Submitted 14 November 2012
Accepted 28 February 2013
Published 12 March 2013
James F. Meadow,
Additional Information and
Declarations can be found on
2013 Meadow et al.
Creative Commons CC-BY 3.0
Significant changes in the skin
microbiome mediated by the sport of
James F. Meadow1, Ashley C. Bateman1, Keith M. Herkert1,2,
Timothy K. O’Connor1,3and Jessica L. Green1,4
1Biology and the Built Environment Center, Institute of Ecology and Evolution, University of
Oregon, Eugene, OR, USA
2Oregon Health & Science University, Portland, OR, USA
3Department of Ecology and Evolutionary Biology, University of Arizona, Tucson, AZ, USA
4Santa Fe Institute, Santa Fe, NM, USA
ties vary by skin location on the body, over time, between individuals, and between
geographic regions. Culture-based studies have shown that human to human and
human to surface contact mediates the dispersal of pathogens, yet little is currently
known about the drivers of bacterial community assembly patterns on human skin.
We hypothesized that participation in a sport involving skin to skin contact would
result in detectable shifts in skin bacterial community composition. We conducted a
study during a flat track roller derby tournament, and found that teammates shared
distinct skin microbial communities before and after playing against another team,
but that opposing teams’ bacterial communities converged during the course of a
roller derby bout. Our results are consistent with the hypothesis that the human
skin microbiome shifts in composition during activities involving human to human
Subjects Biodiversity, Biogeography, Ecology, Microbiology, Dermatology
Keywords Microbial biogeography, Contact sport, Human microbiome, Microbial ecology,
Skin microbiology, Microbial dispersal
Microbial communities living on and in the human skin are diverse and complex. These
human health and well-being. Skin microbial communities have been shown to mediate
skin disorders, provide protection from pathogens, and regulate our immune system
(Costello et al., 2009; Grice & Segre, 2011; Human Microbiome Project Consortium, 2012).
Despite the importance of our skin microbiota, we still know very little about what shapes
As for any other ecosystem, the composition of the skin microbiome is determined
by some combination of two simultaneous ecological processes: the selection of certain
microbial species by the skin environment and the dispersal of microbes from a pool of
How to cite this article Meadow et al. (2013), Significant changes in the skin microbiome mediated by the sport of roller derby. PeerJ
1:e53; DOI 10.7717/peerj.53
available species. Skin moisture, temperature, pH and exposure to ultraviolet light are all
well documented environmental factors that affect skin microbial communities (Grice
& Segre, 2011). The microbial species available for dispersal onto the skin of any given
individual likely stem from many sources including inanimate surfaces, people, pets,
cosmetics, air and water (Capone et al., 2011; Costello et al., 2009; Dominguez-Bello et
al., 2010; Fierer et al., 2008; Fujimura et al., 2010; Grice & Segre, 2011; Hospodsky et al.,
2012; Human Microbiome Project Consortium, 2012; Kembel et al., 2012). Our current
understanding of the relative contributions from these potential sources is nascent.
Human to surface and human to human contact have long been acknowledged as strong
vectors for microbial dispersal in the medical literature, which has been largely focused
on culture-based detection of single-species pathogen transmissions (Boyce et al., 1997;
Casewell & Phillips, 1977; Hamburger, 1947; Noble et al., 1976; Pessoa-Silva et al., 2004;
Pittet et al., 2006). In these culture-based studies, handshaking, as well as hand-contact
with other parts of the body and room surfaces, have been identified as strong vectors
of health care service infections, such as with methicillin-resistant Staphylococcus aureus
Given that human contact with surfaces, and especially the skin surfaces of others, has
Here we explore how activities involving human to human contact influence the skin
microbiome. We use a contact sport, flat track roller derby, as a model study system. Flat
track roller derby is an organized team sport, played worldwide, that involves individuals
roller-skating in close proximity and making frequent contact with other players. Roller
derby teams frequently engage in tournaments, where teams from different geographical
track roller derby tournaments present an ideal setting in which to study the transmission
of skin microbial communities during a contact sport for two main reasons. First, they
into contact with one another – members of the same team – have similar microbiomes.
membership; (2) Were team-specific skin microbiomes altered during a bout; and (3) Did
MATERIALS AND METHODS
Flat-track roller derby
For a full explanation of approved Women’s Flat Track Derby Association rules, refer to
www.wftda.com. Briefly, a bout consists of two 30-min periods, where two competing
teams, each composed of up to 4 “blockers” and 1 “jammer”, circle a track with the goal of
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.532/17
entire 60 min bout. Points are accrued when one team’s jammer makes her second, and
subsequent, pass through the pack of blockers, in effect lapping the pack. Activity occurs
in intervals called “jams”, and a single jam lasts for a maximum of 2 min. Flat track roller
derby is a contact sport; blockers are allowed to initiate contact with another player to
compete for track position using any of the following body parts: upper arm (shoulder
to elbow), torso, hips, “booty” (official WFTDA nomenclature), and mid to upper thigh.
Roller derby tournaments often involve multiple pairwise bouts in a single day between
several teams, one home team and multiple visiting teams from different geographical
locations. Players within a team practice together on a regular basis, and thus come into
frequent physical contact, and live in or near the same city. Teams involved in this study
Written consent forms were signed and collected from all participating subjects. The
Institutional Review Board Initial Application Form for the study was reviewed and
approved by the University of Oregon IRB with the Office for Protection of Human
District Human Resources office granted written permission for the study to take place in
their recreation facility. Written permission was acquired from the three teams’ coaches
Microbial communities inhabiting skin vary greatly across the human body (Grice et al.,
arm (approximately the distal end of the lateral deltoid) as our focal skin sample site. The
upper arm is the one skin region on roller derby skaters that is nearly universally exposed
and frequently contacted during a bout. All players sampled had this area exposed during
the entire bout. All samples were collected Feb. 10, 2012, at the “Big O” Tournament in
Eugene, OR, USA, and all biological samples were taken by trained technicians using
sterile technique. The two bouts that were sampled took place at 12:00pm (Emerald City
vs. Silicon Valley) and 6:00pm (Emerald City vs. DC). DC had already played in one bout
the same day at 10:00am (against Santa Cruz, not considered here), but Emerald City and
were not monitored between bouts. Samples were collected by swabbing individual’s
upper arms in a c. 4 cm by 5 cm area of skin with nylon-flocked swabs (COPAN Flock
Technologies, Brescia, Italy) moistened with sterile buffer solution (0.15M NaCl, 0.1M
were taken within 30 min of the beginning and end of each bout. Samples were stored at
−20◦C until DNA extraction. Total number of jams was recorded for each player, and
multiplied by 2 min (maximum jam length) to approximate total time played per person.
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.533/17
Four swab samples were also taken from the floor of the facility (track) following the
DNA Extraction, amplification and sequencing
Whole genomic DNA was extracted using the MO BIO PowerWater DNA Isolation Kit
(MO BIO Laboratories, Carlsbad, CA) according to manufacturers instructions with the
following modifications: swab tips were incubated with Solution PW1 in a 65◦C water
bath for 15 min prior to bead beating; bead beating length was extended to 10 min since
swab tips were included; and samples were eluted in 50 µL Solution PW6. Dual-arm
A fragment of the 16S rRNA gene including the V4 region was amplified using
a modified F515/R806 primer combination (5?-GTGCCAGCMGCCGCGGTAA-3?,
5?-TACNVGGGTATCTAATCC-3?) (Caporaso et al., 2011b; Claesson et al., 2010). Amplifi-
cation proceeded in two steps using a custom Illumina preparation protocol, where PCR1
was performed with forward primers that contained partial unique barcodes and partial
Illumina adapters. The remaining ends of the Illumina adapters were attached during
PCR2, and barcodes were recombined in silico using paired-end reads. Adapter sequences
are detailed in Supplemental Data. All extracted samples were amplified in triplicate for
PCR1 and triplicates were pooled before PCR2. PCR1 (25 µL total volume per reaction)
consisted of the following ingredients: 5 µL GC buffer (Thermo Fisher Scientific, U.S.A.),
0.5 µL dNTPs (10 mM, Invitrogen), 0.25 µL Phusion Hotstart II polymerase (Thermo
0.5 µL reverse primer, and 5 µL template DNA. The PCR1 conditions were as follows:
initial denaturation for 2 min at 98◦C; 22 cycles of 20 s at 98◦C, 30 s at 50◦C and 20 s at
72◦C; and 72◦C for 2 min for final extension. After PCR1, the triplicate reactions were
pooled and cleaned with the QIAGEN Minelute PCR Purification Kit according to the
manufacturers protocol (QIAGEN, Germantown, MD). Ten µL of 3M NaOAc (pH 5.2)
was added to decrease the pH of the pooled reactions and facilitate efficient binding to the
spin column during cleanup. Samples were eluted in 11.5 µL of Buffer EB. For PCR2, a
single primer pair was used to add the remaining Illumina adaptor segments to the ends
of the concentrated amplicons of PCR1. The PCR2 (25 µL volume per reaction) consisted
ofthe samecombinationof reagentsthatwas used inPCR1,along with5µLconcentrated
PCR1 product as template. The PCR 2 conditions were as follows: 2 min denaturation at
98◦C; 12 cycles of 20 s at 98◦C, 30 s at 66◦C and 20 s at 72◦C; and 2 min at 72◦C for
final extension. Amplicons were size-selected by gel electrophoresis: gel bands at c. 440bp
were extracted and concentrated, using the ZR-96 Zymoclean Gel DNA Recovery Kit
(ZYMO Research, Irvine, CA), following manufacturer’s instructions, quantified using a
preparation for sequencing. Samples were sent to the Georgia Genomics Facility at the
University of Georgia (Athens, GA; www.dna.uga.edu), and sequenced on the Illumina
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.534/17
Table 1 Description of the two roller derby bouts considered in analyses. Two different bouts were
sampled; bout 2 occurred approximately 5 h after bout 1. Emerald City Roller Girls played in both bouts.
Neither team in bout 1 had played a bout previously in the day, but both teams in bout 2 had done so.
Total skin samples considered in analysis = 82. Colored points correspond to those used in all figures.
n PlayersBout1st Bout
of the day
Raw sequences were processed using the FastX Toolkit (http://hannonlab.cshl.edu/fastx
toolkit) and the QIIME pipeline (Caporaso et al., 2010). Barcodes were recombined from
paired-end reads, and forward reads were used for downstream analysis. All sequences
were trimmed to 112 bp, including a 12 bp barcode, and low quality sequences were
removed. Quality filtering settings were as follows: minimum 30 quality score over at
least 75% of the sequence read; no ambiguous bases allowed; 1 primer mismatch allowed.
After quality control and barcode assignment, the remaining 1,368,938 sequences were
binned into operational taxonomic units (OTUs) at a 97% sequence similarity cutoff
using uclust (Edgar, 2010). The highest-quality sequences from each OTU cluster were
taxonomically identified using reference sequences from Greengenes (DeSantis et al.,
Project Consortium, 2012; Kuczynski et al., 2010) we rarefied all samples to 500 sequences
per sample. Samples with fewer than 500 sequences were not used in subsequent analyses
(Table 1); since some low-yield samples were removed from all team groups, players
were not considered as paired samples before vs. after a bout. Samples analyzed for this
study were compiled from two separate MiSeq runs, and additional aspects of this study
were included in the MiSeq runs but are not considered here, so the returned sequence
count does not reflect the full volume of the runs. Three of the track samples contained
samples, but were not used in any ordination analysis. Sequence files and metadata for all
All statistical analyses were performed in R. Community variation among samples, or
β-diversity, was calculated using the quantitative, taxonomy-based Canberra distance,
implemented in the vegan package (Oksanen et al., 2011). Non-metric multidimen-
sional scaling (NMDS) was performed using the bestnmds function in the labdsv
package (Roberts, 2010), using 20 random starts. Discriminant analysis of within-group
similarity was conducted using permutational MANOVA with the adonis function
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.535/17
in vegan. To determine whether skin microbial communities became more similar to
one another after playing in a bout, we used a β-dispersion test with the betadisper
function in vegan. This test is a multivariate analog of Levene’s test for homogeneity of
variances (Anderson, Ellingsen & McArdle, 2006), and it tests for a significant difference in
sample heterogeneity between groups (i.e. the spread of data points in ordination space).
Indicator analysis (Dufrene & Legendre, 1997), using indval in labdsv, was conducted
on each team group, and all players combined before vs. after respective bouts, to identify
between time played and change in community composition was assessed with Pearson’s
sequences. After quality filtering and rarefaction, 82 samples were considered during
analysis, taken from 3 teams and two bouts (Table 1). Emerald City Roller Girls (EC)
played in both bouts and thus were included twice in the analyses as two different team
groups; Silicon Valley Roller Girls (SI) and DC Roller Girls (DC) played in the first and
second bouts, respectively, against EC. Including EC players twice in the study allowed
us to evaluate the change in community composition in a single team after playing
successive bouts. Some EC players were sampled in both bouts, but were not analyzed
on a paired-sample basis. Rarefaction to 500 sequences per sample left 1034 bacterial
OTUs, with the most abundant OTU (Corynebacterium sp.) representing c. 34% of
total sequences, and c. 55% of all OTUs represented by singletons (i.e. a single sequence
represents asingle OTU).The bacterialtaxa identifiedin ourskin sampleswere consistent
with what has been reported in other skin microbiome studies (e.g. Caporaso et al., 2011a;
Costello et al., 2009; Grice & Segre, 2011; Kong, 2011) When considered at the taxonomic
class level, the majority of sequences were Actinobacteria (57.9%), followed by Bacilli
(23.4%), Gammaproteobacteria (7.4%), Betaproteobacteria (3.7%), Alphaproteobacteria
(2.7%), andClostridia(1.3%). Allskin samples weredominated by skin-associatedgenera
(especially Corynebacterium, Micrococcus, Staphylococcus, and Acinetobacter), and by
oral-associated genera (including Neisseria and Rothia), both before and after bouting.
Post-bout samples, however, did contain higher relative abundances of a few soil- and
plant-associated genera, especially Arthrobacter and Xanthomonas. Normalized OTU
richness, at the 500 sequences per sample level (mean richness = 67.6 OTUs), was not
significantlydifferentafterabout(t = 0.007;p = 0.9;fromaWelchtwo-samplet-test).
Were players’ skin microbiomes predicted by team membership?
Bacterial communities detected on players’ upper arms from different teams were signif-
icantly different before playing a bout, as well as after playing a bout (Table 2). In other
words, the skin microbiome of an individual player was predicted by team membership.
Teams clustered together in ordination space using a non-metric multidimensional
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.536/17
tity. Ordination diagrams (axes 1 and 2 from separate 3-dimensional NMDS ordinations) summarizing
vs. SI ) and before bout 2 (ECvs. DC
deviations around community variances from each team. The skin bacterial communities of the four
team groups were significantly different before playing a bout (p<0.001; from permutational MANOVA
on Canberra taxonomic distances). (B) The four team groups are also significantly different after playing
bouts (p < 0.001), though more overlap is observed between teams after bout 1 (EC
after bout 2 (EC
vs. DC). NMDS 3-dimensional stress = 19.66 (A) & 17.55 (B).
). Corresponding-colored ellipses show standard
vs. SI ) and
Table 2 Results from Permutational MANOVA on Canberra distances among skin bacterial commu-
nities sampled from players before and after bouting. Each team was considered individually when
testing for intra-team before/after community shifts, while teams were considered together for the “all
players” before/after test. Team identity was used as a grouping factor to test inter-team clustering (“all
teams”), both before and after bouts. Emerald City was considered to be two different teams (bout 1 and
bout 2) in analyses.
*Significant at p < 0.05 level.
scaling representation of players’ skin microbiomes both before (Fig. 1A) and after
(Fig.1B)playing about,basedonCanberra taxonomic distances.Thoughteamclustering
is significant in both cases (before and after a bout), there is a greater degree of overlap
betweenthe teamsfollowing bouts.Emerald Citywas consideredastwo differentteams in
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.537/17
Figure 2 Home team (EC) players’ skin microbiomes were more similar to the microbial community
detected on the roller derby track than visiting teams. When each player’s pre-bout skin microbiomes
were compared to the microbial communities found on the track surface, Emerald City players’ skin
microbiomes were significantly more similar on average to the three track samples than were the skin
microbiomes of players from Silicon Valley or DC. The same is true when considering teams on a
per-bout basis (p = 0.001 & 0.007; for bouts 1 & 2, respectively).
communities detected on the track than the two visiting teams when considered together
(p < 0.001; from a Welch two-sample t-test; Fig. 2), and when considered separately
(p = 0.001 & 0.007; for bouts 1 & 2, respectively). All players were more dissimilar on
average from track samples (mean Canberra distance = 0.89) than from all other players
(mean Canberra distance = 0.83; p < 0.001; from a Welch two-sample t-test). Player
bacterial communities did not become more similar to the track after a bout, and in fact
bothbout2teams(EC&DC)becamelesssimilartothetrackfollowingabout(p = 0.008
Were team-specific skin microbiomes different after playing a
arms before a bout were significantly different than those detected after the bout in all
cases (Table 2; Fig. 3). We also detected a signal of already having played in a bout. Two
and that was a significant predictor of community composition before the second bout
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.538/17
Figure 3 Team-specific micobiomes are significantly different after playing in a bout. NMDS ordi-
nation diagrams summarizing similarity of skin bacterial community composition when all players
are compared within their own teams before and after a bout. All ordinations are based on Canberra
taxonomic distances. (A) Emerald City before
afterbout 1; (C) Emerald City before and after
2. Corresponding-colored ellipses are standard deviations on community variances for each group. All
stress: A = 8.1, B = 10.47, C = 16.2, D = 17.65.
bout 1; (B) Silicon Valley before
bout 2; (D) DC before
bout and after
Did opposing teams’ skin microbiomes become more similar after
competing in a bout?
All players’ skin microbiomes were more similar to one another after competing in a
bout. To test this we conducted a β-dispersion test, which compared all players before
and after bouting. A significant reduction in β-dispersion between groups (before
F = 11.79;p < 0.001;Table4;Fig.4)whenallplayerswereconsideredtogether,andwhen
players were grouped by the bout in which they played (F = 12.41 & 4.11; p = 0.002
& 0.048; Table 4). A greater proportion of OTUs was shared by competing teams after
bouting compared to before (Table 5), but, interestingly, the opposite is true in 3 out of 4
caseswhencomparing non-competingteams.Wheneach teamwasconsideredseparately,
both teams in bout 1 experienced a significant β-dispersion reduction as did EC after
playing in bout 2, while DC did not (Table 4; Fig. 4). None of the four teams experienced
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.53 9/17
Figure 4 Bacterial community variance is reduced after playing in a bout for all players and for three
of the four teams individually. When all players were considered, regardless of team identity, bacterial
communities were significantly more similar to one another after a bout than they were before a bout
(p < 0.001). Both teams in bout 1 (EC and SI), as well as EC in bout 2, showed the same microbial
community convergence. Points are jittered around the x-axis to more clearly describe distributions. All
that players’ skin micobiomes were more similar to one another after playing in a bout. Colored points
correspond to Table 1 and Figs. 1 and 3.
a significant shift in Shannon-Wiener OTU diversity or evenness after playing in a bout,
nor was there a difference when all players were considered together (all p-values > 0.2).
1 had played during the same day. Changes in bacterial communities before and after a
bout were not correlated with each players time spent in a bout (Pearsons correlation test;
ρ = 0.12;p = 0.45).Jammersandblockersplaydifferentrolesonateam,andthusengage
in different amounts of contact and time played; however, since players are not limited
to a single position during a bout, we did not differentiate between the positions during
We conducted indicator analysis to identify OTUs responsible for the observed difference
comparison adjustment, and 11 were significant after adjustment (Table 3); indicator
taxa are presented at the finest taxonomic resolution provided by Greengenes assignment.
Though the significant indicators include bacterial genera commonly associated with
soil and plants (e.g. Dietzia & Xanthomonas), notable human-associated groups also
stand out (e.g. Coprococcus, Streptococcus, Lachnospiraceae & Brevibacterium). This list
of indicators includes both rare OTUs that account for <1% of sequences, and relatively
common OTUs (Alicyclobacillus made up >25% of pre-bout SI sequences). We also
identified OTUs that were shared between competing teams in post-bout samples but
not in pre-bout samples. Six OTUs fit this description for both bouts, belonging to six
different bacterial genera: Streptococcus, Sphingomonas, Eubacterium, Porphyromonas,
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.5310/17
Table 3 Indicator OTUs from each team prior to sampled bouts. All bacterial OTUs were tested as
indicator taxa for any of the four teams before playing in a bout, using Dufrene and Legendre’s (1997)
procedure. Only those significant at the p<0.05 level after multiple comparison adjustments are shown.
Percentages are calculated as a share of all sequences in the dataset, and as a share of pre-bout sequences
from each team for whom the OTU is indicative. Six of the eight pre-bout indicator OTUs detected in
opposing teams’ samples increased (+) in mean abundance for opposing team members after playing in
OTU GenusTeam (a)Bout Indicator valuePercent of
*Significant at p < 0.05 level after Holm’s correction for multiple comparisons.
aIndicates whether OTU increased (+) or decreased (−) in average abundance when detected in opposing team.
ances from each team before and after, as well as all players regardless of team identity. The first four
tests describe β-dispersion tests (comparison of within-team bacterial community variance) when each
team is considered individually before and after a bout, and the fifth ignores team identity. Results
indicate that skin bacterial communities from Emerald City (bout 1) and Silicon Valley players both
became more similar following a bout, as did Emerald City from their 2nd bout. But this was not the case
for DC after playing in bout 2. Bacterial communities became more similar when players were grouped
by the bout in which they played, and when all players were considered in the same analysis.
*Significant at p < 0.05 level.
Aerococcus and Methylobacterium. It is worth noting that although these OTUs showed a
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.5311/17
Table5 ThenumberofOTUssharedbycompetingteamsincreasedafterabout. When the proportion
(↓) in shared OTUs.
Teams (Bouts) Percent of OTUs shared
Competing teamsEC(1) & SI(1)
EC(2) & DC(2)
EC(1) & EC(2)
EC(1) & DC(2)
EC(2) & SI(1)
DC(2) & SI(1)
Bacteria are ubiquitous. Those inhabiting the human body have received increased
attention in recent years, owing to a greater appreciation of the interrelated nature
of humans and their microbiome, an improved understanding of microbial ecology,
and an unprecedented ability to detect fine-scale microbial community changes with
high-throughput sequencing technology (Human Microbiome Project Consortium, 2012).
The skin is the largest organ and an important barrier that regulates microbial entry
into the human body. Despite the importance of the skin ecosystem to human health
and well-being, we know very little about the forces that shape microbial structure
and composition in the skin environment. The present study was designed as a way
to understand how human to human contact influences the skin microbiome, since
contact has long been acknowledged as a major dispersal vector for skin bacterial
We found that team membership was a strong predictor of skin microbial community
composition, and that differences between teams were partly driven by the presence of
unique indicator taxa that are commonly associated with human skin, gut, mouth, and
respiratory tract. For example, Brevibacterim was the sole indicator taxon for DC. While
short reads are limited in their ability to identify bacterial taxa to the species level, this
been studied for their role in foot odor (Dixon, 1996). The strong microbial fingerprint
linked to each team could be because they were from three distinctly different geographic
locationswithinthe UnitedStates(Eugene, OR;SanJose,CA; andWashington,DC),each
associated with a different climate, urban setting, and outdoor macrobiota. These cities
may also have very different environmental microbiota. Blaser et al. (2012) recently found
communities. Consistent with the idea that humans carry a microbial fingerprint that
reflects where they live, we found that home team (EC) microbiomes were more similar
to their home track than either of the visiting teams prior to bouting. This is also the EC
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.5312/17
with the track surface since they shed skin cells and frequently come into direct contact
with the floor. While a variety of factors likely contribute to this geographic signature, it is
Although each team retained their microbial fingerprint, we found that team microbial
communities became more similar to one another after players competed in a bout. This
bouts, and when considering each team individually, though the latter was not the case
for DC, who played in the second bout. Several reasonable explanations arise given these
results: (1) all players were exercising, and exercise produces predictable changes in skin
acquiring microbial transients from the built environment; and (3) players were coming
using the sampled area of their upper arms, and potentially sharing portions of their
skin microbiomes. With regards to explanation (1), the current study was not set up to
communities. It seems unlikely that 60 min of elevated skin temperature and perspiration
would be long enough for microbial growth dynamics to effect the magnitude of changes
observed, given that bacterial doubling times generally exceed 20 min even in optimal
conditions. It is possible that exercise results in migration from subcutaneous habitats to
the skin surface, but little is known about this potential mechanism. Additionally, both
bouts resulted in a greater proportion of shared OTUs between competing teams, but
not between non-competing teams except when considering only EC over the course
of their two bouts, arguing against an overall exercise effect. Finally, none of the teams
experienced a shift in Shannon-Wiener diversity or evenness, which would be expected
in an exercise-driven community shift, since metabolically active bacteria might come
to dominate the community with a change in pH, temperature and moisture at the skin
Explanations (2) and (3) above both derive from dispersal, either from the built
environment or from other players. Dispersal from the built environment to skaters is
likely, since roller derby and spectator movements stirred up dust from the recreational
venue, and players also frequently fall on the floor. Although humans have been estimated
to contribute more than 106airborne microbial cells per-hour (Qian et al., 2012),
culture-based disease transmission studies suggest that direct contact with humans and
other surfaces is a stronger bacterial dispersal vector than airborne particles (Casewell &
Phillips, 1977; Pessoa-Silva et al., 2004; Pittet et al., 2006). We found that human to track
surface contact did not seem to explain the observed shifts in community composition,
sincenoneof thefourteamgroupsbecame moresimilartothetrack samplesafterplaying
in a bout. Given that the proportion of OTUs shared between competing teams increased
after both bouts, but not between non-competing teams, human to human contact is
the most parsimonious interpretation for the significant changes in skin microbiome we
observed. Future research - particularly over longer time scales - is needed to understand
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.53 13/17
We know very little about how our social, family, and professional interactions shape
our microbial identities. Contact sports are an ideal setting in which to study how human
to human interactions influence our microbial ecosystems. As the rise of mega-cities and
population growth continues, humans may experience an increased rate of person to
person contact mediated by urban living and global travel. To predict the implications of
these changes will require, in part, understanding the ecological and evolutionary forces
that act on the skin microbiome. A thorough comprehension of the drivers of the skin
microbiome is still emerging; novel approaches to studying our skin ecosystems will likely
have lasting implications for health care, disease transmission, and our understanding of
We would like to thank the players and coaches who facilitated and participated in this
office provided permission to conduct sampling in the recreation facility. H. Arnold,
R. Mueller, P. Pillai, J. Reichman, Z. Stephens, A. Womack, M. Naidoo, and Super Cake
helped with sampling and development of molecular protocols. We thank members of the
Bohannan andGreen Labs forvaluable inputon this research,and also threereviewers for
ADDITIONAL INFORMATION AND DECLARATIONS
Research reported in this publication was supported by the Alfred P. Sloan Foundation
under award number 2010-5-22 IEC and the University of Oregon. The funders had no
• Timothy K. O’Connor conceived and designed the experiments, performed the
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.53 14/17
• Jessica L. Green conceived and designed the experiments, performed the experiments,
The following information was supplied relating to ethical approvals (i.e. approving body
Supplemental information for this article can be found online at http://dx.doi.org/
Anderson MJ, Ellingsen KE, McArdle BH. 2006. Multivariate dispersion as a measure of beta
diversity. Ecology Letters 9:683–693 DOI ./j.-...x.
Blaser MJ, Dominguez-Bello MG, Contreras M, Magris M, Hidalgo G, Estrada I, Gao Z,
Clemente JC, Costello EK, Knight R. 2012. Distinct cutaneous bacterial assemblages in a
sampling of south american amerindians and us residents. The ISME Journal: Multidisciplinary
Journal of Microbial Ecology 7:85–95 DOI ./ismej...
Boyce JM, Potter-Bynoe G, Chenevert C, King T. 1997. Environmental contamination due to
methicillin-resistant Staphylococcus aureus: possible infection control implications. Infection
Control and Hospital Epidemiology 18:622–627 DOI ./.
Capone KA, Dowd SE, Stamatas GN, Nikolovski J. 2011. Diversity of the human skin microbiome
early in life. Journal of Investigative Dermatology 131:2026–2032 DOI ./jid...
Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N,
Pena AG, Goodrich JK, Gordon JI, Huttley GA, Kelley ST, Knights D, Koenig JE, Ley RE,
Lozupone CA, McDonald D, Muegge BD, Pirrung M, Reeder J, Sevinsky JR, Tumbaugh PJ,
Walters WA, Widmann J, Yatsunenko T, Zaneveld J, Knight R. 2010. QIIME allows
analysis of high-throughput community sequencing data. Nature Methods 7:335–336
Caporaso JG, Lauber CL, Costello EK, Berg-Lyons D, Gonzalez A, Stombaugh J, Knights D,
Gajer P, Ravel J, Fierer N, Gordon JI, Knight R. 2011a. Moving pictures of the human
microbiome. Genome Biology 12:R50 ISSN 14656914 DOI ./gb----r.
Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Lozupone CA, Turnbaugh PJ, Fierer N,
Knight R. 2011b. Global patterns of 16s rRNA diversity at a depth of millions of
sequences per sample. Proceedings of the National Academy of Sciences USA 108:4516–4522
Casewell M, Phillips I. 1977. Hands as route of transmission for Klebsiella species. British Medical
Journal 2:1315–1317 DOI ./bmj....
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.53 15/17
Claesson MJ, Wang Q, O’Sullivan O, Greene-Diniz R, Cole JR, Ross RP, O’Toole PW. 2010.
Comparison of two next-generation sequencing technologies for resolving highly complex
microbiota composition using tandem variable 16s rRNA gene regions. Nucleic Acids Research
38:e200 DOI ./nar/gkq.
Costello EK, Lauber CL, Hamady M, Fierer N, Gordon JI, Knight R. 2009. Bacterial community
variation in human body habitats across space and time. Science 326:1694–1697
Davis MF, Iverson SA, Baron P, Vasse A, Silbergeld EK, Lautenbach E, Morris DO. 2012.
Household transmission of meticillin-resistant staphylococcus aureus and other staphylococci.
Lancet Infectious Diseases 12:703–716 DOI ./S-()-.
DeSantis TZ, Hugenholtz P, Larsen N, Rojas M, Brodie EL, Keller K, Huber T, Dalevi D,
Hu P, Andersen GL. 2006. Greengenes, a chimera-checked 16S rRNA gene database and
workbench compatible with ARB. Applied and Environmental Microbiology 72:5069–5072
Dixon B. 1996. Cheese, toes, and mosquitoes. BMJ 312:1105 DOI ./bmj....
Dominguez-Bello MG, Costello EK, Contreras M, Magris M, Hidalgo G, Fierer N, Knight R.
2010. Delivery mode shapes the acquisition and structure of the initial microbiota across
multiple body habitats in newborns. Proceedings of the National Academy of Sciences USA
107:11971–11975 DOI ./pnas..
Dufrene M, Legendre P. 1997. Species assemblages and indicator species: the need for a
flexible asymmetrical approach. Ecological Monographs 67:345–366 DOI ./-
Edgar RC. 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics
26:2460–2461 DOI ./bioinformatics/btq.
Fierer N, Hamady M, Lauber CL, Knight R. 2008. The influence of sex, handedness, and washing
on the diversity of hand surface bacteria. Proceedings of the National Academy of Sciences USA
105:17994–17999 DOI ./pnas..
Fujimura KE, Johnson CC, Ownby DR, Cox MJ, Brodie EL, Havstad SL, Zoratti EM,
Woodcroft KJ, Bobbitt KR, Wegienka G, Boushey HA, Lynch SV. 2010. Man’s best friend? the
effect of pet ownership on house dust microbial communities. Journal of Allergy and Clinical
Immunology 126:410–412 DOI ./j.jaci....
Grice EA, Kong HH, Conlan S, Deming CB, Davis J, Young AC, Bouffard GG, Blakesley RW,
Murray PR, Green ED, Turner ML, Segre JA. 2009. Topographical and temporal diversity of
the human skin microbiome. Science 324:1190–1192 DOI ./science..
Grice EA, Kong HH, Renaud G, Young AC, Bouffard GG, Blakesley RW, Wolfsberg TG,
Turner ML, Segre JA. 2008. A diversity profile of the human skin microbiota. Genome Research
18:1043–1050 DOI ./gr...
Grice EA, Segre JA. 2011. The skin microbiome. Nature Reviews Microbiology 9:244–253
Hamburger M Jr. 1947. Transfer of beta hemolytic streptococci by shaking hands. American
Journal of Medicine 2:23–25 DOI ./-()-X.
Holm S. 1979. A simple sequentially rejective multiple test procedure. Scandinavian Journal of
Hospodsky D, Qian J, Nazaroff WW, Yamamoto N, Bibby K, Rismani-Yazdi H, Peccia J.
2012. Human occupancy as a source of indoor airborne bacteria. PLoS ONE 7:e34867
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.53 16/17
Human Microbiome Project Consortium. 2012. Structure, function and diversity of the healthy Download full-text
human microbiome. Nature 486:207–214 DOI ./nature.
Kembel SW, Jones E, Kline J, Northcutt D, Stenson J, Womack AM, Bohannan BJ, Brown GZ,
Green JL. 2012. Architectural design influences the diversity and structure of the built
environment microbiome. The ISME Journal: Multidisciplinary Journal of Microbial Ecology
6:1469–1479 DOI ./ismej...
Kong HH. 2011. Skin microbiome: genomics-based insights into the diversity and role of skin
microbes. Trends in Molecular Medicine 17:320–328 DOI ./j.molmed....
Kuczynski J, Liu ZZ, Lozupone C, McDonald D, Fierer N, Knight R. 2010. Microbial community
resemblance methods differ in their ability to detect biologically relevant patterns. Nature
Methods 7:813–821 DOI ./nmeth..
Noble WC, Habbema JDF, van Furth R, Smith I, de Raay C. 1976. Quantitative studies
on the dispersal of skin bacteria into the air. Journal of Medical Microbiology 9:53–61
Oksanen J, Blanchet FG, Kindt R, Legendre P, O’Hara RB, Simpson GL, Solymos P,
Stevens MHH, Wagner H. 2011. vegan: Community Ecology Package. http://CRAN.R-
project.org/package=vegan, R package version 1.17-6.
Pessoa-Silva CL, Dharan S, Hugonnet S, Touveneau S, Posfay-Barbe K, Pfister R, Pittet D. 2004.
Dynamics of bacterial hand contamination during routine neonatal care. Infection Control and
Hospital Epidemiology 25:192–197 DOI ./.
Pittet D, Allegranzi B, Sax H, Dharan S, Pessoa-Silva CL, Donaldson L, Boyce JM, WHO Global
Patient Safety Challenge, World Alliance for Patient Safety. 2006. Evidence-based model for
hand transmission during patient care and the role of improved practices. Lancet Infectious
Diseases 6:641–652 DOI ./S-()-.
Qian J, Hospodsky D, Yamamoto N, Nazaroff WW, Peccia J. 2012. Size-resolved emission
rates of airborne bacteria and fungi in an occupied classroom. Indoor Air 22:339–351
Roberts DW. 2010. labdsv: Ordination and Multivariate Analysis for Ecology.
http://CRAN.R-project.org/package=labdsv, R package version 1.4-1.
Meadow et al. (2013), PeerJ, DOI 10.7717/peerj.5317/17