Impact of exercise on markers of B cell-related immunity: A systematic review

Highlights

• •Markers of B cell-related immunity are altered by acute exercise and exercise training.
• •B cell counts and levels of secretory and plasma immunoglobulin A are among the most exercise-responsive markers.
• •The large heterogeneity concerning study populations and exercise regimens and the poor study quality and reporting of results hamper evidence-based conclusions on the impact of exercise on B cell-related immunity.
• •Alternative markers (e.g., immunoglobulin free light chains), subpopulations of B cells and functional outcomes of B cell-related immunity should be included more frequently in future exercise studies.

Abstract

Background

B cells represent a crucial component of adaptive immunity that ensures long-term protection from infection by generating pathogen-specific immunoglobulins. Exercise alters B cell counts and immunoglobulin levels, but evidence-based conclusions on potential benefits for adaptive immunity are lacking. This systematic review assessed current literatures on the impact of acute exercise and exercise training on B cells, immunoglobulins, and markers of secretory immunity in human biofluids.

Methods

According to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, MEDLINE, Web of Science, and Embase were searched on March 8, 2023. Non-randomized controlled trials and crossover trials investigating the impact of acute exercise or exercise training on B cell counts and proportions, immunoglobulin levels, salivary flow rate, or secretory immunoglobulin A secretion rate were included. Quality and reporting of exercise training studies were assessed using the Tool for the Assessment of Study Quality and reporting in Exercise. Study characteristics, outcome measures, and statistically significant changes were summarized tabularly.

Results

Of the 67 eligible studies, 22 applied acute exercise and 45 applied exercise training. All included outcomes revealed significant alterations over time in acute exercise and exercise training context, but only a few investigations showed significant differences compared to control conditions. Secretory and plasma immunoglobulin A levels were most consistently increased in response to exercise training.

Conclusion

B cell-related outcomes are altered by acute exercise and exercise training, but evidence-based conclusions cannot be drawn with high confidence due to the large heterogeneity in populations and exercise modalities. Well-designed trials with large sample sizes are needed to clarify how exercise shapes B cell-related immunity.

Graphical Abstract

1. Introduction

The human immune system is comprised of multiple, structurally diverse components that ensure immunological protection against a wide array of pathogens, ranging from simple viruses and bacteria all the way to multicellular organisms, such as parasites or fungi. Regarding the prevention and treatment of viral infections by vaccination, B cells are attracting increasing attention in immunological research.^1,2 As part of the adaptive immune system, a unique feature is their ability to proliferate and differentiate into plasma cells, which produce high amounts of pathogen-specific antibodies.³ This B cell-mediated humoral immune response ensures sustained immunity against infectious diseases, which is amplified by the generation of memory B cells, to ensure a quicker and more effective response upon reinfection with the same pathogen.⁴

Initial tentative evidence suggests that exercise training may constitute an effective adjuvant for the prevention and treatment of various chronic diseases, such as type 2 diabetes mellitus, coronary heart disease, or cancer.^5,6 Besides these non-communicable diseases, regular physical activity also protects against community-acquired infectious diseases, and reduces infectious disease mortality.⁷ Additionally, acute exercise and exercise training increase the humoral immune response to vaccination.⁸ This suggests that exercise may exhibit immunomodulatory properties that reinforce humoral immunity and could therefore be harnessed as a non-pharmacological approach with potential therapeutic value in different diseases settings.9, 10, 11 Additionally, exercise was shown to ameliorate immunosenescence and age-related morbidity, highlighting that immunological adaptions to exercise might also reduce disease burden and improve health span.^12,13 To make sophisticated statements on the immunological implications of exercise, however, it is crucial to separate immune responses induced by acute exercise from immunological adaption processes triggered by exercise training.

Acute exercise is characterized by a transient mobilization of lymphocytes into the bloodstream (i.e., lymphocytosis), followed by a decrease in the subsequent recovery period (i.e., lymphocytopenia). Although the immunological implications of the observed lymphocytopenia are under continued debate,^14,15 it is generally established that exercise characteristics, such as the type, intensity, and duration of exercise, determine the immune response.^10,11 In contrast, exercise training, which is defined as repetitive exposure to acute exercise bouts, results in immunological adaption processes (e.g., enhanced immune cell function),¹⁶ which are thought to mediate the reduced infection risk found in exercise-trained individuals. In detail, a J-shaped relationship between exercise workload and risk for upper respiratory tract infections has been established.¹⁰ This suggests that moderate exercise programs reduce the risk for upper respiratory tract infections, while prolonged and intensive exercise training increases it.¹⁰

Similarly, alterations in secretory and plasma immunoglobulin levels have been reported in response to acute exercise and exercise training. Immunoglobulins are synthesized by plasma cells and offer humoral protection against pathogens across different human biofluids, such as blood, saliva, breast milk, cerebrospinal fluid, or gastrointestinal mucus.^17,18 In the context of exercise, secretory immunoglobulin A (s-IgA) levels are thought to be altered in saliva through β-adrenergic stimulation of salivary glands, with the direction and magnitude of changes being dependent on the applied exercise modality.^19,20 Regarding plasma immunoglobulin levels, alterations in plasma volume and mobilization of peripheral, extravascular, and lymphatic immunoglobulin pools have been proposed as mechanisms for the increased levels observed after acute exercise.²¹ Although the mobilization and redistribution of immunoglobulins serve as a theoretical foundation for the potential long-term effects on humoral immunity, it remains unclear whether they are causally related to the reduced infection risk found in response to exercise training.¹⁰

Since both B cell counts and immunoglobulin levels have been shown to be exercise-sensitive, the objective of this systematic review was to evaluate the current body of evidence on the impact of acute exercise or exercise training on B cell counts and proportions, immunoglobulin levels, salivary flow rate (SFR), and s-IgA secretion rate (SR) in human biofluids. A schematic illustration of the potential relationship between the exercise interventions and outcome measures explored in this systematic review is given in Fig. 1.

Fig. 1.

Schematic illustration of the hypothesized relationship between exercise interventions and outcome measures explored in this review. Hemodynamic and hormonal changes induced by acute exercise are suspected to mobilize and redistribute immune cells and immunoglobulins from peripheral pools and the lymphatic system. Through sympathetic activation, the secretion of immunoglobulins into saliva may also be altered. Repetitive exposure to acute exercise bouts potentially results in immunological adaptions in B cell counts and immunoglobulin levels, which have crucial implications for humoral immunity. s-IgA = secretory immunoglobulin A. Created with BioRender.com.

2. Methods

This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).²² The review protocol can be accesses on PROSPERO (Registration number CRD42021245527).

2.1. Eligibility criteria

Study eligibility was defined using the Population, Intervention, Comparison, Outcomes and Study design (criteria (Table 1).²³ Only studies performed on adult (≥18 years) participants or patients were considered eligible for this review to prevent potential confounding of results due to immunological alterations during childhood or adolescence. Acute exercise and exercise training comprising endurance or resistance exercise components were considered eligible. Any other exercise interventions, such as meditation, yoga, coordination, or balance exercises, were not included. Passive control groups or other exercise groups (e.g., different type, intensity, or duration) served as comparators for exercise-induced changes in the outcomes of interest. If studies contained multiple intervention groups, only the exercise intervention groups and the control group were considered for data extraction. With the aim of including different aspects of B cell-related immunity, we extracted counts and proportions of B cells, s-IgA levels in saliva or any other secreted biofluid, SR, SFR, and plasma immunoglobulin levels (i.e., IgA, IgD, IgE, IgG, IgM) from eligible studies. This heterogenous selection of outcome measures was driven by the fact that all these outcomes are linked to adaptive immunity. For instance, immunoglobulins are secreted into human biofluids by B cells that have differentiated into plasma cells. Therefore, both plasma and salivary immunoglobulin levels, as well as B cell counts and proportions, are markers of interest when it comes to exercise-induced alterations of B cell-related immunity. As defined by the Immunology Guidebook,²⁴ cluster of differentiation (CD)19⁺ and CD20⁺ cells were considered B cells. Only randomized controlled trials (RCTs), non-RCTs (CTs), and crossover trials were included in this review to enable evidence-based conclusions on B cell-related outcomes to be drawn. Cohort studies, case-control studies, cross-sectional studies, case reports, or any other type of study or publication (e.g., conference abstracts, dissertations theses, reports, unpublished manuscripts) were not considered.

Table 1.

PICOS criteria for study eligibility.

PICOS criteria	Inclusion criteria	Exclusion criteria
Population	Healthy or diseased adults (≥18 years)	Children, adolescents, cell cultures, animal models
Intervention	Acute exercise (single bout) or exercise training (multiple bouts) comprising an endurance and/or resistance exercise component	Exercise interventions with uncertain physiological impact (e.g., meditation, yoga, coordination, or balance exercises) and combination treatments, in which exercise was complemented with non-exercise interventions (e.g., nutritional, pharmacological)
Comparison	Passive control group or exercise control group	No passive control group or exercise control group
Outcomes	Counts or proportions of B cells measured in peripheral blood; s-IgA levels measured in saliva or any other secreted fluid; SR, SFR, and plasma immunoglobulin levelsa	None of the defined outcomes measured
Study design	Randomized and non-RCTs, crossover studies	Cohort studies, case-control studies, cross-sectional studies, case reports, or any other type of study

Abbreviations: non-RCTs = non-randomized controlled trials; PICOS = Population, Intervention, Comparison, Outcomes and Study design; s-IgA = secretory immunoglobulin A; SFR = salivary flow rate; SR = s-IgA secretion rate.

^aComprising IgA, IgD, IgE, IgG, and IgM.

2.2. Information sources and search strategy

An electronic literature search was conducted across the 3 databases (MEDLINE, Web of Science, and Embase) on March 8, 2023. MEDLINE was searched through PubMed, which offers coverage from as early as 1946. Web of Science was searched through Web of Science Core Collection, which offers coverage from as early as 1900. Embase was searched through Ovid, which offers coverage from as early as 1974. Each database was searched for relevant publications using a search string with predetermined keywords relating to exercise and exercise training as well as B cells and immunoglobulins. The search string was manually modified to meet the syntactical requirements of each database, respectively. The full search strings used for the 3 databases are presented in the supplementary materials (Supplementary File 1). Since Ovid searches both Embase and MEDLINE, the search was manually limited to Embase.

2.3. Selection process

Titles and abstracts of the obtained search results were imported into a web-based systematic review software (Covidence, Veritas Health Innovation Ltd., Melbourne, VIC, Australia). Duplicate hits within and between the different databases were recognized by an automated filter implemented into the systematic review software. Recognition of duplicates by the software was manually checked for correctness by 1 investigator (SB). If duplicates were overseen by the software, they were manually marked in the process of title and abstract screening and/or full text review. Screening of titles and abstracts was performed separately by 2 independent investigators (SB, MK) at the same time. Subsequently, the 2 reviewers discussed conflicts to achieve a consensus. In case of persisting discrepancies, a third investigator (DW) was consulted. The same procedure was used for screening of full texts. Full texts were obtained from publicly available databases (e.g., PubMed, Google scholar). In case full texts were not obtainable, study investigators were contacted and asked to share a copy of their publication.

2.4. Data collection process and data items

Data collection of eligible full texts was performed by 2 independently working investigators (SB, DW). Any disagreement between the 2 investigators concerning the collected data was discussed until a consensus was achieved. In case of persisting discrepancies, a third independent investigator (PZ) was consulted. If multiple full texts originated from the same study and reported the same outcomes, the full text containing more detailed information on the outcomes of interest was included in the review. Outcomes collected from eligible studies in acute exercise settings included counts and proportions of B cells, secretory and plasma immunoglobulin levels, SR, and SFR (for details see 2.1) during and after exercise compared to baseline (within group difference) and compared to a control group (between group difference and Time × Group interaction effect). Although uncontrolled longitudinal trials were excluded from this review, the time effects reported in the included RCTs, CTs, and crossover trials were collected to explore the impact of exercise on immunological outcomes over time. Significant changes in the specified outcomes were collected for all measurement timepoints implemented in the studies. For exercise training studies, the same outcomes and comparators (baseline for time effects, control group for group effects and Time × Group interaction effects) were considered and significant changes in the outcomes were collected for all reported measurement timepoints. The outcome measures obtained immediately after acute exercise and during or after exercise training (under resting conditions) were considered most relevant and were used for a descriptive synthesis of results since they most closely resemble the immunological impact of exercise. Since post-exercise measurement timepoints differed greatly across acute exercise studies, we included all results obtained up to 2 h post-exercise in our synthesis of results and interpreted these as “immediate” effects of acute exercise. For exercise training studies, measurement timepoints both during (e.g., interim analyses) and after the exercise training intervention were considered. Further data items collected from the studies were the first author's last name and year of publication as well as the study design, sample size, participant characteristics (i.e., age, sex, health status, exercise status), and details on the exercise intervention (i.e., the type, intensity, duration, and frequency of exercise sessions) (Supplementary Tables 1 and 2). In view of the highly heterogeneous participant characteristics and exercise modalities applied, we refrained from performing a meta-analysis after the synthesis of results.

2.5. Study quality and reporting

Study quality and reporting of exercise training studies were evaluated using the Tool for the assEssment of Study qualiTy and reporting in EXercise (TESTEX).²⁵ The TESTEX is a 15-point scale consisting of 12 questions: 5 questions targeting the study quality and 7 questions targeting the study reporting. The TESTEX was designed to account for the specific characteristics of exercise training studies, for which other risk of bias tools (e.g., Cochrane's RoB 2 or ROBINS-I) are hardly applicable. Study quality and reporting were assessed for every included study by 2 independent reviewers (KW and SP). In case of discrepancies, a third independent reviewer (DW) was consulted. Since neither TESTEX nor other risk of bias tools are applicable to acute exercise studies, risk of bias was not assessed for these studies.

3. Results

3.1. Study selection

In the initial literature search, a total of 23,814 records were identified across the 3 databases. After import into the web-based systematic review software, 8302 records were automatically marked as duplicates, and 94 records were manually marked as duplicates in the process of title and abstract screening and/or full text review. Of the 15,418 records included in title and abstract screening, 15,268 were considered irrelevant, and 150 were included for detailed evaluation. During full text review, 83 records were excluded (for reasons, see Fig. 2), resulting in the inclusion of 67 records in this systematic review.

Fig. 2.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram for literature search and study selection.

3.2. Study characteristics of acute exercise studies

Of the 67 studies26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92 included in this review, 22 studies26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47 investigated the impact of acute exercise and 45 studies48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92 assessed the impact of exercise training on the defined B cell-related outcomes. Across the 22 acute exercise studies, 17 studies26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42 applied endurance exercise modalities (e.g., cycling, running, rowing, swimming), 4 studies43, 44, 45, 46 applied resistance exercise modalities (e.g., machine-based or free-weight exercise), and 1 study⁴⁷ applied a combined exercise regimen (i.e., cycling and machine-based resistance exercise). Concerning the study design, 16 studies26, 27, 28^,30, 31, 32, 33^{,35,36,38,39,42,}44, 45, 46, 47 used a crossover study design, 3 studies^29,34,41 used a RCT design, and 3 studies^37,40,43 used a CT design. The total sample size of participants included in acute exercise studies was 483 (330 men, 153 women). Two studies^32,40 investigated diseased populations, including obese participants and patients with amphetamine addiction, respectively. Concerning the outcomes assessed, 6 studies26, 27, 28, 29^,43,47 investigated changes in B cell counts or proportions; 11 studies30, 31, 32, 33, 34, 35, 36, 37, 38, 39^,46 investigated changes in s-IgA levels, SR, or SFR; and 5 studies40, 41, 42^,44,45 investigated plasma immunoglobulin levels. One study³⁹ assessed the impact of acute endurance exercise on breast milk IgA levels. A detailed overview of the study characteristics and significant results of each study separated by exercise modality and sorted by assessed outcomes is given in Supplementary Table 1.

3.3. Study characteristics of exercise training studies

Across the 45 exercise training studies48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92 included in this review, 24 studies48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71 employed endurance training, 10 studies72, 73, 74, 75, 76, 77, 78, 79, 80, 81 employed resistance training, and 11 studies82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92 employed a combined training regimen to assess the impact on B cell-related outcomes. Regarding the study design, 32 studies^48,49,52,56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69^,72, 73, 74^,77,78,80, 81, 82^,84,85,87, 88, 89^,91,92 used an RCT design, 11 studies53, 54, 55^{,70,71,75,76,79,83,86,90} used an CT design, and 2 studies^50,51 used a crossover design. In total, 1837 participants were included in the exercise training studies. Thirteen studies^{48,49,56,58,67,68,71,80,}82, 83, 84^,87,89 were performed on patient collectives, including patients with rheumatoid arthritis, chronic fatigue syndrome, amnestic mild cognitive impairment, obesity, breast cancer, chronic lymphocytic leukemia, coronavirus disease 2019, and Down syndrome. B cell counts and proportions were assessed in 21 studies,48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58^,72, 73, 74, 75, 76, 77, 78^,82, 83, 84 with 2 studies^55,56 also distinguishing between total, naïve, and memory B cells, and 1 study⁸⁴ additionally assessing regulatory B cells. Levels of s-IgA, SR, or SFR were analyzed in 13 studies,56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69^,79, 80, 81^,85,86 and plasma immunoglobulin levels were analyzed in 14 studies.^57,58,66, 67, 68, 69, 70, 71^,87, 88, 89, 90, 91, 92 Of note, several studies assessed multiple outcomes. A detailed overview of the study characteristics separated by exercise modality and ordered by outcomes assessed is given in Supplementary Table 2.

3.4. Assessment of study quality and reporting of exercise training studies

Study quality and reporting of exercise training studies were assessed by 2 independent reviewers (SP, KW) with an inter-rater correlation coefficient of 84.96%. None of the 45 exercise training studies fulfilled all items on the 15-point TESTEX scale. The obtained scores ranged from 2 to 13 points with an average of 6.84 points or 47.15%. Concerning the first 5 items of the TESTEX, which are used to determine the study quality, 8 studies^{49,60,64,69,78,82,84,91} scored ≥4 points and 32 studies^48,50, 51, 52, 53, 54, 55, 56^{,59,61,63,65,66,68,}70, 71, 72, 73, 74, 75, 76^,79, 80, 81^,83,85, 86, 87, 88, 89, 90^,92 scored ≤2 points (mean = 2.09 points). In items 6–12 of the TESTEX, which are used to determine the quality of study reporting, 15 studies^49,56, 57, 58^{,60,64,68,69,72,77,82,83,86,91,92} scored ≥6 points and 15 studies^51,53, 54, 55, 55^,59,61,73, 74, 75, 76^{,79,81,84,88,90} scored ≤4 points (mean = 4.76 points). A graphical depiction of the relative number of studies that fulfilled each of the 15 TESTEX items, respectively, is given in Fig. 3.

Fig. 3.

Study quality and reporting of exercise training studies. Item 1: eligibility criteria specified; Item 2: randomization specified; Item 3: allocation concealment; Item 4: groups similar at baseline for 1 key outcome; Item 5: blinding of assessors for ≥1 key outcome; Item 6.1: outcome measures assessed in ≥85% of participants; Item 6.2: adverse events are reported; Item 6.3: exercise attendance is reported; Item 7: intention-to-treat analysis; Item 8.1: between-group statistical comparisons reported for primary outcome of interest; Item 8.2: between-group statistical comparisons reported for ≥1 secondary outcome; Item 9: point measures and measures of variability for all reported outcome measures; Item 10: activity monitoring in control groups; Item 11: relative exercise intensity remained constant; Item 12: exercise volume and energy expenditure. TESTEX = Tool for the assEssment of Study qualiTy and reporting in EXercise.

3.5. Impact of acute exercise on B cell-related outcomes

The synthesis of results performed with acute exercise studies is summarized in Table 2. A detailed report of all results, irrespective of measurement timepoint, is given in Supplementary Table 1.

Table 2.

Overview of exercise-induced changes in measures of B cell-related outcomes.

Outcome	Number of studies assessing the outcome	Number of studies assessing changes over time (Pre- vs. post-exercisea)		Number of studies assessing changes compared to control (Group effects or group × time interaction effects)
Acute exercise
B cell counts	4	↑ ↔ ↓	1 47 1 27 2 26,28	↑ ↔ ↓	– 3 26,27,47 1 28
B cell proportion	3	↑ ↔ ↓	– 1 43 2 26,29	↑ ↔ ↓	– 2 26,43 1 29
s–IgA levels	11	↑ ↔ ↓	4 30,31,34,46 3 32,35,36 4 33,37, 38, 39	↑ ↔ ↓	– 9 30, 31, 32, 33, 34, 35, 36, 37,46 2 38,39
SR	9	↑ ↔ ↓	3 30,31,35 5 32, 33, 34,36,38 1 37	↑ ↔ ↓	1 31 8 30,32, 33, 34, 35, 36, 37, 38 –
SFR	8	↑ ↔ ↓	3 33,35,37 2 30,31 3 34,36,38	↑ ↔ ↓	– 7 30,31,33, 34, 35, 36, 37 1 38
Plasma immunoglobulins
IgA	4	↑ ↔ ↓	2 40,44 2 41,42 –	↑ ↔ ↓	1 40 3 41,42,44 –
IgD	Not assessed		–		–
IgE	Not assessed		–		–
IgG	4	↑ ↔ ↓	2 40,42 2 41,45 –	↑ ↔ ↓	2 40,42 2 41,45 –
IgM	2	↑ ↔ ↓	– 2 41,42 –	↑ ↔ ↓	– 2 41,42 –
Exercise training
B cell counts	19	↑ ↔ ↓	4 49,52,73,75 11 50,51,53,54,56, 57, 58,74,76, 77, 78 2 72,82	↑ ↔ ↓	2 73,74 15 49, 50, 51, 52, 53, 54,56, 57, 58,72,75, 76, 77, 78,82 –
B cell proportion	3	↑ ↔ ↓	1 55 – 1 84	↑ ↔ ↓	– 2 55,84 –
s-IgA levels	14	↑ ↔ ↓	3 62,80,85 9 59, 60, 61,64, 65, 66,78,79,86 1 81	↑ ↔ ↓	3 62,80,85 10 59, 60, 61,64, 65, 66,78,79,81,86 –
SR	4	↑ ↔ ↓	3 65,85,86 1 66 –	↑ ↔ ↓	2 85,86 2 65,66 –
SFR	5	↑ ↔ ↓	– 5 65,66,78,85,86 –	↑ ↔ ↓	1 85 4 65,66,78,86 –
Plasma immunoglobulins
IgA	11	↑ ↔ ↓	7 58,66, 67, 68,70,87,89 4 57,69,88,90 –	↑ ↔ ↓	5 67,68,70,87,89 5 57,58,66,88,90 1 69
IgD	Not assessed		–		–
IgE	3	↑ ↔ ↓	– 2 66,92 1 70	↑ ↔ ↓	– 3 66,70,92 –
IgG	10	↑ ↔ ↓	7 58,66,70,71,89, 90, 91 2 57,88 2 91,92	↑ ↔ ↓	2 70,71 6 57,58,66,88, 89, 90 2 91,92
IgM	9	↑ ↔ ↓	3 58,70,91 6 57,66,69,88, 89, 90 1 91	↑ ↔ ↓	1 70 6 57,58,66,88, 89, 90 2 69,91

Notes: ↑ means significant increase; ↓means significant decrease; ↔ means outcome unchanged; – means no studies reporting these results.

Abbreviations: Ig = immunoglobulin; s-IgA = secretory immunoglobulin A; SFR = saliva flow rate; SR = s-IgA secretion rate.

^aMeasurement timepoints up until 2 h after the exercise session were treated as post-exercise.

3.5.1. Impact of acute exercise and B cell counts and proportions

Of the 6 studies investigating acute exercise-induced alterations in B cells, 3 studies^27,28,47 investigated B cell counts, 2 studies^29,43 investigated B cell proportions, and 1 study⁴⁷ investigated both. B cell counts increased from pre- to post-exercise in a study applying short but intense continuous cycling exercise⁴⁷ but decreased in the recovery phase of 2 other studies applying continuous or intermittent endurance exercise of different durations.^26,28 Of note, the decrease in B cell count was also significantly different compared to sedentary controls.²⁸ Similarly, B cell proportions decreased after cycling exercise compared to baseline^26,29 and compared to a sedentary control group.²⁹ No changes in B cell counts were observed in participants performing low-intensity cycling²⁷ or light eccentric leg exercise.⁴³

3.5.2. Impact of acute exercise on markers of secretory immunity

Of the acute exercise studies investigating secretory immunity, 11 studies30, 31, 32, 33, 34, 35, 36, 37, 38, 39^,46 investigated s-IgA levels, 9 studies30, 31, 32, 33, 34, 35, 36, 37, 38 investigated SR, and 8 studies^30,31,33, 34, 35, 36, 37, 38, 39 investigated SFR. Levels of s-IgA increased in 4 studies^30,31,34,46 from pre- to post-exercise, remained unchanged in 3 studies,^32,35,36 and decreased in 3 further studies.^33,37,38 Of note, only 1 study showed a significant decrease in s-IgA levels as compared to control conditions.³⁸ Besides saliva IgA counts, acute treadmill exercise until exhaustion also reduced IgA counts in breast milk of postpartum women compared to sedentary postpartum controls.³⁹ Concerning SR, most studies did not show significant alterations induced by acute exercise.32, 33, 34^,36,38 However, 3 studies observed an increase^30,31,35 and 1 study reported a decrease of SR (Table 2).³⁷ Besides these time effects, another study revealed a significant increase in SR after cycling at 80% V̇O_2max until exhaustion compared to cycling at 55% V̇O_2max for 3 h.³¹ For SFR, only 1 out of 8 studies showed a significant decrease in SFR as compared to control conditions (Table 2).³⁸

3.5.3. Impact of acute exercise on plasma immunoglobulin levels

Of the studies investigating plasma immunoglobulin levels, 4 studies investigated IgA40, 41, 42^,44 and IgG40, 41, 42^,45 and 2 studies investigated IgM.^41,42 IgD and IgE were not assessed in the context of acute exercise. For plasma IgA levels, a significant increase after exercise as compared to control conditions was found in 2 studies,^40,44 while 2 others did not show any significant alterations.^41,42 Of note, 1 of the 2 studies showing no immediate changes in plasma IgA levels revealed elevated IgA levels 2 and 5 days after exercise as compared to sedentary controls.⁴¹ Similarly, plasma IgG levels were increased immediately after acute exercise as compared to control in 2 studies,^40,42 and 2 further studies revealed increases between 2 and 5 days after the acute exercise session.^41,45 For plasma IgM levels, no immediate effects of acute exercise, but an increase 2 and 5 days after acute exercise was found.⁴¹ Of note, the studies showing increased plasma immunoglobulin levels several days after the acute exercise session were not included in the synthesis of results, which is displayed in Table 2.

3.6. Impact of exercise training on B cell-related outcomes

The synthesis of results performed with exercise training studies is summarized in Table 2. A detailed report of all results, irrespective of measurement timepoint, is given in Supplementary Table 2.

3.6.1. Impact of exercise training on B cell counts and proportions

Across all exercise training studies, 18 studies 48, 49, 50, 51, 52, 53, 54^,56, 57, 58^,72, 73, 74, 75, 76, 77, 78^,82 investigated B cell counts, 2 studies^55,84, investigated B cell proportions, and 1 study⁸³ investigated both. Most studies did not find differences in B cell counts over time or compared to a control group (Table 2). An increase in B cell counts over time was found in 4 studies applying different types of endurance or resistance training in healthy, young participants^52,73,75 and middle-aged patients with chronic fatigue syndrome.⁵¹ Additionally, 2 studies showed a significant increase in B cell counts after exercise training as compared to a passive control group.^73,74 In contrast, a decrease in B cell counts over time was found in healthy, elderly participants exposed to resistance training⁷² and breast cancer patients who performed a combined endurance and resistance exercise program while receiving chemotherapy.⁸² Concerning B cell proportions, no alterations were found as compared to control conditions. However, an increase in naive B cells over time was found in healthy university students after 14 weeks of endurance exercise.⁵⁵ In contrast, a decrease in regulatory B cell proportions was found in elderly patients with rheumatoid arthritis after 20 weeks of combined exercise training.⁸⁴ Of note, 2 studies did not report statistical comparisons for B cell counts or proportions.^48,83

3.6.2. Impact of exercise training on markers of secretory immunity

Of all the studies assessing secretory immunity, 14 studies59, 60, 61, 62, 63, 64, 65, 66^,78, 79, 80, 81^,85,86 investigated s-IgA levels, 4 studies^65,66,85,86 investigated SR, and 5 studies^{65,66,78,85,86} investigated SFR. As for B cell counts, most studies did not find significant alterations in s-IgA levels after exercise training (Table 2). However, a significant increase in s-IgA levels as compared to a passive control group was found in healthy and diseased participants performing endurance training,⁶² resistance training⁸⁰ or a combination of both.⁸⁵ These results were paralleled by an increased SR in elderly participants exposed to resistance training.^85,86 In contrast, a decrease in s-IgA levels as compared to control conditions was found in a comparable population of elderly participants performing less strenuous chair-based resistance exercises 2–3 times per week for 14 weeks.⁸¹ SFR remained unaltered by exercise training (Table 2). Of note, 1 investigation only compared the response to an acute bout of exercise before and after the exercise training program⁶³ and was therefore excluded from the synthesis of results (Supplementary Table 2 and Table 2).

3.6.3. Impact of exercise training on plasma immunoglobulin levels

Of the plasma immunoglobulins, IgA, IgG, and IgM were most frequently investigated (Table 2). IgE was investigated in 3 studies,^66,70,92 and IgD was not studied at all (Table 2). For IgA and IgG, several studies demonstrated an increase in response to exercise training as compared to controls.^67,68,70,^71,87,89 Of note, this effect was more pronounced for IgA^67,68,70,^71,87,89 than for IgG^70,71 (Table 2). In contrast, decreased IgG levels were found in elderly participants performing 25 weeks of combined exercise⁹² and in a subgroup of participants who received a vaccination against the coronavirus disease 2019.⁹¹ Concerning IgE, no significant alterations compared to control conditions were found (Table 2). IgM levels increased in 1 study⁷⁰ and decreased in 2 studies as compared to the control group.^69,91

4. Discussion

The aim of this systematic review was to assess the impact of acute exercise and exercise training on B cell numbers and proportions, immunoglobulin levels, and markers of secretory immunity (i.e., s-IgA and SR) in human biofluids. Despite high heterogeneity in the type, duration, intensity, and frequency of the applied exercise sessions, several investigations revealed alterations in B cell-related outcomes as compared to control conditions, thereby providing tentative evidence that B cell-related immunity is shaped by acute exercise and exercise training.

4.1. Impact of acute exercise on B cell-related outcomes

4.1.1. Impact of acute exercise on B cell counts and proportions

In most of the studies, acute exercise did not cause significant alterations in B cell counts or proportions as compared to control conditions (Table 2). However, the increase in B cell counts from pre- to post-exercise,⁴⁷ and the decrease in the subsequent recovery phase,^26,28 is consistent with the general exercise kinetics of lymphocytes.⁹³ Since the mobilization of lymphocytes is dependent on shear stress and catecholamine release, this might also explain why no changes in B cell counts were found in 2 studies applying light exercise intensities.^27,43 Although the immunological consequences of reduced B cell counts after acute exercise are still unknown, a potential interpretation might be a transmigration of B cells from the bloodstream into peripheral tissues (e.g., mucosa-associated lymphoid tissue and lymph nodes) to ensure adequate immune surveillance and effective protection against pathogens entering the organism.^10,14 Additional evidence for the impact of physical activity on leukocyte mobilization and redistribution originates from the circadian regulation of immune functions. As reviewed by Scheiermann et al.,⁹⁴ leukocyte trafficking differs considerably between the resting and active phase in mammals, thereby highlighting the potential implications of acute exercise in B cell redistribution.

The decreased proportion of B cells observed during and immediately after acute cycling exercise at different intensities (Table 2) may be explained as a consequence of the strong proportional increase of natural killer cells.^26,29 Although the decrease in B cell proportions is therefore of a secondary nature, this does not rule out the possibility that lower B cell proportions relative to the total immune cell compartment might have a physiological impact.

4.1.2. Impact of acute exercise on secretory immunity

Similar to B cells, most of the studies assessing secretory immunity did not find significant alterations in s-IgA counts, SR, or SFR as compared to control conditions. Although SR revealed an intensity-dependent increase,³¹ s-IgA levels decreased in the recovery period following exercise as compared to sedentary controls.³⁸ Interestingly, this effect was also observed in the breast milk of postpartum women,³⁹ suggesting that the reduction in secretory immunity found post-exercise is not only limited to saliva but might extend to other secreted fluids, such as breast milk, lacrimal fluid, or mucus of the gastrointestinal tract. In fact, the impact of exercise on mucosal immunity has been reviewed multiple times and is a well-appreciated concept in terms of respiratory infections in athletes.^95,96 A review by Nieman et al.¹⁰ confirmed that prolonged and intensive exercise is linked to immune dysfunction and increased infection risk, which might explain the reduced s-IgA levels found after cycling and treadmill exercise.^38,39 In view of the strongly heterogenous results obtained for SFR (Table 2), no clear conclusions on the impact of acute exercise can be drawn. The only investigation showing significant changes in comparison to a control group revealed a decrease in SFR after 60 min of cycling at 75% V̇O_2max,³⁸ which is in accordance with the sympathetic innervation of salivary glands that results in an increased viscosity of saliva. Overall, the results on secretory immunity need to be considered cautiously since there is a considerable number of studies showing no alterations in the associated outcomes from pre- to post-exercise (Table 2). The fact that only 3 studies^31,38,39 showed significant group or interaction effects compared to a control group highlights the limited nature of the results.

4.1.3. Impact of acute exercise on plasma immunoglobulin levels

Compared to secretory immunity, a much clearer pattern can be found for plasma immunoglobulins in response to acute exercise. While plasma IgA and IgG levels revealed exercise-sensitivity, IgM remained unchanged.^41,42 Interestingly, plasma IgA and IgG levels increased both immediately after exercise^40,42,44 and in the subsequent days following the exercise session,^41,45 indicating that an acute exercise bout induces a sustained increase in immunoglobulin levels. Although these results have promising implications for humoral immunity, especially in the context of vaccination regimes, interpretation requires caution since no differences were found in comparison to the control conditions. Considering that acute exercise has been discussed as a low-cost adjuvant approach for increasing the humoral response to vaccinations,^8,97 future studies will have to show whether such approaches provide statistically meaningful benefits.

4.2. Impact of exercise training on B cell-related outcomes

Concerning the low study quality (see Section 3.4 and Fig. 3) and high heterogeneity in populations and exercise modalities, evidence-based conclusions on the long-term effects of exercise training on B cell-related outcomes and immunity cannot be drawn with high confidence. However, there were several investigations that revealed group or interaction effects in comparison to control conditions, indicating that exercise training has the potential to induce immunological adaptions, which might be promising for protection against infectious diseases.

4.2.1. Impact of exercise training on B cell counts and proportions

Overall, B cell counts and proportions remained constant compared to control conditions in most of the studies included in this review (Table 2). However, some investigations were able to show an increase over time^49,52,73,75 and compared to control conditions.^73,74 This effect was almost exclusively observed in healthy participants,^52,73,75 suggesting that a chronic increase in B cell counts is generally possible under certain physiological conditions. Whether this might also apply to patient collectives other than those investigated⁴⁹ remains inconclusive. In line with the increased proportion of naïve B cells (assessed as percentage of total B cells) found in healthy university students after 14 weeks of endurance exercise,⁵⁵ increased hematopoiesis and/or mobilization of naïve B cells from the bone marrow might be responsible for the increase in B cell proportions observed after exercise training. Despite promising implications for adaptive immunity, the immunological consequences of this increase are unknown and require further investigation. The fact that decreased B cell counts were observed after exercise training in elderly participants⁷² and in patients receiving chemotherapy⁸² reinforces the notion that chronic exercise adaptions in B cell counts and proportions might differ between populations.

4.2.2. Impact of exercise training on secretory immunity

The impact of exercise training on secretory immunity revealed largely consistent results across the included investigations. Although s-IgA levels remained unaltered in most studies (Table 2), significant increases compared to control conditions were found in 3 different populations (i.e., healthy young and old participants and participants with Down syndrome) in response to different exercise training programs.^62,80,85 These results highlight that long-term adaptions in s-IgA levels are possible, with potential implications of exercise training for secretory immunity. In fact, cross-sectional analyses have revealed that trained subjects have higher resting s-IgA levels compared to untrained controls,⁹⁸ which is indicative of immunological adaptions in secretory immunity. However, the precise mechanisms mediating these effects remain unresolved.⁹⁶ The decreased s-IgA levels found in another population of elderly participants might be explained by the lower exercise intensity and shorter duration of the exercise program applied.⁸¹ Reinforcing the notion of exercise-induced adaptions in secretory immunity, an increased SR compared to baseline was observed in 3 studies^66,85,86 and compared to controls in 2 studies.^85,86

4.2.3. Impact of exercise training on plasma immunoglobulin levels

Plasma immunoglobulins form a further important component of B cell-related immunity that is characterized by continuous mobilization and redistribution to potential sites of infection via the bloodstream. The increases in IgA and IgG levels observed in response to exercise training shed light on the crucial implications of regular exercise for humoral immunity. Of note, IgA seems to be more responsive to exercise training, as revealed by a higher number of studies showing group or interaction effects for IgA counts as compared to other immunoglobulins (Table 2).^{67,68,70,87,89} Mechanistically, plasma volume changes, influx of extravascular immunoglobulin pools, and increased lymphatic flow have been suggested to mediate alterations in immunoglobulin levels in response to acute exercise.²¹ Since plasma volume has been shown to increase in response to exercise training,⁹⁹ a decrease in immunoglobulin levels would be expected. This might suggest that there are other factors, such as an increased mobilization of immunoglobulins from peripheral and extravascular pools and/or lymph nodes, responsible for the increased values found after exercise training. Although this provides a general mechanism of action for increased immunoglobulin levels, the mechanistic underpinnings related to the distinct responses of different immunoglobulins (i.e., IgA, IgE, IgD, IgG, IgM) have not been elucidated.

4.3. Limitations and future perspectives

To facilitate the drawing of evidence-based conclusions, we limited the literature search of this systematic review to investigations characterized by a high level of evidence (i.e., randomized and non-RCTs and crossover studies; Table 1). One drawback of this approach, however, is that investigations applying other study designs were not considered, although some of these studies have contributed significantly to our understanding of immunological alterations induced by exercise. Since B cell counts were shown to remain unaltered under resting conditions but increase in response to higher intensity exercise,¹⁰⁰ uncontrolled study designs can offer crucial insights into specific research questions, such as the precise B cell phenotype mobilized by exercise. Extensive phenotyping of B cell subsets has revealed that immature, naïve, and memory B cells, as well as B1 cells and plasma cells, increase in response to cycling exercise, with immature B cells showing the largest proportional increase.¹⁰¹ Detailed characterizations like these are crucial for evaluating potential health-related benefits originating from mobilization of specific B cell subsets. For instance, since plasma cells are a crucial component of both the acute and memory-adaptive immune response, exercise-induced alterations could represent a promising avenue for improved immunological protection and reduced infection risk. The fact that plasma cells participate in numerous other physiological processes, beyond antibody secretion,¹⁰² additionally highlights the far-reaching implications that exercise regimens could have. Complicating investigations on this cell type, however, plasma cells are usually not detectable in the peripheral blood of healthy people,^103,104 which is why the impact of acute exercise and/or exercise training on plasma cell counts remains largely unknown so far.

Reinforcing the benefits of alternative study designs, cross-sectional studies have also provided valuable insights into the impact of exercise on B cell-related outcomes. Exercise training studies are usually relatively short in duration (e.g., several weeks to months), and it remains questionable whether this timeframe is sufficient to allow for immunological exercise adaptions. Cross-sectional analyses can provide a remedy for this methodological limitation. A comparison of lifelong master athletes with age-matched sedentary controls and younger sedentary controls revealed that the proportions of several B cell subsets are altered towards a younger phenotype in master athletes.¹² Considering the long half-life of immunoglobulins (typically weeks to months),¹⁰⁵ cross-sectional analyses are also relevant for sensitive detection of exercise-induced adaptions in immunoglobulin levels. For instance, elite rowers and swimmers were shown to have higher resting salivary IgA levels compared to non-athletes and moderately exercising controls, respectively.^98,106 These results highlight how regular exercise can shape secretory immunity, with potential implications for sustained health and prevention of infectious diseases. For a comprehensive overview of investigations on this topic please refer to previous reviews.^21,96

On the contrary, the long half-life of immunoglobulins might also impede sensitive detection of changes in immunoglobulin levels in acute exercise settings. Overcoming this limitation, immunoglobulin free light chains, which are synthesized in surplus and released into the bloodstream during immunoglobulin production,¹⁰⁷ might depict a promising outcome measure. Although the exact biological function of free light chains—especially in healthy populations—remains unknown, they represent a sensitive biomarker for disease diagnostics and progression that can be quantified in different human biofluids (e.g., blood, saliva, urine, or cerebrospinal fluid).^108,109 Interestingly, salivary free light chains exhibit a similar kinetic as salivary IgA in response to acute exercise¹¹⁰ but show lower intra-individual variation,¹¹⁰ which is why they are increasingly recognized as an adjunct biomarker for exercise stress and oral inflammation.^19,111 In addition, their shorter half-life (2–6 h¹¹²) compared to fully matured immunoglobulins facilitates a more sensitive quantification of immunoglobulin production in acute exercise settings.

5. Conclusion

B cell-related immunity is a crucial part of the adaptive immune system that is comprised of both a cellular component formed by B cells and a humoral component formed by pathogen-specific antibodies. Current research in the field of exercise immunology is mostly focused on exercise-secreted immunological factors, such as interleukin-6, T lymphocytes, and natural killer cells. Herein, we systematically assessed the effect of acute exercise and exercise training on B cell-related outcomes to uncover potential effects on immunological defense and protection from infectious diseases. With the aim to draw evidence-based conclusions on the impact of exercise, only RCTs, CTs, and crossover trials were included in this systematic review. Overall, acute exercise studies (n = 22) and exercise training studies (n = 45) revealed heterogenous results, which might be explained by the different populations and exercise regimens investigated. Acute exercise studies indicated changes in all B cell-related outcomes over time, but only a few studies showed significant differences compared to the control conditions. In exercise training studies, most outcomes remained unaltered compared to the control conditions. Although both, s-IgA and plasma IgA levels were increased in some studies, the results remain inconclusive in view of contrary results found in other investigations. Considering the overall heterogeneity in exercise studies included in this review, together with the low methodological study quality and poor reporting, evidence-based conclusions on the immunological impact of exercise cannot be drawn with high confidence. In special consideration of the items of established risk of bias and quality assessment tools (e.g., RoB 2, ROBINS-I, TESTEX), future high-quality studies are warranted to improve the overall evidence on exercise-induced adaptions of the human immune system. Larger sample sizes, lower risk of bias due to methodological shortcomings, and quantification of cellular and secretory functions of B cells are urgently needed to pave the way for solid conclusions on the impact of exercise on B cell-related immunity. Ultimately, this holds the potential to advance our understanding of the immunological implications of exercise, with promising implications for the prevention and treatment of infectious diseases.