPB and PL, while CON rested for 1 h. The protocol designed to elicit muscle damage was preceded by a light warm-up identical to the 1-RM session. Participants then completed six sets of 10 repetitions of the squat, as well as four sets of 10 repetitions of the leg press and leg extension exercises. All exercises were completed at 70% of the participant’s previously determined 1-RM, with 90 s of rest between sets. If participants were unable to complete 10 repetitions, they were provided with assistance, and the weight on subsequent sets was reduced. All testing sessions, including 1-RM, were overseen by a Certified Strength and Conditioning Specialist to monitor adherence to exercise technique.
Blood samples were obtained at seven time points throughout the study (PRE, IP, 1H, 5H, 24H, and 48H). The PRE, IP and 1H blood samples were obtained using a Teflon cannula placed in a superficial forearm vein using a three-way stopcock with a male luer lock adapter and a plastic syringe. The cannula was maintained patent using an isotonic saline solution (Becton Dickinson, Franklin Lakes, NJ). The PRE and 1H blood samples were obtained after a 15-min equilibration period, while IP blood samples were taken within 5-min of exercise cessation. The remaining time points (5H, 24H, and 48H) were obtained by a single-use disposable needle with the subject in a supine position for at least 15 min before sampling. Whole blood (20 mL) was collected in two Vacutainer® tubes (Becton Dickinson), one containing K2EDTA, and one containing no anti-clotting agents. Aliquots were removed from the first tube for hematocrit and hemoglobin measures, as well as flow cytometry analysis, while the second tube was allowed to clot for 30 min before being centrifuged at 3000g for 15 min. The resulting plasma and serum were aliquoted and stored at −80°C for later analysis.
Analysis of circulating analytes
Plasma concentrations of monocyte-chemoattractant protein-1 (MCP-1), and fractalkine (CX3CL1) were analyzed via multiplex assay (EMD Millipore, Billerica, MA). All samples were thawed once and analyzed in duplicate by the same technician using the MagPix (EMD Millipore), with an average coefficient of variation of 6.84%, and 7.18% for MCP-1, and CX3CL1, respectively. Serum concentrations of myoglobin and CK were assessed as previously reported (6).
Fine needle skeletal muscle biopsy procedure
Fine needle muscle biopsies were performed on the vastus lateralis muscle of the participant’s dominant leg using a spring-loaded, reusable instrument with 14-gauge disposable needles and a coaxial introducer (Argon Medical Devices Inc., Plano, TX). After local anesthesia with 2 mL of 1% lidocaine applied into the subcutaneous tissue, a small incision to the skin was made and an insertion cannula was placed perpendicular to the muscle until the fascia was pierced (31). Each muscle sample was removed from the biopsy needle using a sterile scalpel and was subsequently placed in a cryotube, rapidly frozen in liquid nitrogen, and stored at −80°C. A new incision was made for each time point, with approximately 2 cm between all sampling sites. All muscle biopsies were performed by a licensed medical physician.
Intramuscular cytokine protein content
Sufficient sample was not obtained from nine participants (PPB, 2; PL, 3; CON, 4), and therefore were not included in the intramuscular analysis. Tissue samples were thawed and kept on ice for preparation and homogenization. A proprietary lysis buffer with protease inhibitor (EMD Millipore) was added to each sample at a rate of 500 μL per 10 mg of tissue. Samples were homogenized using a Teflon pestle and sonication (Branson, Danbury, CT). Tissue samples were then agitated for 10 min at 4°C and centrifuged at 10,000g for 5 min. The supernatant was then aspirated and used for analysis.
Total protein content was assessed using a detergent compatible (DC) protein assay kit (Bio-Rad, Hercules, CA, USA), and samples were diluted to 0.8 to 1.2 mg·mL−1. The protein content of MCP-1 was then assessed via multiplex assay (EMD Millipore) per manufacturer’s guidelines, and normalized to the total protein content. To limit inter-assay variance, all tissue samples were analyzed in duplicate in the same assay run by a single technician, with an average coefficient of variation of 9.37%. Intramuscular cytokine protein content is expressed in pg/μg total protein (8).
Monocyte subset preparation
Fresh, anti-coagulated (K2EDTA), whole blood (100 μL) was mixed with fluorescent-conjugated monoclonal antibodies specific to CD11b-fluorescein isothiocyanate (Biolegend, San Diego, CA), CD66b-phycoerythrin, CD14-PerCP Cy5.5, and CD16-allophycocyanin (BD Biosciences, San Jose, CA, USA) within 10 min of blood sampling. Samples were mixed and incubated for 15 min in the dark, after which, the samples were lysed with 2 ml of 1× FACS lysing solution (BD Biosciences), mixed and incubated in the dark for an additional 8 min. After incubation, samples were centrifuged at 300g for 8 min and washed with 2 mL of 1× wash buffer containing 1% fetal bovine serum in a 1× phosphate-buffered saline solution. Samples were centrifuged again at 300g for 8 min, and the supernatant was removed. Samples were then fixed in 300 μL of 2% paraformaldehyde in phosphate-buffered saline.
Flow cytometry analysis
Cell preparations were acquired using an Accuri C6 flow cytometer (BD Accuri Cytometers, Ann Arbor, MI) equipped with two lasers providing excitation at 488 and 640 nm, and four band pass filters (FL1, 533/30; FL2, 585/40; FL3, 670LP; FL4, 675/25). Events were recorded based on size (FSC-A), complexity (SSC-A) and mean fluorescence intensity. A total of 200 μL were collected for each sample, with at least 8000 CD14+ events. If 8000 CD14+ events were not obtained, the sample was removed from analysis.
The analysis was completed using BD Accuri analysis software (BD Accuri Cytometers, Ann Arbor, MI). Events were gated as depicted in Figure 2. Initially, cells were discriminated based on SSC-H and SSC-A as a multiplet cell exclusion criteria. Monocytes were then differentiated into classical, intermediate, and nonclassical monocytes initially by FSC/SSC characteristics and secondarily by CD14 and CD16 staining characteristics (16), with CD66b exclusion (15). CD11b expression on each subset was then determined. Monocyte subsets are presented as a percent of the total monocyte population, while CD11b expression is depicted as a fold change from PRE to account for potential nonspecific binding.
Changes in markers of muscle damage, circulating and intramuscular cytokines, as well as monocyte subset distributions and CD11b expression, were analyzed by a two-way, between subjects, repeated-measures ANOVA. In the event of a significant F ratio, a one-way, within-subjects, repeated-measures ANOVA for each group and a one-way, between-subjects ANOVA at each time point with LSD pairwise comparisons were used for post hoc analysis. Significant time and group effects were subsequently analyzed with LSD pairwise comparisons. Furthermore, given the impact of muscle biopsies on the inflammatory process (6), intramuscular MCP-1 was analyzed as one-way repeated measures ANOVA for each group, similar to Della Gatta et al. (8). Data that were not normally distributed (according to the Shapiro–Wilk test) underwent a natural log (LN) transformation. The area under the curve (AUC) was also calculated for changes in circulating cytokines and myoglobin response from PRE to 5H using a standard trapezoidal technique, and a one-way ANOVA was used to examine differences among groups. Raw concentrations from PRE, IP, 1H, and 5H were used to calculate AUC before LN transformation. Additionally, interpretations of effect sizes (η p 2) were based on those used by Wells and colleagues (23) for a similar study design: small effect (η p 2 < 0.06), medium effect (η p 2 < 0.14), and large effect (η p 2 > 0.14). Pearson product moment correlations between markers of muscle damage and monocyte chemoattractants (MCP-1 and CX3CL1), monocyte subpopulations and CD11b expression were assessed for all groups combined. As changes in circulating CK usually do not manifest until 24 to 48 h after resistance exercise (6), markers of muscle damage included CK at 24H and 48H, as well as myoglobin AUC. Significance was accepted at an alpha level of P ≤ 0.05 and all data are reported as mean ± SD of the original, nontransformed data.
No differences were observed between groups for age, height, body mass, BMI, maximal strength (squat and leg press 1-RM), or dietary intake of calories, carbohydrates, protein or fats (6). Furthermore, PPB and PL were significantly greater than CON at IP, 1H, and 5H for circulating myoglobin, while circulating CK was significantly increased in PPB at 24H and 48H, and at 24H in PL compared to CON (6).
Changes in circulating MCP-1 are depicted in Figure 3. A significant interaction was observed for MCP-1 (Fig. 3A; F = 3.23; P = 0.005; ηp 2 = 0.16). Post hoc analysis indicated significantly greater MCP-1 concentrations at 5H in PPB (P = 0.001) and PL (P = 0.012) than CON. Significant increases in PPB from PRE at IP (P = 0.005), 1H (P = 0.010), and 5H (P < 0.001), as well as in PL from PRE to IP (P = 0.006) and 5H (P < 0.001), were also observed, while significant reductions in MCP-1 were noted from PRE to 48H in both PPB (P = 0.048) and PL (P = 0.010). The AUC analysis indicated a significant interaction (Fig. 3B; F = 3.53; P = 0.040), with a significantly greater MCP-1 response in PPB than CON (P = 0.014), and a trend toward a significant increase between PPB and PL (P = 0.091).
No significant interaction was observed for CX3CL1 (F = 1.13; P = 0.345; ηp 2 = 0.05), though a significant effect of time was observed (F = 16.95; P < 0.001; ηp 2 = 0.34) for all groups combined. Pairwise comparisons indicated CX3CL1 was significantly elevated at IP (184.7 ± 70.9 pg·mL−1) and 1H (171.9 ± 73.8 pg·mL−1) compared with PRE (149.1 ± 84.9 pg·mL−1; P < 0.001, P = 0.004, respectively), 24H (138.9 ± 74.3 pg·mL−1; P < 0.001), and 48H (130.5 ± 66.4 pg·mL−1; P < 0.001), while IP was significantly greater than 5H (165.2 ± 107.7 pg·mL−1; P = 0.005). The AUC analysis indicated no significant interaction (F = 0.87; P = 0.429).
Intramuscular cytokine protein content
Changes in intramuscular MCP-1 protein content are depicted in Figure 3C. No significant interaction was observed for intramuscular MCP-1 (F = 0.92; P = 0.473; η p 2 = 0.06); however, due to the nature of this data, each condition was examined individually as previously reported (8). Intramuscular MCP-1 content increased over time for each group individually (CON, F = 12.29; P < 0.001; ηp 2 = 0.67; PPB, F = 0.96; P < 0.001; ηp 2 = 0.82; PL, F = 0.96; P < 0.001; ηp 2 = 0.85). Intramuscular MCP-1 content increased in all groups compared with PRE (P < 0.05); however, only PPB increased from 1H to 5H (P = 0.028). Furthermore, intramuscular MCP-1 content was higher at 5H than 48H for both PPB (P = 0.003) and PL (P = 0.016).
Monocyte subset distributions
Changes in monocyte subset distributions are depicted in Figure 4. A significant interaction was observed for the proportion of classical monocytes (F = 27.99; P < 0.001; η p 2 = 0.47). Post hoc analysis indicated significantly reduced proportions of classical monocytes at IP in PPB (P = 0.008) and PL (P = 0.003) compared with CON, while significant increases in the proportion of classical monocytes at 1H were observed in PPB (P = 0.002) and PL (P = 0.006) compared with CON.
A significant interaction was observed for the proportion of intermediate monocytes (F = 7.76; P < 0.001; η p 2 = 0.33). Post hoc analysis indicated significantly increased proportions of intermediate monocytes at IP in PPB (P = 0.034) and PL (P = 0.001) when compared with CON, while significantly reduced proportions were observed at 1H in PPB (P = 0.003) and PL (P = 0.008) when compared to CON. Furthermore, there was a greater proportion of intermediate monocytes in PPB than CON at 24H (P = 0.016) and 48H (P = 0.007).
A significant interaction was also observed for the proportion of nonclassical monocytes (F = 7.43; P < 0.001; η p 2 = 0.32). Post hoc analysis indicated there was a significantly greater proportion of nonclassical monocytes in PPB (P = 0.014) and PL (P = 0.015) than CON at IP.
Monocyte CD11b expression
Significant differences were observed between groups at PRE for CD11b expression on intermediate monocytes (F = 3.59; P = 0.039) and nonclassical monocytes (F = 7.94; P = 0.002). As such, analyses between groups for expression of CD11b on monocyte subsets were analyzed as the percent of resting values (PRE representing 100%). Changes in CD11b expression on monocyte subsets are depicted in Figure 5.
No significant interaction was observed for the change in CD11b expression on classical monocytes (F = 1.79; P = 0.111; η p 2 = 0.10), however, a significant effect of time was identified (F = 9.66; P < 0.001; η p 2 = 0.23) with all groups combined. Pairwise comparisons indicated a significant increase from PRE to 1H (P < 0.001), 5H (P = 0.033), and 24H (P = 0.004). Furthermore, changes in CD11b expression from PRE to 1H were significantly greater than changes from PRE to IP (P = 0.001), 5H (P = 0.004), 24H (P < 0.001), and 48H (P < 0.001), while the change from PRE to 24H was significantly greater than PRE to 48H (P = 0.002).
No significant interaction was observed for the change in expression of CD11b on intermediate monocytes (F = 1.86, P = 0.083, η p 2 = 0.10); however, a significant main effect of time was observed (F = 5.67, P = 0.001, η p 2 = 0.15) for all groups combined. Pairwise comparisons indicated a significant increase in CD11b expression from PRE to 1H (P < 0.001).
No significant interaction (F = 1.92, P = 0.084, η p 2 = 0.11), nor main effect of time (F = 2.10, P = 0.104, η p 2 = 0.06) was observed for the change in expression of CD11b on nonclassical monocytes; however, a significant main effect for group (F = 4.41, P = 0.020, η p 2 = 0.22) was observed. Pairwise comparisons indicated that the average change in expression of CD11b from PRE to all time points was lower in CON compared with PPB (P = 0.006). No difference was observed between PL and CON (P = 0.121) or PPB (P = 0.106).
Significant correlations were observed between circulating markers of muscle damage and monocyte chemoattractants. Circulating MCP-1 concentrations at 5H and 24H were significantly correlated with CK concentrations at 24H (r = 0.406, P = 0.014; r = 0.505, P = 0.002, respectively), and 48H (r = 0.396, P = 0.017; r = 0.491, P = 0.002, respectively). The MCP-1 concentration at 24H was also significantly correlated with myoglobin AUC (r = 0.435; P = 0.007).
The CX3CL1 concentrations at 5H and 24H were correlated with circulating CK concentration at 24H (r = 0.433, P = 0.011; r = 0.408, P = 0.017, respectively) and 48H (r = 0.394, P = 0.021; r = 0.364, P = 0.034, respectively). The CX3CL1 concentrations were also correlated with myoglobin AUC at 1H (r = 0.340, P = 0.046), 5H (r = 0.503, P = 0.002), and 24H (r = 0.434, P = 0.009).
Significant correlations were observed between markers of muscle damage and changes in CD11b expression on monocyte subsets. CK concentrations at 24H were significantly correlated with the change in CD11b expression on intermediate monocytes at IP (r = 0.439, P = 0.001), and nonclassical monocytes at IP (r = 0.413, P = 0.017), 1H (r = 0.392, P = 0.024) and 5H (r = 0.374, P = 0.032). Furthermore, CK concentrations at 48H were significantly correlated with the change in CD11b expression at IP on classical monocytes (r = 0.365, P = 0.037) and intermediate monocytes (r = 0.452, P = 0.008). Additionally, the change in CD11b expression on nonclassical monocytes was correlated at IP (r = 0.450, P = 0.009), 1H (r = 0.402, P = 0.020), 5H (r = 0.345, P = 0.049), 24H (r = 0.446, P = 0.009), and 48H (r = 0.495, P = 0.003). No significant correlations were observed between the change of CD11b expression on leukocytes and myoglobin AUC. Correlations with leukocyte subset proportions are displayed in Table 1.
The primary findings of this investigation were that resistance exercise elicited significant recruitment, and mobilization of monocytes. As expected, resistance exercise stimulated an increase of circulating MCP-1 concentrations immediately after exercise and a further increase at 5H. While the time course of this response does not appear to be influenced by polyphenol supplementation, PPB did have a significantly greater AUC response than CON, while PL did not. Additionally, resistance exercise induced significant mobilization of intermediate and nonclassical monocyte subtypes immediately after resistance exercise, followed by a supercompensation of the classical subset, while the PPB group experienced an elevated proportion of intermediate monocytes 48 h into recovery. Furthermore, classical and intermediate monocytes increased expression of CD11b at 1H, while only classical monocytes maintained this elevated expression for 24 h. It is important to note, however, that this increase may have been due to the effect of the muscle biopsy, as this increase was noticed for all groups combined. Interestingly, however, although polyphenol supplementation may have enhanced the CD11b response to exercise compared to CON on nonclassical monocytes, though, PPB supplementation may have suppressed CD11b expression on intermediate and nonclassical monocytes at rest.
Skeletal muscle damage produces a robust immune response, characterized by an increased accumulation of phagocytic cells within the damaged tissue (10). MCP-1 plays an integral role in the acute immune response by serving as the primary chemoattractant for classical monocytes (16,21). Our results indicated that the response of circulating MCP-1 to resistance exercise may be biphasic; characterized by an initial increase immediately after exercise, and a second, larger increase at 5H. To the best of our knowledge, only two investigations have examined the acute response of circulating MCP-1 after dynamic resistance exercise (23,24). The immediate increase of circulating MCP-1 observed in this study is in contrast to the decrease in MCP-1 concentrations reported 30 min after resistance exercise in untrained men (24). It is similar, however, to the immediate increase reported after a lower-body resistance exercise protocol in resistance trained men (23). While Ihalainen et al. (24) used a similar population as this study (e.g., previously untrained men), the exercise stimulus only used multiple sets of a single leg press exercise. In contrast, the current study required participants to exercise with multiple lower body exercises, similar to Wells and colleagues (23). The greater number of repetitions (140 total repetitions compared to 50 in the Ihalainen et al. study), and exercises (squat, leg press and leg extension compared to leg press only) performed in this study may have resulted in a greater muscle damage (6,24). Given the observed correlations between markers of muscle damage and MCP-1 in the current study, the observed increase in MCP-1 concentration may be influenced by muscle damage or training volume, and may be independent of training status.
The biphasic MCP-1 response observed was not found in previous studies examining the MCP-1 response to resistance exercise (23,24). Investigations using protocols designed to elicit muscle damage, however, have seen a secondary response approximately 5 h into recovery (1). The participants in those studies had a similar training background as those recruited in this present study. Therefore, in addition to training volume, the biphasic response of MCP-1 may also be a function of novel muscle action.
Within the muscle, MCP-1 was found to increase from PRE in all conditions, and at all time points. Exercise appeared to increase intramuscular MCP-1, and PPB supplementation may delay this response. Previous investigations have demonstrated increased intramuscular MCP-1 in response to both aerobic (32), and resistance exercise (8). Our results are similar to the findings of Della Gatta and colleagues (8), who demonstrated increases at 2 and 4 h after resistance exercise. Increases in intramuscular MCP-1 were also observed in the CON group, however, indicating a potential influence of the muscle biopsy on the MCP-1 response (6). Therefore, it is difficult to isolate the effect of resistance exercise from that of the biopsy. Regardless of the source, the increases observed in circulating and intramuscular MCP-1 likely recruit classical monocytes to the site of tissue damage (21).
While the recruitment of classical monocytes is governed by MCP-1, the recruitment of other monocyte subsets is governed primarily by CX3CL1 (22). Circulating CX3CL1 concentrations increased at both IP and 1H for all groups combined. Others have reported significant increases in CX3CL1 mRNA 2 h postresistance exercise (8), while Della Gatta and colleagues (8) demonstrated a return to baseline levels at 4 h postexercise. To our knowledge, only one investigation has examined circulating CX3CL1 concentrations after exercise, and has reported increases for 2 h after unilateral cycling (33). As this investigation (33) did not have a control group, it is difficult to distinguish between the effects of the exercise and that of the biopsy. Nonetheless, as CX3CL1 is synthesized by endothelial cells (34), it will remain bound to the endothelial surface unless cleaved by TNF-α and IL-1β (35). As TNF-α and IL-1β are proinflammatory cytokines (7), the observed increase in CX3CL1 may be more indicative of a proinflammatory environment, which will be observed after resistance exercise, or potentially after skeletal muscle biopsies (6,36).
As with markers of monocyte recruitment, expansion of the monocyte population has been well documented (9); however, the mobilization of specific subtypes is less defined. Previously, Wells and colleagues (19) have demonstrated no change in classical monocytes immediately after resistance exercise, with significant increases from 1 to 5 h into recovery in resistance-trained men (19). While to our knowledge, this is the first investigation to document changes in circulating intermediate and nonclassical monocyte proportions after resistance exercise, previous studies, using an aerobic exercise model, have demonstrated significant mobilization of the nonclassical subset immediately after exercise (17,18). This selective mobilization of CD16+ monocytes may be mediated by catecholamines (37), though exercise intensity appears to drive the response (38). Increases in both the intermediate and nonclassical monocyte populations have also been demonstrated after aerobic exercise, though this was demonstrated while using a mixed-gender population (17). Considering nonclassical monocytes appear to respond differently to exercise in women compared to men (39), the increased intermediate monocyte proportion observed by Booth et al. (17) may be driven by the sex of the participants. As this investigation used only men, the increased intermediate monocytes observed may be specific to resistance exercise.
Further increases in intermediate monocytes at 24H and 48H in this investigation appeared to occur at the expense of the classical monocyte subset, with no change in the nonclassical proportion. Furthermore, PPB demonstrated a significantly greater intermediate monocyte population than CON at both these time points. While data after ischemic tissue damage has shown elevations in intermediate monocytes 24 and 48 h postinjury (20), to our knowledge, no investigations have examined the monocyte subset response to exercise 24 and 48 h into recovery. Furthermore, Tapp and colleagues (20) have reported that changes in intermediate monocytes were associated with the extent of tissue damage. This is consistent with the correlations observed in this study (r = 0.380 to 0.452) between markers of muscle damage and the intermediate monocyte proportions at 24H and 48H. Given that this resistance exercise bout resulted in a greater increase in CK in PPB versus PL (6), it is possible the observed increase in intermediate monocytes at 24H and 48H is a function of muscle damage as opposed to the effects of polyphenol supplementation. Therefore, it is possible that intermediate monocytes increase in proportion to the magnitude of skeletal muscle damage that occurs postexercise, though future studies must confirm this.
The involvement of CD11b in the transendothelial migration process makes CD11b a key regulator of the phagocyte migration to damaged tissue (26). Previous reports have demonstrated elevated CD11b expression on classical monocytes immediately after and 1 h postresistance exercise (23), while another demonstrated significant increases 30 min after a resistance exercise bout (25). The disparity in the time course reported between these studies may be due to the methods used to identify monocytes, and the lack of differentiation between classical and intermediate monocytes (25). Consequently, Wells et al. (23) suggested that the response of CD14++ monocytes may have been influenced by intermediate monocytes. The delayed increase of intermediate monocytes in this investigation supports the assertion by Wells and colleagues (23); however, the current study also observed a delayed increase (at 1H) of CD11b on classical monocytes. This difference may be due to the lower level of training of the participants in our study, however, the methods of analysis may have also played a role, as we examined changes from PRE, as opposed to raw mean fluorescence intensity expressions.
Polyphenol supplementation (PPB) for 28 d also appeared to reduce CD11b expression on intermediate and nonclassical monocytes at rest compared to CON and PL. CD11b expression on monocytes was not examined before the onset of supplementation, therefore we can only speculate in relating this decreased expression of CD11b to polyphenol supplementation. However, there does appear to be support for this from an in vitro model, which demonstrated a significant downregulation of CD11b, as well as reduced chemotaxis and adherence in response to incubation with polyphenols (28). However, others have suggested that decreased CD11b expression on monocytes may have beneficial health benefits, as polyphenol-associated decreases in CD11b expression on monocytes has been previously reported to have potential beneficial effects in cardiovascular disease (27).
In this study, no differences in the time course of CD11b expression were observed between groups for classical or intermediate monocytes. However, participants in PPB demonstrated significantly greater increases in CD11b expression on nonclassical monocytes during 48 h of recovery than CON. While the expression of CD11b on nonclassical monocytes is reduced compared to other subsets (40), it is unclear why this difference occurred. Nonetheless, the role of CD11b in transendothelial migration (26) and the propensity of nonclassical monocytes to polarize to M2 macrophages (41) may indicate an increased tendency to transmigrate, that would likely have no deleterious effects on recovery. Therefore, polyphenol supplementation may reduce monocyte adherence and chemotaxis at rest, however, may only have a limited effect on the exercise response, though a greater understanding of how other components to the cell adhesion cascade respond to polyphenol supplementation is needed.
Together, the results of this investigation demonstrate that resistance exercise initiates monocyte recruitment, and mobilization. Furthermore, polyphenol supplementation appears to modestly influence the migration of monocytes, reducing CD11b expression at rest, and possibly enhancing expression on nonclassical monocytes after exercise. Therefore, this may modify the transmigration of monocytes into the tissue, although examinations that confirm this within muscle are needed. Unfortunately, the muscle biopsies used in this investigation appear to have influenced some of the results of this study, evidenced by the increased intramuscular MCP-1 and circulating monocyte CD11b expression in CON. While this has been demonstrated to occur previously (6), the biopsies were vital to the acquisition of this data. Although we used microbiopsies, as opposed to percutaneous biopsies with suction (8) to reduce the total quantity of tissue collected, this also increases the ratio of tissue exposed to the trauma of the biopsy procedure itself, in relation to the total volume of tissue collected, potentially inflating the MCP-1 concentration of our sample. Furthermore, as this investigation did not use absolute cell counts, we are unable to indicate absolute changes in monocyte subset populations. Additionally, as CD16 is a marker for NK cells (42), the CD16+ nonclassical monocyte gate may have contained a small number of NK cells. Future research should examine the changes in monocyte subset population counts after resistance exercise. Furthermore, these investigations should examine the expression of various chemotactic receptors, namely CX3CR1 and CCR2, after resistance exercise. Finally, given that polyphenol supplementation appears to influence the expression of cell adhesion molecules, future studies should examine earlier components of the cell adhesion cascade to provide a greater understanding of the propensity for monocytes to undergo transendothelial migration.
This study was funded by Kemin Foods, L.C. (Des Moines, IA). The authors would like to thank Michael J. Redd, Michael B. La Monica, Carleigh H. Boone, Kayla M. Baker, Joshua J. Riffe, Tyler W.D. Muddle, and Ran Wang for their assistance with data collection for this article. The authors would also like to thank Frank Zaldivar, PhD and Fadia Haddad, PhD for their assistance with the flow cytometry protocols and data acquisition. This investigation is registered on clinicaltrials.gov with reference number NCT02442245.
K. A. H. is an employee of Kemin Foods, L.C. All other authors have no actual or potential conflicts of interest to report. As such, the results of this study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. The results of this study do not constitute endorsement by the American College of Sports Medicine.
A. R. J., J. R. H., J. R. T., K. S. B., D. D. C., K. A. H., and J. R. S. participated in the conception and design of research. A. R. J., J. R. T., K. S. B., A. N. V., D. D. C., L. P. O. participated in the acquisition of data. A. R. J., J. R. H., D. D. C., D. H. F. participated in the data analysis and interpretation. A. R. J., J. R. H., K. A. H., S. R. A., D. H. F., J. R. S. participated in the article draft and revision. A. R. J., J. R. H., J. R. T., K. S. B., A. N. V., D. D. C., L. P. O., K. A. H., S. R. A., D. H. F., J. R. S. participated in the approval of final version.