Analysis of human neutrophil phenotypes as biomarker to monitor exercise‐induced immune changes

Abstract The amplitude of the innate immune response reflects the degree of physiological stress imposed by exercise load. An optimal balance of exercise intensity and duration is essential for a balanced immune system and reduces the risk of dysfunction of the immune system. Therefore, it is hypothesized that neutrophils, as key players in the innate immune system, can be used as biomarker in detecting overtraining. The aim was to monitor the state of the innate immune system by phenotyping neutrophils during consecutive bouts of prolonged exercise. Study subjects were recruited from a cohort of walkers participating in a walking event on 3 consecutive days. Participants with immune deficiencies were excluded. Questionnaires to determine the physiological status of the participants were completed. Analysis of neutrophil receptor expression was done by a point‐of‐care fully automated flow cytometer. A total of 45 participants were recruited, of whom 39 participants were included for data analysis. Study participants had a median age of 64 (58‐70) years. The absolute numbers CD16dim/CD62Lbright and CD16bright/CD62Ldim neutrophils were increased after the first 2 days of exercise followed by an adaptation/normalization after the third day. Participants with activated neutrophils (high CD11b expression) had an impaired physical feeling indicated by the participant on a lower visual analog scale compared to participants who did not have activated neutrophils (P = 0.017, P = 0.022). Consecutive days of prolonged exercise results in an initial systemic innate immune response, followed by normalization/adaptation. Increased neutrophil activation was associated with impaired physical feeling measured by a validated VAS score indicated by the participant. Fully automated point‐of‐care flow cytometry analysis of neutrophil phenotypes in a field laboratory might be a useful tool to monitor relevant differences in the systemic innate immune response in response to exercise.

and regular sports are related to a reduction of infections and other diseases. [10][11][12] It has been hypothesized that the relationship between absolute exercise load and risk of illness is a J-shaped curve, with very low or no training being associated with a higher risk of illness, moderate training associated with low illness risk and very high training associated with the highest risk of acute illness. 13 This suggests that an optimal balance in exercise intensity and duration is essential for a balanced immune system. 14 Therefore, a biomarker that can be used to monitor the immune system during and after repetitive bouts of prolonged exercise may help to determine the optimal exercise load and thus training program. 15 Many studies showed an altered adaptive immunity (T-cell changes, NK cell activity, salivary IgA production), cytokine production and innate immunity (granulocyte cell count, granulocyte respiratory burst, neutrophil/lymphocyte ratio, and M activity) for several hours to days during recovery from prolonged exercise. [16][17][18][19][20] The innate immune system demonstrates the most pronounced changes after prolonged exercise. 17 The primary effector cells of the innate immune system are granulocytes and monocytes. [21][22][23] Receptor expression on these cells, measured by flow cytometry, has proven to give insight into the status of the innate immune system. [24][25][26] A recent study showed that intensive exercise led to the mobilization of neutrophil phenotypes associated with systemic inflammation and immunosuppression. 27 After repetitive prolonged exercise on consecutive days the neutrophil phenotypes, CD16 dim /CD62L bright and CD16 bright /CD62L dim increase in the peripheral blood. 27,28 The CD16 dim /CD62L bright cells are young neutrophils with a band shape nucleus, most likely recruited from the bone marrow. 29 The CD16 bright /CD62L dim cells consist of mainly neutrophils containing hyper-segmented nuclei that show a more activated and immunosuppressive phenotype. 28 The presence of this immunosuppressive phenotype during exercise might be involved in the increased risk of developing infections. 27,28 Neutrophil receptor expression can, therefore, be a possible biomarker in monitoring the balance of the immune system during and after repetitive, prolonged exercise. 24 Analysis of neutrophil receptor expression used to have many limitations that precluded broad application in a non-laboratory setting.
These limitations include a short time window-of-opportunity for analysis, because inflammatory cells are easily activated by ex vivo manipulation and flow cytometry requires a fully functional immunological laboratory with experienced laboratory personnel. 25,30 The technique of fully automated flow cytometry has become available that circumvents most of these limitations and is applicable in a field laboratory setting. 25,30 This now allows analysis of the immune system in athletes at the sports site.
To get more insight into the potential role of neutrophil phenotypes in the exercise-induced immune response, we investigated neutrophils as read out for the innate immune system point-of-care during repetitive, prolonged exercise in a large-scale public walking event where athletic accomplishment of distance goals is closely monitored. The primary aim of our study was to find a biomarker that is associated with overreaching. Our secondary study goal was to study neutrophil kinetics during repetitive, prolonged walking.

Baseline measurements
Baseline data were collected 1 or 2 days before the start of the event, after a minimum resting period of 24 h (Fig. 1). A period of 24 h recovery was based on the kinetics of immune cell count recovery postexercise. 4 At baseline, body height and weight (Seca 888 scale, Hamburg, Germany) were measured to calculate body mass index (BMI).
Waist circumference was measured with a measuring tape (Seca 201, Chino, CA, USA). Resting heart rate (HR), systolic blood pressure (SBR), and diastolic blood pressure (DBP) were measured in the supine position after a 5 min resting period. All participants completed a general questionnaire on demographics, level of education, smoking, and medication use. Heart rate (HR) was used to estimate exercise intensity as a percentage of the maximum HR (exercise intensity = measured HR/expected maximal HR × 100%, where expected max HR = 208 -132 (0.7 × age)). 31 Exercise intensity was determined using the guideline of the America Heart Association. 32 Performing 50-70% of your maximum heart rate is considered moderate intensity. From 70% to 85% of your maximum heart rate is regarded as vigorous intensity. For determination of the exercise intensity, heart rate at day 1 was measured during the prolonged exercise at every 5-km checkpoint. The mean heart rate per person was used for exercise intensity calculation.

Visual Analog Scale questionnaire
At baseline, and after every day of walking exercise, participants were instructed to score their physical feeling, exercise effort, and perceived F I G U R E 1 Design of the study. Baseline data were collected 1 or 2 days before the start of the event and each day of walking within 30 min after completion of the exercise muscle pain using the Visual Analog Scale (VAS), which consists of a horizontal line of 100 mm. 33 This method has been extensively validated and used in many studies. [34][35][36] The aspect of physical feeling was determined with the question: "Show with a vertical line how you feel today," bad feeling (0 mm) to best feeling (100 mm). The exercise effort was determined with the question: "Show with a vertical line how much effort the exercise costed you today," it felt like a short walk (0 mm) to it felt like an extremely long walk (100 mm). The aspect perceived muscle pain was determined with the question, "Show with a vertical line how many muscle complaints you had today," no complaints (0 mm) to worst pain ever (100 mm). Furthermore, participants were asked whether they took painkillers after every day of walking.

Blood sampling
Venous blood was drawn at baseline and after each day of walking <30 min after completion of the exercise (Fig. 1

Flow cytometry analysis
The AQUIOS CL R combines robotic automated sample preparation with analysis of cells using flow cytometry. 25 The AQUIOS CL R has one 488 nm diode laser, 2 light scatter channels (forward scatter and side scatter), 5 fluorescence channels, and an electronic volume measure. Absolute leukocyte count was based on an electronicvolume measurement. A cassette filled with blood tubes was placed in the machine and the barcodes of the samples were saved. After automatic blood mixing, the samples were cap-pierced, and 43 l was pipetted into a 96-deep wells plate, that is used for Ab staining. The .lmd data files were exported from the AQUIOS CL R and imported into FlowJo R analysis software (Tree Star Inc., Ashland, OH).
The absolute total leukocyte count was extracted from the .pdf files generated by the AQUIOS CL R . The gating strategy is shown in Supplemental Figure 2. Hematology test reference ranges of the American board of internal medicine were used regarding absolute cell counts. 37 Reference values of flow cytometry were determined with in the field laboratory with the use of healthy control non-walkers. Granulocytes and monocytes were gated based on forward scatter and side scatter. Neutrophils and eosinophils in the granulocyte gate were identified based on positivity or negativity of CD16 expression. Neutrophil phenotypes were identified by the expression of CD16 and CD62L, as described in detail before. 28 The gating strategy was checked for every individual by 2 independent researchers. The median fluorescent intensity (MFI) of gated granulocytes was used to describe the expression of CD35, CD11b, and CD10.

Statistical analysis
The statistical analyses were conducted in IBM SPSS R Statistics for

RESULTS
Demographics of the total study group are shown in Table 1

Prolonged walking induced a systemic innate immune response after day 1
The absolute number of white blood cells, neutrophils, monocytes, and eosinophils are shown in Figure 2. The total white blood cell count increased after 1 day of walking (6.1 ± 1.8 × 10 6 /ml at baseline to 9.8 ± 2.4 × 10 6 /ml after 1 day of walking; P < 0.0001). This effect was mainly caused by an increase of neutrophilic granulocytes (4.0 Eosinophilic granulocytes show the opposite pattern of neutrophilic granulocytes, with a small but significant decrease after 1 day (0.11 (0.07-0.16) × 10 6 /ml to 0.10 (0.05-0.14) × 10 6 /mL; P < 0.05).

Normalization of systemic innate immune response after 2 and 3 days of walking
Repetitive prolonged walking exercise on consecutive days resulted in a partial normalization of tWBC from 9.8 ± 2.4 × 10 6 /ml on day 1 to 8.8 ± 2.0 × 10 6 /ml on day 2 (P < 0.005) and to 8.0 ± 1.9 × 10 6 /ml on day 3 (P < 0.0005). The decrease in tWBC count on day 2 was mainly the result of a partial normalization in neutrophil numbers (7.9   were still present at day 2 ( Fig. 3A and B). The decrease of total

F I G U R E 2 The absolute count of total white blood cells (A), monocytes (B), and neutrophils (C) show a significant increase on day 1, followed by a decrease in the following days (n = 39). Eosinophils (D)
show the opposite effect, a slight decrease on day 1 and after that a significant increase. The total white blood cell count was obtained from the cell counter in the automated flow cytometer. The percentage monocytes and granulocytes were gated base of forward versus side scatter. The percentage of neutrophils and eosinophils were determined by CD16 + and CD16in the CD16 granulocyte histogram. Total white blood cells are shown as scatter plot with mean and SD, tested with ANOVA repeated-measures with a Bonferroni post hoc multiple comparison correction. The rest of the data are presented as a scatter plot with median and interquartile range, tested with Friedman's test, with Dunn's post hoc multiple comparison correction. Reference values (grey area) show laboratory test reference ranges of the American board of internal medicine; ns, not significant; * P < 0.05; ** P < 0.005; *** P < 0.0005; **** P < 0.0001 neutrophil count after day 2 of walking, was mainly the result of a decrease in CD16 bright /CD62L bright neutrophils (4.58 (3.33-5.70) × 10 6 /ml to 3.12 (2.41-4.57) × 10 6 /ml; P < 0.0001; Figure 3C).
From day 2 to day 3 CD16 dim /CD62L bright neutrophils decreased

Demonstration of an exponential relationship between CD62L and CD11b expression on neutrophils after repeated, prolonged walking
The MFIs of the neutrophil activation markers CD35, CD11b, and CD10 are shown in Figure 3D

Increased expression of CD11b on neutrophils was associated with slow walking speed and bad physical feeling
Participants with a high neutrophil CD11b expression after day 1 (n = 21) were compared to participants with normal CD11b expression after day 1, n = 18; Table 2). A total of 66% of the participants with increased CD11b expression walked 30 km compared to 34% who    Continuous data are shown as median (IQR), the significance is tested with the Mann-Whitney U-test. Dichotomous data are shown as an absolute amount (percentage), the significance is tested with the Pearson chi-square test; MFI, median fluorescent intensity; *significant.

DISCUSSION
To our knowledge, this is the first study in the literature that applied fully automated flow cytometry in a field laboratory to determine the effect of repetitive, prolonged walking exercise on consecutive days on the innate immune system. The study demonstrated the mobilization and activation of different neutrophil phenotypes in peripheral blood in response to prolonged exercise. The most pronounced immune response was found after day 1, followed by a partial normalization/adaptation in the days after that. Participants with a high neutrophil CD11b expression, indicative for neutrophil activation, were characterized by a higher degree of physiological stress imposed by the exercise load. Therefore, CD11b expression on neutrophils might be a valuable marker to monitor changes in the immune system after exercise. 38 We used the neutrophil activation markers CD35, CD11b, and CD10, since they have been validated to demonstrate neutrophil activation in a fully automated flowcytometer. 25 The complement receptors type 1 (CR1/CD35) and 3 (CR3/Mac1/CD11b) are known markers for neutrophil activation. 39,40 A reservoir of CR1 and CR3 receptors in neutrophils are present in cytoplasmic secretory vesicles, which are translocated to the plasma membrane upon cell activation. 39,40 The receptor CD10 is a known maturation marker and activation marker. 25,41,42 The only other field study focusing on neutrophil phenotyping during exercise was done by Van Staveren et al. 27 They investigated the expression of neutrophil markers of participants of an 8 days cycling tour with a mean daily distance of 160 km and 2300 altimeters. 27 Blood sampling was performed in the morning before and after 4 and 8 days of cycling. This study showed an increased overall neutrophil count and an increased CD16 dim /CD62L bright and CD16 bright /CD62L dim after day 4 and 8 of cycling. Even though the same neutrophil phenotypes were studied it is challenging to compare the data to our study for several reasons: (i) our study was performed with a fully automated flow cytometer in a field laboratory whereas Van Staveren et al. 27 collected, fixated, and froze the samples for later analysis in the laboratory; (ii) the first exercise data were obtained after 4 days, whereas this study determined the effects on neutrophils after 1, 2, and 3 days; and (iii) the exercise volume was not comparable.
Nevertheless, we found a similar increase in CD16 dim /CD62L bright neutrophils and CD16 bright /CD62L dim neutrophils after 1 day of walking.
Hereafter, a decrease in the number of these cells indicated adaptation of the innate immune system to a certain level of exercise. This is in contrast to van Staveren et al., 27 who showed a cumulative increase of systemic immune response after 4 and 8 days, which could be explained by the higher exercise intensity and load in their study, fitness level, and/or the sampling methods. However, one might speculate that low intensity prolonged walking exercise allows the immune system to adapt, while high-intensity cycling at high altitude leads to an overreacting immune system preventing adaptation. In a previous study on repeated prolonged exercise, conducted during the same walking event, an increase in cytokines levels was shown after the first day and a normalization/adaptation the consecutive days. 43 This leads to a similar conclusion regarding innate immune response, as we describe in this study.
After the first bout of prolonged exercise, several changes were visible in the neutrophil compartment in the peripheral blood. The CD16 dim /CD62L bright neutrophils were increased in the peripheral blood. These cells are mainly young neutrophils with a banded shaped nucleus released from the bone marrow. 44 It is tempting to speculate that these neutrophils were mobilized from the bone marrow in response to damage-associated molecular patterns (DAMPs) originating from the muscle damage in exercise. Furthermore, there was an increase of CD16 bright /CD62L dim neutrophils in all walking participants after day 1. This immune-suppressive subset of neutrophils might be involved in immune regulation during physical exercise. 28,29 After days 2 and 3, a decrease in these neutrophil phenotypes was demonstrated indicative for adaptation of the innate immune system to exercise-induced DAMPs.
All participants with a high neutrophil CD11b expression had an increase in number of CD16 bright /CD62L dim neutrophils as well.
Therefore, the mere presence of high expression of the neutrophil activation marker CD11b also seems to correlate with the presence of immunosuppressive, CD16 bright /CD62L dim neutrophils in prolonged exercise. 28 An increase of the immunosuppressive CD16 bright /CD62L dim neutrophils might be involved in the development of infectious illnesses during overtraining periods.
A limitation of this study is that a relatively elderly population was studied, which might not be representative of the general population.
This age was in line with the mean age of participants who generally participate in these walking events.
With fully automated flow cytometry in a field laboratory, it is now possible to implement neutrophil phenotyping as a fast (20 min) and easily accessible test for studying the immune system in field settings. Future studies should also focus on high-intensity exercise and should add more blood drawing moments to get a better understanding of the response of the innate immune system. For future studies in exercise immunology, it might be considered to integrate a fully automated flow cytometer in a moving mobile laboratory.

CONCLUSION
Repetitive, prolonged walking exercise on consecutive days results in an initial systemic innate immune response, followed by normal-