From what little we do know, and as we learn more with better and cheaper sequencing technology, it is clear that within all the teeming diversity there are some important players in our bodies. From our perspective, there are the "good" bugs that produce vitamins, help us digest our food, train the immune system, and may even affect our physiological development; then there are the "bad" bugs that cause infections and ulcers (and, ahem... what about those pesky chocolate cravings?)
We might wonder then, who could be called the "ugly"? Perhaps they're the shady characters of our microflora - the ordinarily harmless bacteria that occasionally turn on us, emerging as pathogens. Understanding how a common commensal of our skin (like Staphylococcus aureus) can morph into a deadly antibiotic-resistant global epidemic (like MRSA) has obvious practical implications. Antibiotics are our last line of defense against bacterial infections and in recent years multiple-antibiotic resistant pathogens have proliferated. The question that begs to be answered is how - and where - does antibiotic resistance emerge in potential pathogens?
Whole-genome sequencing has lent some molecular insight to the "how" of this question, by showing that most resistance genes in pathogens bear the trademarks of lateral gene transfer. This indicates resistance genes commonly evolve in other bacteria but that pathogens fortuitously pick them up through mobilized DNA. As for the "where," a recent Science paper shows there is great potential for this to happen within our own bodies, among our numerous native bacteria.
Authors Sommer, Dantas and Church sought out antibiotic resistance genes within our own intestinal communities. They cataloged genes that we've already identified in clinical pathogens and identified novel genes, all of which may be available to potential pathogens as they percolate with the bacterial populations in our bodies.
As it turns out, our individual reservoir of functional antibiotic resistance genes is extensive, evolutionarily diverse, and (based on depth of sequencing as well as a sample size of merely two people) it is fairly safe to say that Sommer et al. only scratched the surface! This surprising and magnificent view is largely due to their inclusion of the unculturable but numerically dominant bacteria (through metagenomics) as well as culturable bacteria.
The authors isolated DNA from saliva and fecal samples of two healthy individuals. They cloned metagenomic DNA fragments into expression vectors which were transformed into E. coli hosts. The E. coli were then screened for resistance to 13 diverse antibiotics. Sequencing the resistant clones revealed an unprecedented number of evolutionarily distant, yet functional, antibitoic resistance genes. Most of the sequences came from the phyla Bacteroidetes and Firmicutes, which dominate our guts; however only 20% of the genes were closely related homologs (to, e.g., Bifidobacterium spp. or opportunistic Bacteroides fragilis). On average, the resistance genes in the natural population of the gut were only distantly (60.7% at the nucleotide level) related to the resistance genes we know from pathogenic isolates. One would have thought they would be more closely related, considering that the method selects resistance genes that function in E. coli (a member of the Proteobacteria, a phyla known for its many pathogens and a relative minority in the human gut community). This suggests there might be more of a barrier to gene transfer between the gut microflora and common pathogens then thought before.
The authors also used rich media and aerobic conditions to directly cultivate bacteria from fecal matter. Among these isolates, multiple resistance among cultured bacteria turned out to be way too common for comfort. The isolates from Subject One and Two were, on average, resistant to 9/13 and 5/13 antibiotics! And in contrast to the metagenomic survey that was dominated by Bacteroidetes and Firmicutes, these strains were primarily Proteobacteria with resistance genes that were closely related (or identical) to genes from clinical pathogens. Interestingly, in comparing the similarity of the two people sampled, 65% of their culture-derived genes were similar to one another, whereas only 10% of their metagenome-derived genes were similar to one another.
This study was unique in its functional metagenomic approach, but like other human microbiome studies preceding it, it confirmed that the bacterial communities each person hosts are incredibly unique. We're learning this diversity reflects, to different degrees, transmission from our parents and environment as well as dynamic community responses to selective pressures like diet and antibiotic regimes. I would love to know more detail about the two people sampled - was one a vegetarian? Did they grow up in the same place? Despite being healthy (defined here as not being on antibiotics within the past year), had "Subject One" spent a lot of time in hospitals? As we come to better understand the natural diversity of our microbial communities, how they respond to change, and what factors are significant in shaping species composition, there are tantalizing possibilities for eventually achieving more personalized medicine - and finally ferreting out who we might call the good, the bad and the ugly in our own, personal microbiomes.
Sommer MOA, Dantas G and GM Church (2009) Functional Characterization of the Antibiotic Resistance Reservoir in the Human Microflora. Science 325: 1128-1131
For more on the roles bacteria play in our body and how the human microbiome project is elucidating their roles in human health, read:
Turnbaugh PJ et al. (2007) The human microbiome project. Nature 449:804-810
**If you are specifically interested in the "chocolate-desiring" bugs you may be hosting, see:
Rezzi S. et al. (2007) Human Metabolic Phenotypes Linked Directly to Specific Dietary Preferences in Healthy Individuals. Journal of Proteome Research 6:4469-4477
Information about bacteria swapping genes (lateral gene transfer):
Ochman H, Lawrence JG and EA Groisman (2000) Lateral gene transfer and the nature of bacterial innovation. Nature 405:299-304
And lastly, for specific studies looking at differences in the gut microbiome of lean and obese people, or the effects of antibiotics, see:
Turbaugh PJ et al. (2009) A core gut microbiome in obese and lean twins. Nature 457:480-484
Dethlefsen L, Huse S, Sogin ML and DA Relman (2008) The pervasive effect of an antibiotic on the human gut microbiome, as revealed by deep 16S rRNA sequencing. PLoS Biol. 6(11):e280. doi10.1371/journal.pbio.0060280.