Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Since January 2020 Elsevier has created a COVID-19 resource centre with
free information in English and Mandarin on the novel coronavirus COVID-
19. The COVID-19 resource centre is hosted on Elsevier Connect, the
company's public news and information website.
Elsevier hereby grants permission to make all its COVID-19-related
research that is available on the COVID-19 resource centre - including this
research content - immediately available in PubMed Central and other
publicly funded repositories, such as the WHO COVID database with rights
for unrestricted research re-use and analyses in any form or by any means
with acknowledgement of the original source. These permissions are
granted for free by Elsevier for as long as the COVID-19 resource centre
remains active.
Brain, Behavior, and Immunity 102 (2022) 1–10
Available online 5 February 2022
0889-1591/© 2022 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Full-length Article
Exercise after inuenza or COVID-19 vaccination increases serum antibody
without an increase in side effects
Justus Hallam
a
,
b
, Tyanez Jones
c
, Jessica Alley
a
,
b
, Marian L. Kohut
a
,
b
,
c
,
*
a
Department of Kinesiology, Iowa State University, Ames, IA, USA
b
Program of Immunobiology, Iowa State University, Ames, IA, USA
c
Nanovaccine Institute, Iowa State University, Ames, IA, USA
ARTICLE INFO
Keywords:
Exercise
Vaccine
Inuenza
COVID-19
ABSTRACT
Vaccination is an effective public health measure, yet vaccine efcacy varies across different populations. Ad-
juvants improve vaccine efcacy but often increase reactogenicity. An unconventional behavioral “adjuvant” is
physical exercise at the time of vaccination. Here, in separate experiments, we examined the effect of 90-minute
light- to moderate-intensity cycle ergometer or outdoor walk/jog aerobic exercise performed once after immu-
nization on serum antibody response to three different vaccines (2009 pandemic inuenza H1N1, seasonal
inuenza, and COVID-19). Exercise took place after inuenza vaccination or after the rst dose of Pzer-
BioNTech COVID-19 vaccine. A mouse model of inuenza A immunization was used to examine the effect of
exercise on antibody response and the role of IFN
α
as a potential mechanism by treating mice with anti-IFN
α
antibody. The results show that 90 min of exercise consistently increased serum antibody to each vaccine four
weeks post-immunization, and IFN
α
may partially contribute to the exercise-related benet. Exercise did not
increase side effects after the COVID-19 vaccination. These ndings suggest that adults who exercise regularly
may increase antibody response to inuenza or COVID-19 vaccine by performing a single session of light- to
moderate-intensity exercise post-immunization.
1. Introduction
Physical exercise performed near the time of immunization may in-
crease antibody response to vaccination. Several studies have reported
such ndings, demonstrating that exercise preceding immunization
improved antibody response (Edwards et al. 2012; Edwards et al. 2006;
Edwards, et al. 2007; Ranadive et al. 2014). One explanation given for
these results is that exercise may act as an acute stressor. There are
examples in the literature that demonstrate that acute stress may in-
crease antibody response when applied before immunization (Edwards,
et al. 2008; Edwards et al. 2006; Silberman et al. 2003). It has also been
suggested that eccentric exercise produces a local inammatory
response, potentially resulting in greater antigen-presenting cell acti-
vation. In some studies, eccentric exercise before vaccination increased
antibody response (Edwards, et al. 2007). As another possibility, it is
recognized that an increase in serum IL-6 accompanies exercise (Reih-
mane and Dela, 2014; Vasconcelos and Salla, 2018), and the exercise-
associated change in IL-6 may be another mechanism by which exer-
cise may enhance antibody response (Edwards et al., 2006). Studies
show the administration of IL-6 at the time of inuenza immunization
increases IgG, mediated indirectly by CD4
+
T cells (Dienz, et al. 2009).
IL-6 has a role in T follicular helper (TFH) CD4
+
T cell differentiation
(Choi, et al. 2013; Eto, et al. 2011), permitting germinal center TFH cells
to receive continued T-cell receptor signaling (Papillion, et al. 2019).
Therefore, evidence supports the possibility that increased IL-6 at the
time of exercise could contribute to exercise-induced increases in serum
antibody. However, a consistent association between IL and 6, exercise,
and antibody response to vaccine has not been observed, and as a result,
the mechanisms to explain the association between exercise and vaccine
response remain speculative.
The current research on exercise and vaccination shows promising
ndings in several studies (Edwards, et al. 2006; Edwards, et al. 2007;
Edwards, et al. 2008; Ranadive, et al. 2014), but signicant challenges
remain. For example, the results across studies are inconsistent, as some
studies demonstrate no benet of exercise (Bohn-Goldbaum, et al. 2020;
Bohn-Goldbaum, et al. 2019; Bruunsgaard, et al. 1997; Campbell, et al.
2010; Long, et al. 2012). Others report an exercise-related response to
one antigen in a vaccine but no effect on the response to other antigens
* Corresponding author at: Department of Kinesiology, Iowa State University, 5203 ATRB, Ames, IA 50011-4008, USA.
E-mail address: mkohut@iastate.edu (M.L. Kohut).
Contents lists available at ScienceDirect
Brain Behavior and Immunity
journal homepage: www.elsevier.com/locate/ybrbi
https://doi.org/10.1016/j.bbi.2022.02.005
Received 26 October 2021; Received in revised form 27 January 2022; Accepted 1 February 2022
Brain Behavior and Immunity 102 (2022) 1–10
2
contained in the same vaccine (Edwards, et al. 2006; Ranadive, et al.
2014). Ideally, an effective adjuvant would be expected to demonstrate
a consistent enhancement across all antigens in a vaccine. The theory
that eccentric exercise might elicit an inammatory response to boost
antibody has not held up consistently, given the ndings showing no
benet of eccentric exercise or a difference only for one sex (Campbell,
et al. 2010; Edwards, et al. 2007). Given the inconsistencies across
studies, it has been suggested that an effect of exercise may be observed
only under conditions in which antigen dose is low, or participants tend
towards a reduced antibody response (Edwards, et al. 2012). This
interpretation of existing ndings implies that exercise effects are small
in magnitude compared to the large immune stimulus from a vaccine,
and therefore may be difcult to detect.
In order to advance the concept of exercise as a vaccine “behavioral
adjuvant,” with translational public health relevance, it is essential to
dene the exercise parameters that consistently result in enhanced
serum antibody. It is also critical to identify the vaccine platforms and
pathogens for which an effect of exercise may be present. In this study,
we evaluated the effect of standardized aerobic exercise administered
after immunization in contrast to other studies that focused on exercise
prior to vaccine. In the stress and immunity literature, a stressor applied
after immunization has been shown to enhance antibody response to
vaccination (Karp, et al. 2000; Wood, et al. 1993). Therefore, we eval-
uated the effect of exercise on antibody response when exercise was
administered after immunization rather than before.
Additional rationale for selecting 90 min of exercise was based partly
on unpublished ndings demonstrating that 90 min of exercise results in
a signicant increase in the type I interferon, interferon-alpha (IFN
α
)
production by plasmacytoid dendritic cells upon activation (Supplement
Table S1). Type I IFNs promote dendritic cell activation (Montoya, et al.
2002), increase antibody production, promote class switching (Le Bon,
et al. 2001), and may have a direct stimulatory effect on B cells and T
cells (Le Bon, et al. 2006), Adjuvants that induce type I IFN potently
increase antibody response to vaccination (Junkins, et al. 2018).
Therefore, in addition to studies with human participants, a rodent
model was included. In the rodent experiments, anti-IFN
α
antibody was
used to block IFN
α
at the time of immunization to evaluate whether
IFN
α
may be one mechanism contributing to the exercise-induced
enhancement of antibody response. We also evaluated 45 min of exer-
cise in some experiments involving young and aged adults as aged adults
may be more readily able to complete 45 min of light-intensity exercise
instead of 90 min. We examined the role of exercise in response to a
“novel” H1N1 antigen with vaccine response to 2009 H1N1 Pandemic
(H1N1pdm09) virus and the response to a trivalent seasonal inuenza
virus in which neither type of inuenza A would be considered novel.
Finally, in early studies with human participants, we examined the in-
uence of exercise on antibody response to an mRNA-based vaccine,
PzerBioNTech BNT162b2, against the disease caused by the novel
coronavirus SARS-CoV-2 (COVID-19), using the same exercise approach
that we found to be effective in inuenza experiments. As the inuenza
vaccine platforms consisted of split-virus preparations and the P-
zerBioNTech BNT162b2 vaccine was based on an mRNA platform, we
compared the effect of exercise across different vaccine platforms. Ex-
ercise has been proposed as a potential “behavioral” adjuvant for
COVID-19 immunization (Valenzuela, et al. 2021), and the experiments
conducted here investigated that possibility.
2. Methods
For all experiments involving human subjects, participants were
recruited by yers posted in the local community as well as by email sent
to university staff, students, or community organizations and businesses
that schedule group vaccination clinics.
2.1. Inuenza vaccine research participants
A total of 20 participants were enrolled in the monovalent Inuenza
A/California/7/09 H1N1 vaccine experiment, and 16 were included in
the nal analysis (see Supplement Fig. 1 CONSORT diagram). A total of
28 participants were enrolled in the trivalent seasonal inuenza vaccine
experiment, and 26 were included in the nal analysis (see Supplement
Fig. 2 CONSORT diagram). Individuals were excluded if they were:
taking medications for psychological disorders or medications that
altered immune variables of interest (e.g., oral corticosteroids); had any
medical condition that may directly impact immune outcomes,
including autoimmune disorders; or were unable to perform the pre-
scribed exercise safely. Participants were included if they had been
exercising regularly for at least the previous six months and met the
criteria set forth for moderate-intensity exercise in accordance with
American College of Sports Medicine Guidelines (American College of
Sports Medicine, 2018). In the rst experiment, participants were
immunized with a monovalent vaccine of a novel strain (single antigen
A/California/7/09 Inuenza H1N1, pdm09), hereafter referred to as
“monovalent”. In a second experiment, participants were vaccinated
with the trivalent inuenza vaccine ((Inuenza A/California/7/09, A/
Perth /16/2009, B/Brisbane/60/2008), termed “seasonal” vaccine. All
participants received the current inuenza Vaccine Information State-
ment and were asked to report any concerning side effects to study
personnel. The Institutional Review Board approved all procedures at
Iowa State University.
2.2. COVID-19 vaccine research participants
A total of 36 individuals that received the Pzer BNT162b2 (Pzer-
BioNTech COVID-19 Vaccine) between March 2021-June 2021 were
enrolled in the study (see Supplement Fig. 3 CONSORT diagram). Data
from eight participants who were possibly previously infected with
SARS-CoV-2 based on higher pre-immunization antibody levels and a
larger magnitude of change in response to the rst vaccine was reported
but not considered part of the primary analysis. Participants were
included in the study if they regularly participated in moderate or
vigorous-intensity exercise two or more times a week, with, on average,
at least one session lasting 50 min or longer. Participation in the study
followed the American College of Sports Medicine recommendations for
preparticipation health screening (Riebe, et al. 2015). Individuals were
excluded if they had an immune condition that would be expected to
impact the variables of interest or if they were taking a medication that
signicantly alters immune response. Individuals who were not preg-
nant, who planned to receive the COVID-19 vaccine, and who were
willing to donate blood were included in the study.
2.3. Psychosocial surveys for inuenza experiments
All participants in the inuenza vaccine experiments completed
several psychosocial surveys to determine whether there was an asso-
ciation between antibody response to the vaccine and psychosocial
factors. All participants completed the following surveys: Perceived
Stress Scale (PSS) (Cohen, et al. 1983), Sense of Coherence (SOC)
(Antonovsky, 1993), and Prole of Mood States (POMS) (McNair, et al.
1992).
2.4. Blood Collection, Vaccination, and timeline
Blood was taken from an antecubital vein (30 ml) in subjects just
before immunization with either the monovalent or seasonal inuenza
vaccine. Blood was collected at two weeks and four weeks post-
immunization.
A pre-immunization blood sample was collected within the week
preceding the COVID-19 vaccination. After the rst Pzer BioNTech
COVID-19 vaccine was administered, subjects returned two weeks later
J. Hallam et al.
Brain Behavior and Immunity 102 (2022) 1–10
3
for blood collection. The second dose of vaccine was given three weeks
after the rst vaccine dose. An additional blood sample was collected
one week following the second Pzer BioNTech COVID-19 vaccine (1
week post dose 2). All participants received side effect report forms
listing the side effects described on the Pzer BioNTech emergency use
authorization. Side effects were recorded every 24 h for the rst 72 h
after each vaccine by placing a check next to each side effect if present. A
score of 1 was given each day a given side effect was recorded by the
participant. Therefore, scores ranged from 0 (not present on any day) to
3 (present on each of the three days following the vaccine) for each side
effect listed on the emergency use authorization form.
2.5. Exercise conditions
In the experiment in which monovalent H1N1 vaccine was admin-
istered, subjects were randomly assigned to a light- to moderate-
intensity exercise group (90 min) or no exercise control group. If ran-
domized to exercise, participants began the exercise session within 30
min of receiving the single antigen vaccine. If assigned to control, the
participants started a sedentary period within 30 min of vaccination.
The sedentary period consisted of sitting while watching videos for 90
min. Adults randomized to exercise performed 90 min of exercise on a
cycle ergometer at 60–70% of estimated maximal heart rate (HR max)
with the range calculated as [(220 – age) ×0.6] to [(220-age) ×0.7],
typically corresponding to an intensity perceived as light to somewhat
hard based on the Rating of Perceived Exertion (RPE) Borg 6–20 scale
(Borg, 1982). After a 10-minute warm-up period, the cycling rate and
workload were adjusted to maintain heart rate in the appropriate range.
Heart rate and RPE (Borg 6–20 scale) were assessed every 10 min, and
water was available for the participant.
In the experiment in which seasonal vaccine was administered,
young (age 18–33) subjects were randomly assigned to one of three
groups: no exercise, 45 min of exercise, or 90 min of exercise, whereas
aged subjects (age 62–87) were randomly assigned to one of two groups:
no exercise, or 45 min of exercise. The exercise or rest period
commenced within 30 min after receiving the seasonal inuenza vac-
cine. The same exercise intensity was adhered to throughout exercise
(60–70% of age-estimated HR max on a cycle ergometer). Heart rate and
RPE were assessed every 10 min. Participants assigned to the no exercise
condition remained sedentary for 90 min post-vaccine, seated watching
videos.
In the experiments involving the COVID-19 vaccine, participants
were randomly assigned to either 90 min of exercise or instructed to go
about their daily routine but avoid exercise on the day of the rst
vaccination. All participants were asked to avoid exercise on the day the
second vaccine dose was given. Exercise took place outdoors at the
location where participants were vaccinated to limit SARS-CoV-2
infection risk for participants and study personnel. Study personnel
identied a safe walking/jogging route, and routes were designed to
monitor heart rate and RPE approximately every 10 min. An exercise
heart rate zone of approximately 120–140 beats per minute was targeted
as this range was consistent with the average heart rate range performed
at the target zone of 60–70% of HR max in the inuenza vaccine studies.
As exercise took place outdoors to minimize risk of SARS-CoV-2 infec-
tion, it was not possible to precisely control heart rates as compared to
the inuenza vaccine studies in which a cycle ergometer was used to set
a workload. Therefore, although target heart rate was approximately
equal to the target zone of 60–70% of HR max as in the inuenza vaccine
experiments, there was slightly more variability due to terrain condi-
tions. The exercise intensity was also monitored by RPE and consistent
with the inuenza vaccine studies in which target RPE was light to
somewhat hard. RPE was also used to adjust exercise intensity for any
participants treated with beta-blocker medications.
2.6. Serum antibody Detection assays (Human Sera)
After collection, blood sat at room temperature for 45 min until
clotted, was then centrifuged at 180 ×g, and was frozen at −80 C
o
until
subsequent measurement of anti-inuenza antibodies by ELISA. Samples
from the same individual for all time points were run together on the
same plate to minimize variability. Briey, for A/California/7/09, plates
were coated with 1 µg/ml hemagglutinin peptide in carbonate coating
buffer, followed by overnight incubation. For A/Perth/16/2009, plates
were coated with 256 hemagglutination (HA) units/ml, followed by
overnight incubation. Plates were washed with phosphate-buffered sa-
line (PBS) containing 0.05% Tween 20 between each step. Dilutions of
serum were added to the wells in duplicate with an optimal dilution of
1:200 for IgG based on preliminary experiments in which the optimal
dilution yielding a difference from pre-immunization values was iden-
tied. After overnight incubation, AP-conjugated goat anti-human IgG
(Southern Biotech) was added at 1:100 dilution. Phosphatase substrate
(Sigma) was added, and optical density (OD) was assessed at 405 nm on
a Fluostar plate reader. Data is shown as OD or fold change in OD from
pre-immunization to post-immunization time points. The OD repre-
senting the level of detectability for each assay is indicated in the
appropriate gure legend. This endpoint was calculated by a dilution
series carried out in a subset of participants to determine the OD at the
titer at which subsequent dilutions no longer detected a change in OD.
The detectable limits were calculated separately for each vaccine anti-
gen and in the appropriate group of participants. For experiments
involving COVID-19 vaccination, antibody level was measured using
GenScript SARS-CoV-2 Spike S1-RBD IgG&IgM ELISA Detection Kit. The
manufacturer’s directions were followed. Briey, serum was diluted
1:100 and the HRP-conjugated mouse anti-human IgG Fc was added to
the plate to detect anti-RBD IgG antibody. Banked pre-pandemic serum
was used as a negative control.
2.7. Mice, exercise conditions, vaccination, and IFN
α
antibody treatment
In all experiments, male BALB/c mice at 10 to 12 weeks of age
(Charles River Labs) were used. First, we compared different durations
of moderate-intensity exercise to determine the extent to which exercise
duration altered antibody response to intramuscular injection of inu-
enza vaccine. These experiments were replicated to conrm 90 min as
the duration in which increased antibody to vaccination was observed.
Next, we tested the hypothesis that IFN
α
mediates exercise-induced
changes in serum antibody concentration by administering anti-IFN
α
antibody to a subset of mice. Vaccinated mice in both experiments
received 50
μ
L of Binary ethyleneimine inactivated A/PR/8/34 virus
(512 HA units) administered intramuscularly into the quadriceps. Saline
control mice received 50 µL of saline and did not exercise. All mice were
acclimated to the treadmill running for three days in the week preceding
experiments by exposing mice to gradually increasing speeds on a
motorized treadmill for 10–15 min. All studies were performed ac-
cording to Institutional Animal Care and Use Committee guidelines at
Iowa State University and within the guidelines set by the NIH for the
care and use of laboratory animals.
In the rst experiment, thirty-eight BALB/c mice were randomly
assigned to one of four treatment groups (8–10 mice per group): no
exercise, or moderate-intensity exercise for 45 min post-immunization,
90 min post-immunization, or 3 h post-immunization. In a subsequent
experiment, we tested the role of IFN
α
as a mechanism by which exercise
may inuence antibody response to vaccine. Mice were randomly
assigned to one of four groups (9–10 mice per group): no exercise, no
exercise +anti-IFN
α
antibody, 90 min moderate-intensity exercise, 90
min moderate-intensity exercise +anti-IFN
α
antibody. The anti-IFN
α
treatment consisted of daily injection with 20 µg/mouse of rat IgG1 anti-
mouse IFN
α
, clone RMMA-1 (PBL Interferon Source) diluted in saline
containing 0.01% bovine serum albumin (BSA). Antibody administra-
tion began the day before vaccination (day −1) and continued until day
J. Hallam et al.
Brain Behavior and Immunity 102 (2022) 1–10
4
three post-immunization. Mice that did not receive anti-IFN
α
treatment
were injected with 20 µg/mouse of an irrelevant antibody, rat IgG1, at
the same dose and time of day as anti-IFN
α
treated mice. The 90-minute
exercise duration was selected based upon results from the initial ex-
periments, which demonstrated this exercise duration resulted in
increased antibody response to vaccine.
Mice began their respective exercise protocols within 15 min
following vaccination. Mice performed exercise on a treadmill at a speed
of 15 m/min, which has been shown to be moderate intensity (Fernando,
et al. 1993; Hoydal, et al. 2007). No-exercise mice were placed in
housing cages afxed to the top of the treadmill to mimic stressors
associated with the treadmill environment (noise, vibration) as closely
as possible. All mice were acclimated to the treadmill environment on
two separate occasions before immunization.
2.8. Serum antibody response to vaccine (mouse sera)
Blood was collected from all mice at two weeks post-immunization or
four weeks post-immunization. Blood was allowed to clot at room
temperature and then centrifuged at 180 ×g for 15 min. Serum was
collected and frozen at −80 ◦C until subsequent measurement of IgG,
IgG1, and IgG2a anti-inuenza antibodies by ELISA. Briey, plates were
coated overnight at 4◦C with inuenza virus A/PR/8/34 diluted in
carbonate coating buffer (pH 9.6) at a 200 HA units/ml concentration
for anti-inuenza IgG, IgG1, and IgG2a. The wells were blocked with
0.1% BSA solution at 37 ◦C for one hour. Plates were washed three times
with PBS containing 0.05% Tween 20 between each step. Diluted serum
(1:50 for IgG and 1:5 for IgG1 or IgG2a) was added to the wells and
incubated for 3 h at 37 ◦C. These dilutions were chosen based on pre-
liminary assay optimization (data not shown). After incubation, AP-
conjugated rat anti-mouse IgG, IgG1, or IgG2a were added (at 1:100
dilution for IgG and 1:10 for IgG1 and IgG2a), then plates were incu-
bated overnight at 4 ◦C. Finally, phosphatase substrate (4-Nitrophenyl
phosphate disodium salt hexahydrate, (Sigma) was added, and OD was
assessed at 405 nm at 30 min (IgG) and 50 min (IgG1, IgG2a) using a
Fluostar plate reader.
2.9. Statistical analysis
Statistics for all surveys, exercise data, and ELISA results in experi-
ments involving human participants were calculated using SPSS statis-
tical software (PASW/IBM Inc.) to perform ANOVAs. A mixed ANOVA
(exercise treatment * time) was used to compare serum anti-inuenza
immunoglobulins (IgG) to the different exercise durations post-
vaccination (either direct OD readings at respective serum dilutions or
fold change in OD relative to pre-immunization level, as indicated in
results). In experiments that included different age groups (seasonal
Inuenza vaccine), age and exercise were included in the model. In the
experiment with a wide range of ages (COVID-19 vaccine), initial ana-
lyses examined the effect of exercise only. A secondary analysis was
performed in which the top quartile for age was used to dene a middle-
aged population (ages 44–62) compared with participants considered
young adults (ages 18–43). Pearson correlations were used to compare
psychological survey outcomes and antibody response to the Inuenza
A/California/7/09 H1N1 antigen, combining the monovalent and sea-
sonal vaccine results. All data are reported as mean ±standard error of
the mean (SEM) unless otherwise indicated. In mouse experiments, a
one-way ANOVA was used to evaluate the effect of exercise duration on
antibody response followed by post-hoc analysis (Sidak). A two-way
ANOVA (exercise group and anti-IFN
α
antibody treatment group) was
used to compare antibody response in experiments to test the role of
IFN
α
as a mechanism by which exercise may impact antibody response.
Values of p <0.05 were considered statistically signicant, while values
of 0.05 <p <0.1 were considered a statistical trend.
3. Results
3.1. Demographics and response to exercise
Age, exercise heart rate, and RPE data are shown in Supplement
Table 2 (S2). There were no age differences between non-exercisers or
exercise treatment groups within each separate vaccine experiment.
Within the inuenza seasonal vaccine experiment, heart rate among
young participants was not different between the 45- or 90-minute ex-
ercise condition, but RPE in the 90-minute condition was slightly greater
than the 45-minute condition.
3.2. Antibody response to monovalent or seasonal inuenza vaccine,
effect of exercise, and association with psychosocial factors
The serum IgG response to monovalent H1N1 vaccine increased as
expected following vaccination (signicant main effect of time), as
shown in Fig. 1a. A signicant treatment by time interaction suggested
that the antibody response between the exercise group and no exercise
group responded differently over time with a greater response in the
exercise group (Fig. 1a). When results were calculated as fold change
relative to pre-immunization, a trend towards greater fold change in
antibody was observed in the exercise group relative to no exercise
(main effect of exercise, Fig. 1b). Individual antibody data for mono-
valent vaccine is also shown in Supplement Figure S4a-S4c. Antibody
response to seasonal inuenza vaccine was compared in young or aged
adults that exercised moderately for 45 min, 90 min (young only), or did
not exercise. The results show that anti-H1N1 serum IgG increased in
response to the vaccine as expected (signicant effect of time, Fig. 2a).
Over time, the change in antibody response differed between groups
(signicant time by exercise interaction), and the 90-minute exercise
condition resulted in greater serum antibody than no exercise. However,
antibody response in those assigned to the 45-minute exercise condition
was not different than no exercise in young or aged participants (Fig. 2a
and 2b, main effect of exercise, post-hoc analyses 90-min exercise >no-
ex). When data was calculated as fold change relative to pre-
immunization, the results were similar in that fold change of antibody
level in response to 90 min of exercise was signicantly greater than no-
exercise or 45 min of exercise (Fig. 2c), (main effect of exercise, 90-min
>no-exercise or 45-min ex in post-hoc analyses). There was no benet in
fold change in antibody level for the 45-minute exercise condition
relative to no exercise for either young or aged adults. The results con-
cerning anti-H3N2 antibody response as shown in Fig. 2d – 2f are similar
to H1N1. Again, 90 min of exercise treatment resulted in greater anti-
body level (main effect of exercise, Fig. 2d), and as expected, antibody
increased in response to vaccination (main effect of time, Fig. 2d and
2e). The 45-minute exercise condition did not improve antibody
response in young or aged participants, and there was no effect of age on
antibody response. Individual data for the response to seasonal vaccine
are shown in Supplement Figures S5a-S5h.
Given that the same Inuenza A/California/7/09 H1N1 antigen was
present in the monovalent and seasonal vaccine experiments, we com-
bined the psychosocial survey data from these two experiments to
analyze associations between antibody and psychosocial factors. A
negative association between antibody titer and perceived stress and a
similar negative association between antibody level and total mood
disturbance were noted, but the correlations did not meet statistical
signicance (Supplement Figures 8a-8c). A positive association between
the Sense of Coherence score and antibody was observed, but again this
relationship did not meet statistical signicance.
3.3. Antibody response to Pzer BioNTech COVID-19 vaccine and effect
of exercise
Antibody to SARS-CoV-2 was not measured before enrollment.
Therefore, we could not determine which participants may have
J. Hallam et al.
Brain Behavior and Immunity 102 (2022) 1–10
5
experienced asymptomatic infection before immunization or experi-
enced an infection with COVID-19-like symptoms but did not have a
conrmed positive COVID antigen test. Upon analysis of antibody level
pre- and post-immunization, we noted that 8 of 36 participants were
possibly infected based upon >1.5 fold greater pre-immunization OD
value as compared those assumed to be not previously infected, and a
signicant increase in antibody level after the rst vaccine nearly
equivalent to the antibody level observed after the second dose in those
assumed to be not previously infected. Based on these criteria, the
antibody level of “possibly infected” participants was signicantly
higher at the pre-immunization time point and two weeks after the
initial vaccine (Fig. 3a). The nding that antibody titer in response to the
rst dose of vaccine is typically greater in previously infected as
compared to naïve individuals is a common nding in the literature,
whereas differences in antibody response between previously infected
relative to naïve may or may not be present in response to the second
Fig. 1. Serum IgG response to monovalent inuenza H1N1 vaccine. 1a. Serum anti-inuenza IgG shown as optical density at 405 nm by ELISA assay from serum samples
collected pre-immunization, two, or four-weeks post. Serum dilution series indicate assay level of detectability at OD =203. A signicant main effect of time (**p =
0.001, F =15.07), and a signicant time by treatment interaction between the no-exercise and 90-minute exercise condition (*p =0.04, F =3.68, df =15,
η
2
partial
=0.235) were found. 1b. Calculated fold change in antibody response, with a trend to main effect of exercise treatment (+p =0.059, F =4.36, df =15,
η
2
partial
=0.255).
Fig. 2. Serum IgG response to seasonal inuenza vaccine. 2a & 2b. Serum antibody level assessed as optical density at 405 nm by ELISA. Serum dilution series indicate
assay level of detectability for H1N1 at OD =224. A signicant main effect of time was observed as antibody increased in response to vaccination (main effect of
time, p <0.0001, F =111.59). 90-minute exercise increased antibody response (time * exercise interaction, p <0.001, F =10.5, df =24,
η
2
partial =0.489; main
effect of exercise, p =0.027, F =4.379, df =24,
η
2
partial =0.277; post-hoc analysis (Sidak)) with no difference between no-ex and 45-min ex or 90-min ex and 45-
min ex). No signicant effect of age. 2c. Fold change in antibody level calculated relative to pre-immunization, * indicates greater fold change in 90-min exercise
condition than either no-ex or 45-min exercise (main effect of exercise, p =0.005, F =7.13, df =24,
η
2
partial =0.387, post hoc analysis (Sidak) with no difference
between no-ex and 45-min ex, and 90-min ex >45-min ex (p =0.001), 90-min ex >no-ex (p =0.005). 2d & 2e. Serum antibody level assessed as optical density at
405 nm by ELISA. Serum dilution series indicate assay level of detectability for H3N2 at OD =219. In the overall model, a signicant main effect of time as antibody
increased in response to vaccination (main effect of time, p <0.001, F =31.7, df =25). There was a signicant time by exercise interactions (p =0.021, F =3.2, df =
25,
η
2
partial =0.235, and the 90-minute exercise was associated with increased antibody response (* main effect of exercise, p =0.014, F =5.217, df =25,
η
2
partial
=0.332) post-hoc analysis (Sidak) with no difference between no-ex and 45-min ex, but 90-min was signicantly greater than no ex (p =0.015) and also>45-min ex,
(p =0.04). 2f. Fold change in antibody titer calculated relative to pre-immunization. A trend toward main effect of exercise was noted (p =0.07, F =2.93, df =25,
η
2
partial =0.219).
J. Hallam et al.
Brain Behavior and Immunity 102 (2022) 1–10
6
vaccine (Ali et al., 2021, Ebinger, et al. 2021, Fraley, et al., 2021, Geers,
et al., 2021, Gobbi et al., 2021, Goel, et al., 2021, Krammer et al., 2021).
However, the presence of antibody to nucleocapsid protein can further
differentiate these populations, which was not measured in this study,
and this is a limitation in the method we used to dene “possibly
infected.” As the initial vaccination likely served as a second dose for
possibly infected participants, these participants were removed from the
primary data analysis because the fold change in antibody response
differed signicantly from participants who were assumed to be naive.
The serum anti-RBD IgG antibody level assessed over time shows a
signicant time by treatment interaction (Fig. 3b) with a greater in-
crease in the exercise group over time. When results are expressed as
fold change in antibody relative to pre-immunization, exercise partici-
pants have signicantly greater antibody (Fig. 3c). Individual data as
fold change or OD are shown in Supplement Figures S6a-S6c. We also
observed a signicant negative correlation between age and antibody
level at two weeks following the initial immunization and one week
following the second vaccine dose when antibody was measured either
as OD or fold change. The Pearson correlation between age and OD two
weeks post-immunization was 0.534, p =0.006; at one-week post-dose
2, the Pearson correlation between age and OD was 0.603, p =0.002.
Given the effect of age, the results of a secondary analysis in which age
group was included as a factor (young adult as 18–43, middle-aged as
44–64) showed a signicant interaction between age, exercise, and
antibody level over time (pre, 2 wk post, 1 wk post-dose 2; p =0.008, F
=5.38; graph not shown). When antibody was assessed as fold change
relative to pre-immunization, a similar pattern was observed with a
trend towards a signicant interaction between age, exercise, and fold
change in antibody (2 wk post, 1 wk post-dose 2; Fig. 3d). Middle-aged
participants had signicantly lower fold changes in antibody response
compared to young (Fig. 3d). Participants recorded side effects for three
days after receiving each vaccine. The side effect scores were not
signicantly different between 90-minute exercise participants
compared to no exercise (Supplement, Table S3). Reported side effects
on a percentage basis are also shown (Supplement, Table S4). Initial
ndings for the participants that were in the “possibly infected” category
included only eight subjects, but this data is reported (n =5 no exercise,
n =3 exercise) (Supplement Figure S9).
Fig. 3. Serum anti-RBD IgG response to Pzer BioNTech COVID-19 vaccine 3a. SARS-CoV-2 Spike S1-RBD IgG from serum of participants categorized as “possibly
infected” prior to immunization was signicantly greater at pre-immunization and at 2 wk after the initial vaccine, but not different at 1 wk post-dose 2 (main effect
of time, p <0.001, F =61.2, main effect of group, p <0.001, F =19.1, time by group interaction, p , F =19.6; and follow up one-way ANOVA to identify time points
of difference, * indicates difference at pre-immunization p <0.001, F =37.8; and * indicates difference at 2 wk post p <, 0.001, F =46.2). 3b. SARS-CoV-2 Spike S1-
RBD IgG as optical density (OD) in serum collected pre-immunization, 2 wk after initial vaccine or 1 wk after second vaccine dose (the second vaccine was
administered 3 weeks after the rst vaccine dose). * indicates signicant treatment group by time interaction (p =0.039, F =3.50, df =25,
η
2
partial =0.137). A
main effect of time (p <0.001, F =473.6) was observed and a trend to main effect of treatment (p =0.055, F =4.075, df =25,
η
2
partial =0.149). 3c. Fold change in
SARS-CoV-2 Spike S1-RBD IgG relative to pre-immunization, (90 m exercise >no exercise *p =0.048, F =4.37, df =25,
η
2
partial =0.160; 1wk post dose 2 >2wk
post initial vaccine, p <0.001, F =200.4 df =25). 3d. Fold change in SARS-CoV-2 Spike S1-RBD IgG relative to pre-immunization as compared by exercise, age, and
time post-immunization. Antibody response in middle aged adults <young adults (main effect of age, p =0.003, F =11.3, df =25,
η
2
partial =0.350) and exercise
treatment trended towards a difference based on age (exercise by age ×antibody interaction, p =0.057, F =4.06 df =25).
J. Hallam et al.
Brain Behavior and Immunity 102 (2022) 1–10
7
3.4. Antibody response to inuenza A/PR/8/34 inactivated vaccine and
effect of exercise duration in mouse model
A rodent model of inuenza immunization was used to examine the
effect of varying durations of exercise on antibody response to the
vaccine. The results showed no effect of exercise on antibody level
measured at two weeks post-immunization, but at four weeks after im-
munization, mice exercising for 90 min had signicantly greater serum
antibody level than no exercise or 180 min of exercise (Fig. 4a and in-
dividual data shown in Supplement Figure S7a). The IgG1 subclass of
antibody was measured at four weeks post-immunization (Fig. 4b), and
results showed a similar pattern to total IgG, although the overall effect
of exercise met the criteria for a statistical trend (p =0.09).
3.5. Requirement for IFN
α
at time of immunization for optimal antibody
response
Separate experiments examined the potential role of IFN
α
as a
mechanism that may contribute to the effects of exercise on antibody
response. Mice treated with anti-IFN
α
antibody or control antibody at
Fig. 4. Mouse Serum anti-Inuenza A/PR/8/34 IgG, IgG1, or IgG2a response to vaccination following varying durations of exercise, and the role of IFN
α
(mice). 4a. Serum
was collected at either 2 weeks or 4 weeks post-immunization (separate mice at 2 weeks and 4 weeks). No signicant effect of exercise at 2 weeks. At 4 weeks of
exercise, exercise treatment altered antibody response (overall signicant effect of exercise, p =0.044, F =3.046, df =34,
η
2
partial =0.233) and post hoc analysis
results indicated serum IgG in mice completing 90 min of exercise >no ex (*p =0.004). 4b. Serum anti-inuenza IgG1 assessed at 4 weeks post-immunization. There
was a statistical trend (overall effect of exercise, p =0.09, F =2.92, df =34,
η
2
partial =0.168) towards a difference between groups. 4c. Total anti-inuenza serum
IgG was measured four-weeks post-immunization. Anti-IFN
α
antibody-treated mice had signicantly less IgG but exercise increased IgG in both anti-IFN
α
treated
mice and mice treated with irrelevant antibody (main effect of antibody, * p <0.001, F =28.3, df =38,
η
2
partial =0.447 ; main effect of exercise, p =0.002, F =
11.5, df =38,
η
2
partial =0.247). There was no signicant exercise group * anti-IFN
α
interaction, p =0.128). 4d & 4e. Anti-inuenza serum IgG1 (6b) or IgG2a (6c)
were measured at four weeks post-immunization. Without IFN
α
, antibody response was signicantly reduced for IgG1 (main effect of IFN
α
antibody treatment, p <
0.001, F =24.5, df =38,
η
2
partial =0.447) and exercise treatment signicantly increased IgG1 (p <0.001, F =23.5, df =38,
η
2
partial =0.402), and there was a
trend towards an anti-IFN
α
antibody by exercise treatment interaction (p =0.06). With respect to IgG2a there was a signicant interaction between exercise
treatment and anti-IFN
α
antibody (exercise * antibody interaction, p =0.014, F =6.63, df =38,
η
2
partial =0.159) implying the exercise effect was not consistent
across antibody treatment. For IgG2a, effect of exercise (p <0.001, F =23.73, df =38,
η
2
partial =0.404) and IFN
α
antibody treatment, p <0.001, F =31.6, df =38,
η
2
partial =0.475). 4f. The signicant interaction for IgG2a (from 4f) is shown in a line graph format.
J. Hallam et al.
Brain Behavior and Immunity 102 (2022) 1–10
8
the time of immunization completed either 90 min of exercise or no
exercise. The results of serum anti-inuenza IgG antibody, IgG1, and
IgG2a were similar in that a lack of IFN
α
at the time of immunization
signicantly impaired antibody response (Fig. 4c −4f). Mice assigned to
the 90-minute exercise intervention had higher IgG, IgG1, and IgG2a
than no exercise mice, regardless of whether mice received anti-IFN
α
antibody treatment. However, there was a signicant interaction with
respect to IgG2a in which the anti-IFN
α
antibody treatment attenuated
the effect of exercise (signicant exercise by IFN
α
antibody treatment
interaction, Fig. 4e and 4f), suggesting that IFN
α
may contribute to some
extent to the effect of exercise on antibody class switching. Individual
mouse data is shown in Supplement Figures S7b-S7d.
4. Discussion
The ndings presented here demonstrate that 90 min of exercise
after immunization increases antibody response several weeks later
across several immunization models. To the best of our knowledge, these
ndings are the rst to show that light- to moderate-intensity long-
duration exercise enhances antibody response across several vaccine
formulations, including COVID-19 vaccination. These results may have
immediately translatable public health relevance as the exercise para-
digm is straightforward to implement and does not require special
equipment. The exercise intervention is feasible for people who exercise
regularly at light intensities such as walking, and persons with a range of
health characteristics were able to complete the exercise. For example,
nearly half of the participants in the COVID-19 vaccination trial had a
BMI in the overweight or obese category, and the distance covered in 90
min ranged from approximately four miles to over 10 miles, represent-
ing a variety of tness levels as heart rate and relative perceived exertion
level were maintained within a constant range. It will be essential to
determine the length of time post-vaccination for which an exercise-
associated increase in serum antibody may be present. Longer term
antibody response will be assessed as these ndings are an early report.
It would also be helpful to dene how the change in antibody translates
to protection from infection. However, appropriate study designs to
address that question typically require thousands of participants, and
may not be feasible, in which case inference from antibody level and
protection studies will be necessary. Serum antibody is recognized as an
immune correlate of protection for inuenza (Potter and Oxford, 1979),
including IgG measured by ELISA (Trombetta, et al. 2018), and therefore
we expect that the increase in antibody that we report would confer
some benet. The immune correlates of protection from COVID-19
vaccination are under investigation, but early evidence supports
serum antibody as a potential correlate (Sadarangani, et al. 2021).
In comparing the results from our studies with the existing literature
on exercise and vaccines, one limitation that arises is the wide variety of
exercise approaches used. To our knowledge, there are no other studies
that investigated the immunomodulatory effects of a 90-minute aerobic
exercise session. However, of the studies that examined aerobic exercise
rather than eccentric exercise, there were conicting results. For
example, one study showed no benet in antibody response to either
pneumococcal or inuenza vaccine after 45 min of aerobic exercise
(Long, et al. 2012). In other studies, mixed results were found in
response to 45 min of dynamic exercise or aerobic exercise, with an
increase in antibody response to one antigenic component of inuenza
vaccine, but not both inuenza A antigens, and variable results by sex
were present (Edwards, et al. 2006; Ranadive, et al. 2014). Although the
timing of exercise in relation to vaccination was different in our ex-
periments with exercise performed after immunization, we observed
that 45 min of exercise was of insufcient duration to increase antibody
response to either inuenza A antigen (in young or aged adults) or in a
mouse model of inuenza A vaccination. Therefore, our ndings align
with the current literature reporting that aerobic exercise interventions
of 45 min or less do not enhance the antibody response following
inuenza A vaccination.
Notably, there were no differences in the total number of side effects
or duration of side effects in exercise compared to no exercise subjects in
response to the vaccine with greater reactogenicity (COVID-19), indi-
cating a potential benet of exercise with no change in side effect pro-
le. No adverse reactions to either of the inuenza vaccines were
reported. Two separate studies found decreased side effects of exercisers
versus non-exercisers (Bohn-Goldbaum, et al. 2020; Lee, et al. 2018). We
did not see a similar exercise-associated reduction of vaccine side ef-
fects, but exercise mode and duration differed, and COVID-19 vaccine
reactogenicity may also differ, thereby limiting the ability to make
direct comparisons.
The mechanisms by which exercise may increase antibody response
to vaccines remain to be elucidated, although our results provide some
initial support for a potential role of IFN
α
based on ndings from the
mouse model. The exercise-induced enhancement of IgG2a was atten-
uated in mice that received anti-IFN
α
antibody treatment, and a similar
trend was apparent for IgG1. Type I IFN promotes class switching (Le
Bon, et al. 2001), supported by our data showing that mice lacking IFN
α
at the time of immunization had signicantly reduced IgG1 and IgG2a.
IFN
α
also promotes dendritic cell costimulatory molecule expression
and dendritic cell activation (Montoya, et al. 2002; Tough, 2004),
germinal center formation (Le Bon, et al. 2001), and may have direct or
indirect stimulatory effects on B cells and T cells (Le Bon, et al. 2006).
Type I IFN, inducers of Type I IFN, or IFN
α
alone have adjuvant prop-
erties as demonstrated in an inuenza model (Junkins, et al. 2018;
Proietti, et al. 2002), may specically stimulate IgG2c or IgA antibody in
response to inuenza vaccine (Ye, et al. 2019), and may serve as a
mucosal adjuvant in porcine inuenza vaccination (Liu, et al. 2019).
Inducers of IFN
α
(imiquimod) delivered intradermally with inuenza
vaccine improved antibody response in young or older adults (Hung,
et al. 2016; Hung, et al. 2014). Altogether, the evidence from these
studies suggests that IFN
α
can have immunostimulatory effects and
adjuvant activity for vaccines. Therefore, if exercise results in signicant
increases in IFN
α
, or the ability to produce greater IFN
α
upon stimula-
tion, IFN
α
may be one mechanism by which exercise improves antibody
response to vaccines. With 90 min of exercise, we had previously
observed a signicant increase in IFN
α
by human plasmacytoid den-
dritic cells (Table S1). We acknowledge that 90 min of exercise in a
mouse is not directly translatable to humans, but the mouse model af-
fords the opportunity to begin identifying potential mechanisms.
Considering other possible mechanisms, the lack of response to 45
min of aerobic exercise compared with enhanced antibody response to
immunization with 90 min of exercise may provide some insight. Ex-
ercise duration and intensity inuence the metabolic and neuroendo-
crine responses to exercise. Multiple aspects of the vaccine itself,
including the dose of antigen, the inclusion of antigens, the delivery
route, the vaccine platform (lipid nanoparticle encapsulated-mRNA
compared to subunit/split virus preparation), may impact the kinetics
and quality of antigen-presenting cell response and subsequent activa-
tion of T cell and B cell response (Bachmann and Jennings, 2010;
Chappell, et al. 2014; Liang, et al. 2017; Zeng, et al. 2020). These factors
may inuence the immunomodulatory effects of exercise. The ndings
presented here provide initial insights into mechanisms to further
explore in future studies. Our results suggest a possible partial role for
IFN
α
and show that exercise benets extend across different vaccine
formulations but require 90 min of light- to moderate-intensity exercise
instead of only 45 min. Our ndings also cannot rule out or conrm a
role for IL-6 as a potential mechanism by which exercise may contribute
to enhanced antibody response. One might expect a dose response such
that as exercise duration increases, IL-6 increases (Fischer, 2006). The
data presented in the mouse model of vaccination in which 45, 90, and
180 min of exercise were compared do not support a potential dos-
e–response effect of IL-6, but the experimental design did not specically
address IL-6.
A limitation of the ndings from these experiments is the relatively
small number of participants. However, the reproducibility of the results
J. Hallam et al.
Brain Behavior and Immunity 102 (2022) 1–10
9
across three different human vaccine experiments with varying vaccine
formulations (inactivated, mRNA-based) that contain either novel an-
tigens or antigens to which participants had previous exposure (seasonal
inuenza vaccine) lends credence to these results. Studies in a mouse
model replicated the ndings in human experiments. Although the
conclusions of the COVID-19 vaccination are an early report with rela-
tively limited sample size, analysis of long-term antibody response will
be performed. Due to the potential public health relevance, these early
ndings with COVID-19 were reported. Larger scale trials should be
undertaken that conrm these ndings and examine the extent to which
similar effects may occur after booster immunizations. From a public
health perspective, it would be worthwhile to determine whether an
exercise duration that falls between 45 and 90 min would confer some
benet as more adults may be able to complete an exercise session that is
less than 90 min. Another limitation of the ndings is that the mouse
experiments involved only male mice, and therefore it remains possible
that these ndings may not apply to female mice. The human studies
included males and females but with a limited number of participants, it
was not possible to establish whether there were signicant sex effects.
In future studies, it will be important to establish whether sex by exer-
cise interactions exist and inuence any exercise-associated changes in
the antibody response to vaccine.
In summary, our results are the rst to demonstrate an exercise-
induced enhancement of antibody response to COVID-19 immuniza-
tion without an increase in reported side effects. Our ndings also show
longer-duration light- to moderate-intensity exercise increases antibody
response across different vaccine formulations, and exercise-induced
alterations of IFN
α
may partially contribute to this effect.
Declaration of Competing Interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Acknowledgements
This research did not receive any specic grant from funding
agencies in the public, commercial, or not-for-prot sectors.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bbi.2022.02.005.
References
American College of Sports Medicine, 2018. ACSM’s Guidelines for Exercise Testing and
Prescription, Tenth edition. Wolters Kluwer, Philadelphia.
Ali, H., Alahmad, B., Al-Shammari, A.A., Alterki, A., Hammad, M., Cherian, P.,
Alkhairi, I., Sindhu, S., Thanaraj, T.A., Mohammad, A., Alghanim, G., Deverajan, S.,
Ahmad, R., El-Shazly, S., Dashti, A.A., Shehab, M., Al-Sabah, S., Alkandari, A.,
Abubaker, J., Abu-Farha, M., Al-Mulla, F., 2021. Previous COVID-19 Infection and
Antibody Levels After Vaccination. Front Public Health. (9), 778243 https://doi.org/
10.3389/fpubh.2021.778243. Dec 1 PMID: 34926392; PMCID: PMC8671167.
Antonovsky, A., 1993. The structure and properties of the Sense of Coherence Scale. Soc
Sci Med. 36 (6), 725–733. https://doi.org/10.1016/0277-9536(93)90033-z.
Bachmann, M.F., Jennings, G.T., 2010. Vaccine delivery: a matter of size, geometry,
kinetics and molecular patterns. Nat Rev Immunol. 10 (11), 787–796. https://doi.
org/10.1038/nri2868.
Bohn-Goldbaum, E., Lee, V.Y., Skinner, S.R., Frazer, I.H., Khan, B.A., Booy, R.,
Edwards, K.M., 2019. Acute exercise does not improve immune response to HPV
vaccination series in adolescents. Papillomavirus Res. 8, 1–3. https://doi.org/
10.1016/j.pvr.2019.100178.
Bohn-Goldbaum, E., Pascoe, A., Singh, M.F., Singh, N., Kok, J., Dwyer, D.E.,
Mathieson, E., Booy, R., Edwards, K.M., 2020. Acute exercise decreases vaccine
reactions following inuenza vaccination among older adults. Brain Behav. Immun. -
Health 1, 1–7. https://doi.org/10.1016/j.bbih.2019.100009.
Borg, G.A., 1982. Psychophysical bases of perceived exertion. Med Sci Sports Exerc. 14
(5), 377–381. https://doi.org/10.1055/s-2007-1021244.
Bruunsgaard, H., Hartkopp, A., Mohr, T., Konradsen, H., Heron, I., Mordhorst, C.H.,
Pedersen, B.K., 1997. In vivo cell-mediated immunity and vaccination response
following prolonged, intense exercise. Med Sci Sports Exerc. 29 (9), 1176–1181.
https://doi.org/10.1097/00005768-199709000-00009.
Campbell, J.P., Edwards, K.M., Ring, C., Drayson, M.T., Bosch, J.A., Inskip, A., Long, J.E.,
Pulsford, D., Burns, V.E., 2010. The effects of vaccine timing on the efcacy of an
acute eccentric exercise intervention on the immune response to an inuenza
vaccine in young adults. Brain Behav Immun. 24 (2), 236–242. https://doi.org/
10.1016/j.bbi.2009.10.001.
Chappell, C.P., Giltiay, N.V., Dresch, C., Clark, E.A., 2014. Controlling immune responses
by targeting antigens to dendritic cell subsets and B cells. Int Immunol. 26 (1), 3–11.
https://doi.org/10.1093/intimm/dxt059.
Choi, Y.S., Eto, D., Yang, J.A., Lao, C., Crotty, S., 2013. Cutting edge: STAT1 is required
for IL-6-mediated Bcl6 induction for early follicular helper cell differentiation.
J Immunol. 190 (7), 3049–3053. https://doi.org/10.4049/jimmunol.1203032.
Cohen, S., Kamarck, T., Mermelstein, R., 1983. A global measure of perceived stress.
J Health Soc Behav. 24 (4), 385–396. https://doi.org/10.2307/2136404.
Dienz, O., Eaton, S.M., Bond, J.P., Neveu, W., Moquin, D., Noubade, R., Briso, E.M.,
Charland, C., Leonard, W.J., Ciliberto, G., Teuscher, C., Haynes, L., Rincon, M.,
2009. The induction of antibody production by IL-6 is indirectly mediated by IL-21
produced by CD4+T cells. J Exp Med. 206 (1), 69–78. https://doi.org/10.1084/
jem.20081571.
Ebinger, J.E., Fert-Bober, J., Printsev, I., Wu, M., Sun, N., Prostko, J.C., Frias, E.C.,
Stewart, J.L., Van Eyk, J.E., Braun, J.G., Cheng, S., Sobhani, K., 2021. Antibody
responses to the BNT162b2 mRNA vaccine in individuals previously infected with
SARS-CoV-2. Nat Med. 27 (6), 981–984. https://doi.org/10.1038/s41591-021-
01325-6.
Edwards, K.M., Burns, V.E., Reynolds, T., Carroll, D., Drayson, M., Ring, C., 2006. Acute
stress exposure prior to inuenza vaccination enhances antibody response in women.
Brain Behav Immun. 20 (2), 159–168. https://doi.org/10.1016/j.bbi.2005.07.001.
Edwards, K.M., Burns, V.E., Allen, L.M., McPhee, J.S., Bosch, J.A., Carroll, D.,
Drayson, M., Ring, C., 2007. Eccentric exercise as an adjuvant to inuenza
vaccination in humans. Brain Behav Immun. 21 (2), 209–217. https://doi.org/
10.1016/j.bbi.2006.04.158.
Edwards, K.M., Burns, V.E., Carroll, D., Drayson, M., Ring, C., 2007 Jul. The acute stress-
induced immunoenhancement hypothesis. Exerc Sport Sci Rev. 35 (3), 150–155.
Edwards, K.M., Burns, V.E., Adkins, A.E., Carroll, D., Drayson, M., Ring, C., 2008.
Meningococcal A vaccination response is enhanced by acute stress in men.
Psychosom Med. 70 (2), 147–151. https://doi.org/10.1097/
PSY.0b013e318164232e.
Edwards, K.M., Pung, M.A., Tomfohr, L.M., Ziegler, M.G., Campbell, J.P., Drayson, M.T.,
Mills, P.J., 2012. Acute exercise enhancement of pneumococcal vaccination
response: a randomised controlled trial of weaker and stronger immune response.
Vaccine. 30 (45), 6389–6395. https://doi.org/10.1016/j.vaccine.2012.08.022.
Eto, D., Lao, C., DiToro, D., Barnett, B., Escobar, T.C., Kageyama, R., Yusuf, I., Crotty, S.,
Poh, L.N.F., 2011. IL-21 and IL-6 are critical for different aspects of B cell immunity
and redundantly induce optimal follicular helper CD4 T cell (Tfh) differentiation.
PLoS One. 6 (3), e17739. https://doi.org/10.1371/journal.pone.0017739.
Fernando, P., Bonen, A., Hoffman-Goetz, L., 1993. Predicting submaximal oxygen
consumption during treadmill running in mice. Can J Physiol Pharmacol. 71
(10–11), 854–857. https://doi.org/10.1139/y93-128.
Fischer, C.P., 2006. Interleukin-6 in acute exercise and training: what is the biological
relevance? Exerc Immunol Rev. 12, 6–33. PMID: 17201070.
Fraley E, LeMaster C, Geanes E, et al. Humoral immune responses during SARS-CoV-2
mRNA vaccine administration in seropositive and seronegative individuals. BMC
Med. 2021;19(1):169. Published 2021 Jul 26. doi:10.1186/s12916-021-02055-.
Geers, D., Shamier, M.C., Bogers, S., den Hartog, G., Gommers, L., Nieuwkoop, N.N.,
Schmitz, K.S., Rijsbergen, L.C., van Osch, J.A.T., Dijkhuizen, E., Smits, G.,
Comvalius, A., van Mourik, D., Caniels, T.G., van Gils, M.J., Sanders, R.W., Oude
Munnink, B.B., Molenkamp, R., de Jager, H.J., Haagmans, B.L., de Swart, R.L.,
Koopmans, M.P.G., van Binnendijk, R.S., de Vries, R.D., GeurtsvanKessel, C.H., 2021.
SARS-CoV-2 variants of concern partially escape humoral but not T-cell responses in
COVID-19 convalescent donors and vaccinees. Sci Immunol. 6 (59) https://doi.org/
10.1126/sciimmunol.abj1750 eabj1750. PMID: 34035118.
Gobbi, F., Buonfrate, D., Moro, L., Rodari, P., Piubelli, C., Caldrer, S., Riccetti, S.,
Sinigaglia, A., Barzon, L., 2021. Antibody Response to the BNT162b2 mRNA COVID-
19 Vaccine in Subjects with Prior SARS-CoV-2 Infection. Viruses. 13 (3), 422.
https://doi.org/10.3390/v13030422. PMID: 33807957; PMCID: PMC8001674.
Goel, R.R., Apostolidis, S.A., Painter, M.M., Mathew, D., Pattekar, A., Kuthuru, O.,
Gouma, S., Hicks, P., Meng, W., Rosenfeld, A.M., Dysinger, S., Lundgreen, K.A., Kuri-
Cervantes, L., Adamski, S., Hicks, A., Korte, S., Oldridge, D.A., Baxter, A.E., Giles, J.
R., Weirick, M.E., McAllister, C.M., Dougherty, J., Long, S., D’Andrea, K.,
Hamilton, J.T., Betts, M.R., Luning Prak, E.T., Bates, P., Hensley, S.E., Greenplate, A.
R., Wherry, E.J., 2021. Distinct antibody and memory B cell responses in SARS-CoV-
2 naïve and recovered individuals following mRNA vaccination. eabi6950 Sci
Immunol. 6 (58). https://doi.org/10.1126/sciimmunol.abi6950.
Hoydal, M.A., Wisloff, U., Kemi, O.J., Ellingsen, O., 2007. Running speed and maximal
oxygen uptake in rats and mice: practical implications for exercise training. Eur J
Cardiovasc Prev Rehabil. 14 (6), 753–760. https://doi.org/10.1097/
HJR.0b013e3281eacef1.
Hung, I.F., Zhang, A.J., To, K.K., Chan, J.F., Li, C., Zhu, H.S., Li, P., Li, C., Chan, T.C.,
Cheng, V.C., Chan, K.H., Yuen, K.Y., 2014. Immunogenicity of intradermal trivalent
inuenza vaccine with topical imiquimod: a double blind randomized controlled
trial. Clin Infect Dis. 59 (9), 1246–1255. https://doi.org/10.1093/cid/ciu582.
Hung, I.F., Zhang, A.J., To, K.K., Chan, J.F., Li, P., Wong, T.L., Zhang, R., Chan, T.C.,
Chan, B.C., Wai, H.H., Chan, L.W., Fong, H.P., Hui, R.K., Kong, K.L., Leung, A.C.,
Ngan, A.H., Tsang, L.W., Yeung, A.P., Yiu, G.C., Yung, W., Lau, J.Y., Chen, H.,
Chan, K.H., Yuen, K.Y., 2016. Topical imiquimod before intradermal trivalent
J. Hallam et al.
Brain Behavior and Immunity 102 (2022) 1–10
10
inuenza vaccine for protection against heterologous non-vaccine and antigenically
drifted viruses: a single-centre, double-blind, randomised, controlled phase 2b/3
trial. Lancet Infect Dis. 16 (2), 209–218. https://doi.org/10.1016/S1473-3099(15)
00354-0.
Junkins, R.D., Gallovic, M.D., Johnson, B.M., Collier, M.A., Watkins-Schulz, R.,
Cheng, N., David, C.N., McGee, C.E., Sempowski, G.D., Shterev, I., McKinnon, K.,
Bachelder, E.M., Ainslie, K.M., Ting, J.P., 2018. A robust microparticle platform for a
STING-targeted adjuvant that enhances both humoral and cellular immunity during
vaccination. J Control Release. 28 (270), 1–13. https://doi.org/10.1016/j.
jconrel.2017.11.030.
Karp, J.D., Smith, J., Hawk, K., 2000. Restraint stress augments antibody production in
cyclophosphamide-treated mice. Physiol Behav. 70 (3–4), 271–278. https://doi.org/
10.1016/s0031-9384(00)00267-5.
Krammer, F., Srivastava, K., Alshammary, H., Amoako, A.A., Awawda, M.H., Beach, K.F.,
Bermúdez-Gonz´
alez, M.C., Bielak, D.A., Carre˜
no, J.M., Chernet, R.L., Eaker, L.Q.,
Ferreri, E.D., Floda, D.L., Gleason, C.R., Hamburger, J.Z., Jiang, K., Kleiner, G.,
Jurczyszak, D., Matthews, J.C., Mendez, W.A., Nabeel, I., Mulder, L.C.F., Raskin, A.
J., Russo, K.T., Salimbangon, A.-B., Saksena, M., Shin, A.S., Singh, G., Sominsky, L.
A., Stadlbauer, D., Wajnberg, A., Simon, V., 2021. Antibody Responses in
Seropositive Persons after a Single Dose of SARS-CoV-2 mRNA Vaccine. N Engl J
Med 384 (14), 1372–1374.
Le Bon, A., Schiavoni, G., D’Agostino, G., Gresser, I., Belardelli, F., Tough, D.F., 2001.
Type i interferons potently enhance humoral immunity and can promote isotype
switching by stimulating dendritic cells in vivo. Immunity. 14 (4), 461–470. https://
doi.org/10.1016/s1074-7613(01)00126-1.
Le Bon, A., Thompson, C., Kamphuis, E., Durand, V., Rossmann, C., Kalinke, U.,
Tough, D.F., 2006. Cutting edge: enhancement of antibody responses through direct
stimulation of B and T cells by type I IFN. Journal of immunology) 176 (4),
2074–2078. https://doi.org/10.4049/jimmunol.176.4.2074.
Lee, V.Y., Booy, R., Skinner, S.R., Fong, J., Edwards, K.M., 2018. The effect of exercise on
local and systemic adverse reactions after vaccinations - Outcomes of two
randomized controlled trials. Vaccine. 36 (46), 6995–7002. https://doi.org/
10.1016/j.vaccine.2018.09.067.
Liang, F., Lindgren, G., Lin, A., Thompson, E.A., Ols, S., Rohss, J., John, S., Hassett, K.,
Yuzhakov, O., Bahl, K., Brito, L.A., Salter, H., Ciaramella, G., Lore, K., 2017. Efcient
Targeting and Activation of Antigen-Presenting Cells In Vivo after Modied mRNA
Vaccine Administration in Rhesus Macaques. Mol Ther. 25 (12), 2635–2647. https://
doi.org/10.1016/j.ymthe.2017.08.006.
Liu, L., Fan, W., Zhang, H., Zhang, S., Cui, L., Wang, M., Bai, X., Yang, W., Sun, L.,
Yang, L., Liu, W., Li, J., 2019. Interferon as a Mucosal Adjuvant for an Inuenza
Vaccine in Pigs. Virol Sin. 34 (3), 324–333. https://doi.org/10.1007/s12250-019-
00102-7.
Long, J.E., Ring, C., Drayson, M., Bosch, J., Campbell, J.P., Bhabra, J., Browne, D.,
Dawson, J., Harding, S., Lau, J., Burns, V.E., 2012. Vaccination response following
aerobic exercise: can a brisk walk enhance antibody response to pneumococcal and
inuenza vaccinations? Brain Behav Immun. 26 (4), 680–687. https://doi.org/
10.1016/j.bbi.2012.02.004.
McNair, D.M., Lorr, M., Droppleman, L.F., 1992. EdITS manual for the Prole of Mood
States. Educational and Industrial Testing Service, San Diego, CA.
Montoya, M., Schiavoni, G., Mattei, F., Gresser, I., Belardelli, F., Borrow, P., Tough, D.F.,
2002. Type I interferons produced by dendritic cells promote their phenotypic and
functional activation. Blood. 99 (9), 3263–3271.
Papillion, A., Powell, M.D., Chisolm, D.A., Bachus, H., Fuller, M.J., Weinmann, A.S.,
Villarino, A., O’Shea, J.J., Le´
on, B., Oestreich, K.J., Ballesteros-Tato, A., 2019.
Inhibition of IL-2 responsiveness by IL-6 is required for the generation of GC-T
FH
cells. Sci Immunol. 4 (39) https://doi.org/10.1126/sciimmunol.aaw7636.
Potter, C.W., Oxford, J.S., 1979. Determinants of immunity to inuenza infection in man.
Br Med Bull. 35 (1), 69–75. https://doi.org/10.1093/oxfordjournals.bmb.a071545.
Proietti, E., Bracci, L., Puzelli, S., Di Pucchio, T., Sestili, P., De Vincenzi, E., Venditti, M.,
Capone, I., Seif, I., De Maeyer, E., Tough, D., Donatelli, I., Belardelli, F., 2002. Type I
IFN as a natural adjuvant for a protective immune response: lessons from the
inuenza vaccine model. J Immunol. 169 (1), 375–383. https://doi.org/10.4049/
jimmunol.169.1.375.
Ranadive, S.M., Cook, M., Kappus, R.M., Yan, H., Lane, A.D., Woods, J.A., Wilund, K.R.,
Iwamoto, G., Vanar, V., Tandon, R., Fernhall, B., 2014. Effect of acute aerobic
exercise on vaccine efcacy in older adults. Med Sci Sports Exerc. 46 (3), 455–461.
https://doi.org/10.1249/MSS.0b013e3182a75ff2.
Reihmane, D., Dela, F., 2014. Interleukin-6: possible biological roles during exercise. Eur
J Sport Sci. 14 (3), 242–250. https://doi.org/10.1080/17461391.2013.776640.
Riebe, D., Franklin, B.A., Thompson, P.D., Garber, C.E., Whiteld, G.P., Magal, M.,
Pescatello, L.S., 2015. Updating ACSM’s Recommendations for Exercise
Preparticipation Health Screening. Med Sci Sports Exerc. 47 (11), 2473–2479.
https://doi.org/10.1249/MSS.0000000000000664.
Sadarangani, M., Marchant, A., Kollmann, T.R., 2021. Immunological mechanisms of
vaccine-induced protection against COVID-19 in humans. Nat Rev Immunol. 21 (8),
475–484. https://doi.org/10.1038/s41577-021-00578-z.
Silberman, D.M., Wald, M.R., Genaro, A.M., 2003. Acute and chronic stress exert
opposing effects on antibody responses associated with changes in stress hormone
regulation of T-lymphocyte reactivity. J Neuroimmunol. 144 (1–2), 53–60. https://
doi.org/10.1016/j.jneuroim.2003.08.031.
Tough, D.F., 2004. Type I interferon as a link between innate and adaptive immunity
through dendritic cell stimulation. Leuk. Lymphoma 45 (2), 257–264. https://doi.
org/10.1080/1042819031000149368.
Trombetta, C.M., Remarque, E.J., Mortier, D., Montomoli, E., 2018. Comparison of
hemagglutination inhibition, single radial hemolysis, virus neutralization assays, and
ELISA to detect antibody levels against seasonal inuenza viruses. Inuenza Other
Respir Viruses. 12 (6), 675–686. https://doi.org/10.1111/irv.12591.
Valenzuela, P.L., Simpson, R.J., Castillo-García, A., Lucia, A., 2021. Physical activity: A
coadjuvant treatment to COVID-19 vaccination? Brain, Behavior, and Immunity. 94,
1–3. https://doi.org/10.1016/j.bbi.2021.03.003.
Vasconcelos, E.S., Salla, R.F., 2018. Role of interleukin-6 and interleukin-15 in exercise.
MOJ Immunol. 6 (1), 17-19. 10.15406/moji.2018.06.00185.
Wood, P.G., Karol, M.H., Kusnecov, A.W., Rabin, B.S., 1993. Enhancement of antigen-
specic humoral and cell-mediated immunity by electric footshock stress in rats.
Brain Behav Immun. 7 (2), 121-34. 10.1006/brbi.1993.1014.
Ye, L., Ohnemus, A., Ong, L.C., Gad, H.H., Hartmann, R., Lycke, N., Staeheli, P., 2019.
Type I and Type III Interferons Differ in Their Adjuvant Activities for Inuenza
Vaccines. J Virol. 93 (23), e01262–e1319. https://doi.org/10.1128/JVI.01262-19.
Zeng, C., Zhang, C., Walker, P.G., Dong, Y., 2020. Formulation and Delivery
Technologies for mRNA Vaccines. In: Current Topics in Microbiology and
Immunology. Springer, Berlin, Heidelberg, pp. 1–40. https://doi.org/10.1007/82_
2020_217.
J. Hallam et al.
