Introduction

The biogeographical phenomenon known as the latitudinal diversity gradient is defined as the gradual decrease of species diversity when one moves away from the tropics; this gradient has been observed since the age of dinosaurs (MacArthur, 1965; Pianka, 1989; Mittelbach et al., 2007; Lomolino et al., 2006; Janzen, 1967; Hillebrand, 2004). With this phenomenon, it is concluded that the tropics hold the greatest number of species on Earth (Lomolino et al., 2006). A number of scientists have attempted to tackle the question of what variables and factors drive this gradient.

Pianka (1989) proposed a number of variables that potentially have a driving force in this pattern, which include climatic stability within seasons, spatial heterogeneity, competition, predation, productivity, and time. MacArthur (1965) also proposed very similar explanations to what drives this pattern. Focusing in on climate, Pianka (1989) proposed the time hypothesis where the duration of ‘same-ness’ – how long over a number of years that the tropics have been stable – is a major driving force that allows the tropics to harvest all these species. As mentioned above, the climatic stability hypothesis focuses on minimal seasonal variation to explain the pattern. Janzen (1967) observed that in the tropics, Central America, the “temperature regimes are generally more uniform than in North America” being a temperate region.

Just as the earth is an organisms’ playground, Homo sapiens are a microbes’ playground. The relationship between microbes and humans is an interesting one, because as beneficial as microbes can be to us, they can also be as dangerous and infectious (Importance of Microbes). The human body is home to trillions of diverse microorganisms, which make it a great model system that enable us to apply biogeographical principles on a much smaller scale (Importance of Microbes). Microbe communities are greatly influenced by a wide range of factors such as climate, body location, age, sex, medications, soaps and detergents, etc. (Roth & James, 1988; Fierer et al., 2008). However, one can control for these variables to some extent.

The latitudinal diversity gradient has been widely studied across different systems. Here, we apply this biogeographical phenomenon to the human microbial system to investigate if this theory holds true to the microbial human system. First, we hypothesize that the microbe samples from the axilla (i.e. underarm) will result in higher species diversity, richness, and abundance, compared to the microbe samples from the antecubital region (i.e. the inner elbow). Second, participants who use deodorant will have higher microbe species diversity, richness, and abundance, compared to participants who use antiperspirant. Since antiperspirant contains aluminum that blocks the pores, the use of this product would change the humidity of the environment, causing it to be drier than deodorant users where deodorant controls for the odor by neutralizing the odor-causing bacteria (Jones, 2013). Both products do have an effect on microbe communities, yielding different differences. The underarm is a region with increased temperature and humidity level that allow a greater range of microbes to be colonized, in comparison to the upper arm region, which is a drier environment and has a lower microbial colonization count (Roth & James, 1988). In addition, the axilla region harvests one of the densest skin populations of microbes, compared to other skin regions (Taylor et al., 2003).

In theory, the underarm microenvironment is analogous to the tropics, because of its stable environment of warm temperatures and relatively high humidity. On the other hand, the inner elbow microenvironment is analogous to a temperate region due to higher climatic variations and lower humidity relative to the underarm.

MATERIALS AND METHODS

Sampling

We used a paired design test to analyze the differences between microbial species richness, abundance, and diversity between the underarm and the inner elbow of each participant. In an attempt to retrieve samples immediately following a change in temperature, participants were asked to be sampled as they walked into the Mohr Student Center or Moore Hall ensuring exposure to a colder environment (outside) then entering into a warmer environment (indoors). To control for disturbances of each region of interest, only participants who were wearing long sleeve shirts were asked to be sampled. We obtained samples from thirty females, for a total of sixty swabs, which were all taken from the side of the body opposite from the participant’s dominant hand. Sampling took place on April 14-16, 2015 and the outside temperature were 44°C, 42°C, and 49°C respectively.

Sampling Criteria

Females were selected to participate in this study to control for underarm hair differences between males and females. We accepted to sample females who have shaven within the last twenty-four hours to further control for habitat complexity. Sampling took place at night to ensure that participants were relatively dirty assuming that they had at least four hours since their last shower and application of underarm product. All prospected females were asked these questions and if they did not fit the criteria were excluded. Since the chemical differences between the deodorant and antiperspirant could result in microbial differences, participant’s usage of either product was recorded.

I performed all the samples on the female participants, while Peter Nesper asked and recorded all the additional information. While using latex gloves, I retained the same swabbing technique across all participants. I placed the cotton swab on the inner elbow or in the underarm and rotated the swab in the same position for about ten seconds. Then, the swab was bent at the place opposite of what side I used to mark which side contains the microbes and then placed in a labeled sandwich bag. All participants remained anonymous by labeling the samples 01-30 and labeling the underarm sample as ‘A’ and the inner elbow sample as ‘K’.

Microbial Growth

Swabs were taken out of the bag and the side that contained the microbes was thoroughly mixed with PBS in a 0.5 mL microfuge tube. The tube was shaken and 50 μL was pipetted into a second microfuge tube of 950 μL of PBS. The second microfuge tube was mixed thoroughly and 50 μL was pipetted onto a labeled trypticase soy agar (TSA) plate. Several sterile glass beads were used to spread the suspension across the surface of the agar. Used beads were dumped into a beaker. All TSA plates were labeled with the same identification from sampling. TSA plates were placed in a 36°C incubator for a total of five days.

TABLE 1. Visual description of morphospecies. Morphospecies 0 (TTMS) is potentially a possible contaminant and thus why it is labeled 0 and not with a letter.

Morphospecies Identification

To quantify species richness abundance and diversity across all our samples, we characterized and grouped the microbe colonies into morphospecies based on similar morphology such as texture of surface, color, edge growth pattern, and other obvious morphological traits that were distinguishable. A dissecting microscope was used to visualize the colonies’ morphology. In addition, we took pictures of all the plates with an iPhones 5S to refer to any plate when needed. Refer to Table 1 for morphospecies descriptions.

Richness, Abundance, and Diversity

Species richness and abundance was calculated for each sample. A number of samples did not have any microbial growth and therefore, richness and abundance were recorded as zero. Diversity was calculated using the Shannon-Weiner diversity index. However, plates that did not have any microbial growth and their pair were automatically excluded from the diversity calculation since a value of ‘0’ in the Shannon-Weiner diversity index does signify one species present. Diversity was calculated for only fifteen pairs of samples.

A paired sample t-test was run to compare the means of richness, abundance, diversity between the inner elbow colonies and the underarm colonies using IBM SPSS Statistics software. Since our data was skewed by morphospecies 0 (TTMS), the raw dataset was log-transformed and the same statistics were run on the log dataset. By taking another look at TTMS and its abundance, we are very confident that this might have been a contaminant in the TSA plates. Out of the ten paired samples (20 plates) that were plated out on the second night of sampling, fifteen plates had the TTMS with colonies over 700 per plate. In addition, the belly button biodiversity website provides a large variety of microbes that they grew from their project and not any of those microbes share the same morphological traits as TTMS (Belly Button biodiversity showcase). With this potential contaminant, another set of statistics were run excluding TTMS.

Pertaining to our second hypothesis, an independent t-test was run to compare the means of species richness, abundance, and diversity between antiperspirant and deodorant from all the underarm samples using SPSS. We had a total of five females who use antiperspirant, twenty-three females who use deodorant, one female who uses baking soda, and one female who does not use any product. The final two were excluded from this analysis. Since diversity values were only calculated for fifteen samples, out of those fifteen, one sample was labeled as antiperspirant and the rest as deodorant. Overall, the diversity analysis had a very weak sample size in this case and when comparing inner elbow to underarm. Lastly, the statistics were run on all three datasets (i.e. raw, no TTMS, log).

RESULTS

Comparison between underarm and inner elbow

As a general trend, mean species richness, species abundance, and species diversity was higher for the underarm microbe samples, compared to the inner elbow microbe samples across all datasets (Fig. 1, 2, 3). However, mean differences were not significant (Table 2). The log transformed abundance means had the lowest p-value of 0.084 (t29 = 1.787)

FIigure 1. A) The mean species richness plotted for underarm and inner elbow for the raw dataset. Error bars represent standard error. Mean differences were not significant (Table 2). B) Means were plotted for the No TTMS dataset and were not significantly different (P > 0.05) (Table 2).

Figure 2. The mean species abundance plotted for underarm and inner elbow for the raw, no TTMS, and log datasets. Error bars represent standard error. Mean differences were not significant (P > 0.05) (Table 2).

TABLE 2. Paired sample t-test statistical output for the means comparison between underarm colonies and inner elbow colonies of species richness, species abundance, and species diversity. There was no statistical significance ( P > 0.05).

Figure 3. The mean species diversity plotted for underarm and inner elbow for the raw, no TTMS, and log datasets. Error bars represent standard error. Mean differences were not significant (P > 0.05) (Table 2).

FIGURE 4. The mean species richness plotted for antiperspirant and deodorant use in the underarms for the raw and no TTMS datasets. Error bars represent standard error. Mean differences were not significant (P > 0.05) (Table 3).

TABLE 3. Independent sample t-test statistical output for the means comparison between antiperspirant and deodorant underarm samples for species richness, species abundance, and species diversity. There was no statistical significance ( P > 0.05).

FIGURE 5. The mean species abundance plotted for antiperspirant and deodorant use in the underarms for the raw, no TTMS, and log datasets. Error bars represent standard error. Mean differences were not significant (P > 0.05) (Table 3).

FIGURE 6. The mean species diversity plotted for antiperspirant and deodorant use in the underarms for the raw, no TTMS, and log datasets. Error bars represent standard error. Mean differences were not significant (P > 0.05) (Table 3). Plot A and C did not have standard error bars due to the calculation being done on one data point. Plot B had a diversity of 0 in the antiperspirant sample due to presence of only one morphospecies.

but it was not less than α which is set at 0.05 (Fig. 2C, Table 2).

Comparison between antiperspirant and deodorant

Across all underarm samples, the species richness, abundance, and diversity were higher in the deodorant samples, compared to the antiperspirant samples (Fig. 5, 6, 7). However, the mean differences across all datasets were not significant (p > 0.05) (Table 3). For the diversity calculations for antiperspirant, there was insufficient amount of data to give credible results (Fig. 6).

Characterizing Morphospecies

Morphospecies distribution varies across all the plates and varies between the underarm environment and the inner elbow environment.

FIGURE 7. A) Percent occurrence of morphospecies was plotted as the number of plates the morphospecies is present over the total number of plates possible (30). Percent occurrence was sorted by underarm percentages. B) Percent abundance of morphospecies was calculated by the total number of colonies across all thirty plates in one morphospecies divided by the total number of all colonies across all thirty plates.

Morphospecies 0 (TTMS) was excluded from this analysis due to the gathering evidence that it was a contaminant. Morphospecies A was most common and most abundant in both environments (Fig. 7A, B). Morphospecies B, C, E, and H were common in both environments, but not abundant. Only pertaining to the inner elbow, morphospecies I had an occurrence of about 8% across all plates, but had the second highest abundance of 45% (Fig. 7A, B). Overall, majority of the morphospecies had a low abundance of under 5% in both environments. The percent occurrence was relatively evenly distributed between the inner elbow and the underarm.

DISCUSSION

The first hypothesis was not statistically supported that species richness, abundance, and diversity for microbes would be higher in the underarm environment compared to the inner elbow environment (Fig. 1, 2, 3; Table 2). Pertaining to the second hypothesis, the microbe species richness, abundance, and diversity was not significantly different between the use of antiperspirant and deodorant (Fig. 4, 5, 6; Table 3).

The lack of significance can be due to a number of errors. When calculating the diversity index, half of the samples had no microbial growth and had to be excluded from the analysis. We do not know if the lack of growth was due to poor sampling methods or actual absence of microbes. Due to time constraints, the samples that did not have any growth were not re-plated. This would be an ideal procedure that should have occurred if more time was allowed. In addition, a larger sample size would be employed to enhance the credibility and reliability of this study. This potential error is consistent with both hypotheses.

Pertaining to the second hypothesis, a recording error most likely occurred when female participants were asked what underarm product they use. Some participants did not know the difference between the two products and some quickly responded with deodorant, but the majority of feminine underarm products are antiperspirants. This should have been enforced during sampling.

The lack of significance in our study is inconclusive with other relevant studies. McBride et al. (1976) carried out a more extensive study that was investigating the role of the environment on the microbial ecology of the skin. They sampled five different regions on the skin: hands, back, axillae, groin, and feet. There subjects came from three different environment conditions, which were 1) high humidity and high temperature, 2) high humidity and low temperature, and 3) low humidity and moderate temperature (McBride et al., 1976). One of their findings was that microbe populations were significantly greater in the high temperature and high humidity environment, compared to the moderate temperature and low humidity environment when looking at the back, axillae, and feet skin sites (McBride et al., 1976). This was also supported by Roth & James (1988) in their Microbe Ecology of the Skin review. The sole condition of either temperature or humidity does not have a significant effect on microbial growth (McBride et al., 1976; Roth & James, 1988). Even though this is not direct evidence that underarm environment will have a higher species composition than inner elbow environment, it does support that the environment differences, such as temperature and humidity, do have an effect on microbial growth, where you see increased species composition in the hotter and more humid environment. Taking into account the rest of their results, McBride et al. (1976) concluded that climatic fluctuations do have an impact on microbial growth, but it is not the major driving factor.

Environmental factors such as hospitalization, age, sex, race, socioeconomics, occupation, soaps and detergents, medication, and ultraviolet light all influence microbial growth on our skin and inside our bodies (Roth & James, 1988). There is a whole other dimension of factors that come from the physiological relationship between microbes and humans such as microbial adherence, the immune system, and characteristics of the microbes themselves (Roth & James, 1988). All these factors add to the complexity of the human microbial ecosystem, as also seen in the nature’s ecosystem. Pianka (1989) proposed six different factors that could be potential drivers for the latitudinal diversity gradient. This pattern is not easily observed when there could be a number of driving forces, which is also apparent with the microbe communities on human skin where only certain variables were controlled. In conclusion, it is important to realize that the latitudinal diversity gradient can potentially be driven by multiple factors operating on the system to yield such a bold pattern.

Acknowledgements

This study was supported by the Department of Biology at Lake Forest College. Special thanks to Dr. Sean Menke for his guidance and support and Beth Herbert, and her lab assistants, for preparing and providing the appropriate materials for carrying out this study. text article text article text article text article text article text article text artcle text article text article text article text article text article text article text article text article text article text.

References

Belly Button Biodiversity Showcase. Belly button biodiversity. Retrieved from http://navels.yourwildlife.org/

Fierer, N., Hamady, M., Lauber, C. L., & Knight, R. (2008). The influence of sex, handedness, and washing on the diversity of hand surface bacteria Proceedings of the National Academy of Sciences, 105(46), 17994-17999.

Hillebrand, H. (2004). On the generality of the latitudinal diversity gradient. The American Naturalist, 163(2), 192-211.

Importance of Microbes. Microbes.org. Retrieved from http://microbes.org/microbes- importance

Janzen, D. H. (1967). Why mountain passes are higher in the tropics. American Naturalist, 233-249.

Jones, A., August 16, 2013). All Natural Deoderants and Antiperspirants. LiveStrong. Retrieved from www.livestrong.com/article/276146 -all-natural-deodorants-antiperspirants/

Lomolino, M. V., Riddle, B. R., & Brown, J. H. (2006). Biogeography.Sunderland, MA: Sinauer Associates.

MacArthur, R. H. (1965). Patterns of species diversity. Biological reviews, 40(4), 510-533.

McBride, M. E., Duncan, W. C., & Knox, J. M. (1977). The environment and the microbial ecology of human skin. Applied and environmental microbiology, 33(3), 603-608.

Pianka, E. R. (1989). Latitudinal gradients in species diversity. Trends in Ecology & Evolution, 4(8), 223.

Roth, R. R., & James, W. D. (1988). Microbial ecology of the skin. Annual Reviews in Microbiology, 42(1), 441-464.

Taylor, D., Daulby, A., Grimshaw, S., James, G., Mercer, J., & Vaziri, S. (2003) Characterization of the microflora of the human axilla. International journal of cosmetic science, 25(3), 137-145.

 

Related Links:

• A comparison of species_edited.pdf (PDF)