- <div style="background-image:url(/live/image/gid/32/width/1600/height/300/crop/1/41839_V14Cover_Lynch_Artwork.2.rev.1520229233.png)"/>
Sweating Up a Storm: Areas of Regular Flooding Disturbance Modeled Through Human Skin Bacteria on Athletes and Non-Athletes
Department of Biology
Lake Forest College
Lake Forest, Illinois 60045
Disturbances have a major role in a number of processes involving diversity. While disturbance is usually thought of as either large scale or man-made, often they are smaller scale and environmental, such as flooding or mudslides. It has been observed that disturbance regularity may play a key factor in a variety of systems. Disturbance can affect the environment by reducing species numbers. One study found that the number of tree species would increase with a decrease in the amounts of disturbance found in the region those trees are growing (Dzwonko & Loster, 1997). Species richness was higher in areas with lower disturbance. There have also been observations that regular disturbance events have a tendency to lower overall biodiversity, in both terms of species composition and abundance. One study found that some plant species were more affected by disturbance than others, as well as a distinct change in the plant cover and total biomass of those species (Wollgast et al, 2008). This could suggest that large species are also much more susceptible to the effects of large scale disturbance, with regular disturbance being a driver for local extinctions. In its most extreme cases, it has been noted that disturbances of large scales that happen regularly could also change environmental conditions completely, such as landslides in the Andes causing changes in amounts of biospheric carbon (Clark et al, 2015). This also suggests that disturbances could indirectly change the species composition of a region by changing atmospheric or water conditions in that region.
One specific and common type of regular disturbance is flooding, which happens to many regions during high precipitation. The importance of flooding has two factors: the first of which is that it occurs due to increased rainfall, making its regularity variable and the second being that the length in time of the disturbance varies depending upon the habitat it occurs in. The length of this disturbance type could be affected by variables such as temperature, elevation and plant matter. Flooding has also been seen having an interesting effect upon individuals within floodable areas, specifically with the length of the flooding period. In one study, it was observed that plants of the Salicaceae family had developed a specific life history as a way of surviving both the flooding disturbance and erosion effects that occurred while species that weren’t using these traits weren’t found in floodplains (Karrenberg et al, 2002). This suggests that longer term regular disturbance is something which species need to be adapted for. It was also noted that plant species in extreme flooding or changes in water conditions actually changed to fit the environmental need in wet grasslands, such as changing their flowering time or their root system as needed (Brotherton and Joyce, 2015). Through this effect, flooding may act as a driver for speciation, and that resilience to large scale disturbance may be a trait that is selected for in a number of high flooding environments.
While there are a number of environments in which the effects of disturbance can be observed, studying in these environments can be costly and unpredictable, since people cannot easily cause disturbance events such as flooding. Instead, a more appropriate model would be bacteria on human skin. On any given human body, there are innumerable amounts of bacteria present. Humans and other organisms have a mutualistic relationship to bacteria: bacteria help us in processes such as digestion or protection from more harmful illnesses. For example, biodiverse and high in species richness groups of bacteria have been observed to have beneficial effects for Panamanian gold frogs from harmful fungi (Becker et al, 2015). The number of bacteria could prevent species from dispersing through the varying ways which they get onto the human skin, including by touch or through other forms of contact. In another study, it was seen that the composition of bacteria on the skin was also important, specifically that certain bacteria prevented fungal growth on toads (Park et al, 2013). Thus it can be seen that the assemblage of bacteria has an important role for bodily health and function.
Due to the biodiversity among bacteria, specifically that on the human skin, we can assume that the human skin acts as an ecosystem similar to that of others, and is also susceptible to disturbances. One study found that feather length was positively correlated with specific changes in skin bacteria of pied flycatchers, suggesting that with continued growth, there was a change in the overall diversity and composition of that microhabitat (Gonzalez-Braojos et al, 2015). This would show that there are also successional events with bacteria, making it clear they have a similar dynamic to environments which have disturbance events. More similarly, skin bacteria also have their own version of flooding, specifically sweating. Sweating has the effect of both increasing salinity and moisture, thus creating defined differences in the human skin environment in ways which bacteria are somewhat sensitive to. In one study, it was observed that the biodiversity of bacteria was decreased with regular use of antiperspirants and deodorants (Callewaert et al, 2014). It can be assumed that since antiperspirants and deodorants are used as a way to deal with excess sweating, that those who were using them more in the study sweated more and that therefore the sweating was correlated to the lowered number of bacteria.
Using bacteria as a model system, this study observed four separate potential environments in which the variable of regular sweating could be observed. The way which regular flooding was emulated was using athletes to represent more regular flooding, while non-athletes represented non-regular disturbance or lack of disturbance. Length of the flooding disturbance would instead be modeled by two body parts, the neck and the crease of the elbow. On the front of the neck, sweat evaporates much more quickly, while the crease of the elbow retains moisture for longer. The four sample types show changes in the regularity and length of sweating. From these groups, three hypotheses were made. First, the similarity of two sites with regular disturbance should be higher than that of two sites without regular disturbance due to selection for resilience to disturbance. If this is true, then there should be more similarity between the neck and elbows of athletes than the necks of athletes. Secondly, regular disturbance should lower both species richness and the abundance of bacterial colonies, both in total bacterial numbers and across species. If this hypothesis is true, we should see species numbers and abundances reduced in athletic individuals who sweat more often. Finally, the length of disturbance should lower the abundance and the total species richness overall, thus making the elbow crease area be less biodiverse overall.
We collected samples from the Lake County area of Illinois. The samples were split into two groups, one of them being non-athletes and the other being athletes, with 8 male subjects and 8 female subjects in each group, giving us a total of 16 individuals per group. Subjects were contacted a few days before the study and asked to not shower and exercise for 24 hours prior to sample taking to ensure that there is no large scale disturbance of bacteria on their body. Each subject swabbed themselves rather than being swabbed by a tester. Two samples were taken from each person, one on the crease of their elbow and the other on the front of their neck. The subjects were asked to swab in small circular motions to get proper amounts of bacteria on the cotton swab. After sampling was complete, the sample was placed in a 0.5 PBS solution for approximately 1-2 hours. Each sample was then diluted to 1/200 the original concentration and grown on an agar plate after being in the dilution gel for 3 days in an incubator at 37 degrees Celsius.
Statistical Analysis: For each subject in the experiment, the number of morphologically different bacteria was counted. Then, the similarity between the two samples from each subject was calculated using a Sorenson coefficient calculation.
(a=shared, b=species unique to the elbow, c=species unique to the neck)
Following this, a two sample t-test was used to test for the similarity between the two types of subjects and determine whether there was a statistically significant difference between how similar the two types of individuals are in terms of biodiversity. T-tests were also used to compare athletes and non-athletes on species numbers and abundances, while the same tests were used between elbows and necks. An ANOVA single factor test was run when species incidence was found across the different groups.
Across 64 samples, 14 separate species were found. Amongst them, 7 appeared on all four sample types, while the endemic one that we termed ‘brown abnormal’, was only found on a singular non-athlete neck (Table 1). Elbow samples were 95.65% similar while neck samples were 81.62% similar overall. Elbow sites were 88% similar to neck sites. Athletes were overall 96.3% similar to non-athletes.
In terms of Sorenson comparisons, the mean similarity for athletes was 28.9% and the mean similarity for non-athletes was 30.81% (T_15= -0.97, 0.4229) (Figure 1A). The mean number of species for athletes was 4.25 and for non-athletes it was 4 (T_15=0.32, 0.374) (Figure 1B). The mean number of colonies per athlete individual was 21 and the mean number of colonies per individual for non-athletes was 26.813 (T_15=-0.46, 0.324) (Figure 1C).
When elbows were compared, for elbows the mean number of species was 2.938 and the mean number on non-athlete elbows was 2.25 (T_15=1.337, 0.10054) (Figure 2A). The mean abundance on each plate for athlete elbows was 12.98 and for non-athlete elbows it was 7.875 (T_15=0.909, 0.189) (Figure 2B). When necks were compared, the mean species number found on athletes was 2.125 and for non-athletes it was 2.688 (T_15= -0.929, 0.184) (Figure 2C). The mean abundance on neck plates 8.0625 for athletes and for non-athletes was 18.94 (T_15= -0.975,0.172) (Figure 2D). When compared overall, athlete elbows had the highest mean species amount from all four categories and the lowest mean species amount was found on athlete necks (F_63= 0.948, 0.424) (Figure 3A). In terms of abundance, non-athlete necks had the highest amount of mean abundance and the lowest amount was found on non-athlete elbows (F_63=0.795, 0.501) (Figure 3B).
The species with the highest incidence was the clear white circular bacteria and the species with the lowest incidence was the brown abnormal bacteria. The sample type with the highest average incidence across species was the athlete elbow while the lowest average incidence across species was athlete necks (F_55= 0.340, 0.797) (Figure 4A). The highest average abundance was the tiny circles and the lowest average abundance was brown abnormal. Overall the sample type with the highest average abundance across the species was athlete elbows and the lowest average abundance was athlete necks (F_55=0.811, .494) (Figure 4B).
Overall, there was no support for any of the predictions made by the hypotheses listed above. The similarity in sites with regular disturbance had a lower mean similarity than those without the regular disturbance events, and had lesser abundances of the different species found in our study and lower mean species numbers. Also, regions with longer disturbances did show higher numbers of species, but the highest abundances were found in regions with less lengthy disturbances. Interestingly, areas with both lengthier disturbances and more regular disturbances or less lengthy and less regular disturbances showed the highest amount of species and abundances, suggesting that the bacterial species present were either adapted to one condition or another. This can also be observed that certain species were more present in different types based on how close they were to disturbance. However, there was no support for any of the claims made earlier since there were no statistically significant trends throughout.
There were a few issues with the study’s capabilities throughout the work. The first issue was the lack of consistency throughout the study with sampling. One issue was the assumption that sweat regularity and length could be equated with how often a person worked out, specifically calling athletes anyone who did three to four or more exercise periods per week. However, this doesn’t necessarily mean their schedule is regular, nor does it correct for times when there is less exercise or sweating overall. There is also the possibility that the exercise conditions had sanitation issues, allowing for bacteria which wouldn’t be common on athletes to appear within their plates. Throughout the experiment, we performed the work with only minor sanitation practices throughout, suggesting that some bacterial parties may have been over-exaggerated during the work due to contamination. The major issue is that non-athletes and athletes could actually overlap in terms of how much they sweat, specifically since there wasn’t a way in which the non-athletes were prevented from sweating in non-exercising conditions (such as times when temperatures were higher). Most likely, there would need to be more controlled protocols for both sample selection and sampling process to ensure there are no such issues afterwards. This would include changing the selection of athletes to be a specific amount of exercise and ensuring that participants are not sweating if they aren’t athletes.
There are several important implications to the study, despite the primarily insignificant results. The first being that these results were not similar to papers which discussed the outcome of changes in abundances or species numbers referred to earlier. This could suggest that potentially sweating may not act similarly to a flooding event or that the human biome is not analogous to the environments which go through flooding systems. Alternatively, it may be that human body does act as a disturbed environment, but that bacteria do not act similarly to other species. One example of this is that species numbers were higher in areas with either high amounts of lengthy disturbance or areas with less regular disturbance for less time. This would suggest that those bacteria were able to adapt to one region more or another much more quickly than other species to the events.
The majority of bacteria species appearing on all sample types could be explained by the possibility that the disturbances of sweating created a patchier environment than previously thought, which allowed for bacteria with lower resilience to sweating to be able to survive on athlete plates. In one study on birds in areas of high disturbance, it was noted that less disturbed areas had more homogenous environments (Lent and Capen, 1995). If this is the case, then the areas with more regular and lengthy disturbances may have more species richness throughout the area because of their patchiness. The possibility of spatial heterogeneity being a factor could also explain how in many cases the areas with more regular disturbance had higher species numbers and higher abundance. In another case, it was actually seen that this patchiness actually provided resources for birds in Southwestern Riparian forests, since the gradient of flooding created a number of areas more or less suitable for the birds and allowing for a higher biodiversity of birds (Brawn et al, 2001). This could also be a possibility since there is not instantaneous drying of the skin, but like all disturbances, a gradient of areas affected to different degrees throughout the skin. Another factor is that bacteria may be grouped together, which allows for higher amounts of species. Clustering has been observed as having a benefit for biodiversity in comparison to random habitat differences (Matlack et al, 2007). This could be evident that the clustering in high bacteria areas (such as areas where there is a lot of contact or isn’t cleaned as regularly) may lead to higher species numbers alone, even with disturbance.
The second factor is that bacteria may not be similar to the organisms studied in these previous works. Most of the works mentioned previously involve larger scale organisms. However, bacteria are extremely small, to the point where they could be completely understood to be more affected by the microhabitat than the large-scale habitats which affect the plants and birds mentioned previously. For example, proximity to a water source might not be as important to bacteria as moisture levels in the environment. Another case is that bacteria do have tendencies to congregate in moist parts of the body, and will vary dependent upon that. In one study, 165 different bacterial species were found on individuals’ bodies in various moist body parts (Ling et al, 2013). This could be reflective that the bacteria numbers found on the skin were more likely suited to those regions specifically, and lowered due to bacteria primarily needing some form of regular moisture. The other portion of this is that bacteria show a different form of assemblages than many other habitats, since they don’t act on trophic levels. One study found over 113 species, but only 10 species accounted for 90% of the bacteria on the skin (Pennisi, 2008). This could suggest that rather than having more complex trophic levels (which accounts for biodiversity in multicellular organisms), bacteria could have much more unbalanced and different compositions in terms of communities, and may be more prone to competitive exclusions in those communities. This also could explain how the four sites saw larger numbers of seven of the species.
Another potential difference is the role of dispersal in this study. In many cases, the differences in sites could also be explained through some dispersal effects, as could the species numbers. As mentioned before, bacteria are often disturbed by events such as their habitat coming into contact with different objects or other people. This also acts as a way for passive dispersal of these bacteria, and could potentially explain why so many similar bacteria were found in all sample types. It could also explain why bacteria were so low in areas where they were predicted to be more abundant. Also, it has been observed that bacteria could use low salinity regions as a way to disperse into larger saline lakes (Engstrom-Ost et al). This could also be indicative that bacteria may actually use sweat as a passive way of dispersal, making it so that when athletes or other wipe off sweat, it would result in the movement of the bacteria.
Sweat could still be used as a model for flooding, and there are several options with which this study could be improved. The first option would be to repeat the study with higher sample numbers and with a controlled environment rather than one that is variable. If this were the case, then it would also be suggested to perform this with controlled exercise patterns of the subjects, making sure that the regular exercise of the athlete is steady and that non-athletes did not unintentionally sweat. Another future possibility for improvement of this study is using the sweating phenomenon to observe the movement of bacteria, such as having samples be taken prior to and after an intense sweating time period. Another option would be to use people living in temperate climates and people living in tropical climates instead of athletes and non-athletes; primarily since people in tropical climates will sweat more regularly than in temperate climates. This may also be useful to see if the latitudinal diversity gradient also applies to molecular forms.
Becker M.H., Walke J.B., Cikanek S., Savage A.E., Mattheus N., Santiago C.N., Minbiole K.P.C., Harris R.N., Belden L.K., & Gratwicke B. (2015) Composition of symbiotic bacteria predicts survival in Panamanian golden frogs infected with a lethal fungus. Proceedings of the Royal Society B: Biological Sciences, 282, 1–10.
Brawn J.D., Robinson S.K., & Thompson F.R. (2001) The role of disturbance in the ecology and conservation of birds. Annual Review of Ecology & Systematics, 32, 251–277.
Brotherton S. & Joyce C. (2015) Extreme climate events and wet Grasslands: plant traits for ecological resiliance. Hydrobiologia, 750, 229–243.
Callewaert C., Hutapea P., Wiele T., & Boom N. (2014) Deodorants and antiperspirants affect the axillary bacterial community. Archives of Dermatological Research, 306, 701–710.
Capen D.E. & Lent R.A. (1995) Effects of small-scale habitat disturbance on the ecology of breeding birds in a Vermont (USA) hardwood forest. Ecography, 18, 97–108.
Clark K.E., West A.J., Hilton R.G., Asner G.P., Quesada C.A., Silman M.R., Saatchi S.S., Farfan-Rios W., Martin R.E., Horwath A.B., Hallady K., New M., & Malhi Y. (2016) Storm-triggered landslides in the Peruvian Andies and implications for topography, carbon cycles and biodiversity. Earth Surface Dynamics, 4, 47–70.
Dzwonko Z. & Loster S. (1997) Effects of dominant trees and anthropogenic disturbances on species richness and floristic composition of secondary communities in southern Poland. Journal of Applied Ecology, 34, 861–870.
Engstrom-Ost J., Rasic I.S., Brutemark A., Rancken R., Simic G.S., & Laugen A.T. (2015) Can Cylindrospermopsis raciborskii invade the Baltic Sea? Environmental Reviews, 23, 161–169.
Gonzalez-Braojos S., Vela A., Ruiz-de-Castaneda R., Briones V., Cantarero A., & Moreno J. (2015) Bacteria on nestling skin in relation to growth in pied flycatchers. Journal of Ornithology, 156, 327–330.
Karrenberg S., Edwards P.J., & Kollman J. (2002) The life history of Salicaceae living in the active zone of floodplains. Freshwater Biology, 47, 733–748.
Li K., Bihan M., & Methe B.A. (2013) Analyses of the stability and core taxonomic memberships of the human microbiome. PLoS ONE, 8, 1–25.
Matlack G. & Leu N. (2007) Persistence of dispersal-limited species in structured dynamic landscapes. Ecosystems, 10, 1287–1298.
Park S.T., Collingwood A.M., St-Hilaire S., & Sheridan P.P. (2014) Inhibition of Batrachochytrium dendrobatidis caused by bacteria isolated from the skin of boreal toads, Anaxyrus (Bufo) boreas boreas, from Grand Teton National Park, Wyoming, USA . Microbiology Insights, 7, 1–8.
Pennisi E. (2008) Bacteria are picky about their homes on human skin. Science, 320, 1001.
Wollgast S., Lenz M., Wahl M., & Mollis M. (2008) Effects of regular and irregular temporal patterns of disturbance on biomass accrual and species composition of a subtidal hard-bottom assemblage. Helgoland Marine Research, 62, 309–319.
Table 1: Species list showing which sample types each species occurred on. (AE=Athlete Elbow, NE=Non-athlete Elbow, AN=Athlete Neck, NN=Non-athlete
Figure 1: Comparison of athletes and non-athletes in terms of similarity (1A), mean species richness (1B) and mean abundance (1C).
Figure 2: A comparison of the two body parts elbows and necks separately. The first graphs compare elbows of athletes and non-athletes in terms of mean species number (2A) or in terms of means abundance (2B). The second two graphs compare athlete necks to non-athlete necks based on mean species richness (2C) or mean abundance (2D)
Figure 3: A comparison of all four sample types in terms of mean species number (3A) and mean abundance (3B). For clarification, AE is athlete elbows, NE is non-athlete elbows, AN is athlete necks and NN is non-athlete necks.
Figure 4: Incidence (4A) and abundance (4B) across all species
Eukaryon is published by students at Lake Forest College, who are solely responsible for its content. The views expressed in Eukaryon do not necessarily reflect those of the College.
Articles published within Eukaryon should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.