 |
| E. coli click for source. |
Everyone knows the last place you want to get stuck for a long time is a hospital. Hospitals are breeding grounds for bacteria. Disease-causing bacteria. Bacteria that are getting smarter. “Getting smarter” is a great euphemism for evolution. The same way we learn and adapt to new situations, bacterial pathogens adapt better to human hosts, develop ways to evade the immune response of their host, and become more resistant to antibiotics. (See MRSA) They do this simply by accumulating mutations in their genomes. The ones that are beneficial cause them to multiply, the ones that aren’t, well, those don’t do so well. That, my friends, is evolution.
Evolution is easy to see on this small scale because of a relatively small genome size (compared with say, mammals, but this isn’t necessarily indicative of organism complexity) and generation time. Given the right conditions, E. coli (your general run-of-the-mill bacterium) can double in 20 minutes. I can’t think of anything that I can do in 20 minutes. To put it in perspective, while you’re busy watching 30 Rock on Netflix, Your Favorite Bacterium has completed a generation time and is working on another. How’s that to make you feel lazy. Actually it shouldn’t. That has no bearing on your personal TV choices. However, because this bacterium is constantly churning out new generations, a random mutation that say, allows one bug to use oxygen more efficiently, and say, generate then 2x faster than the one without the mutation, when you’re speaking in time frames of twenty minutes, that makes a big difference to the bacteria whose sole purpose is to produce as much offspring as it possibly can. (Live long and prosper, etc.)
The scary part is that while scientists and doctors are at the metaphoric first floor of disease etiology, bacterial pathogens are racing up flights of stairs, using the elevator, and are miles ahead of us as fast as you can say “candidate gene”.
 |
| click for source |
The cool thing is, we’re starting to catch up. In the most recent issue of Nature Genetics Lieberman et al use some pretty interesting techniques to take a comprehensive look at bacterial evolution. High throughput sequencing has made it possible to actually trace the specific mutations driving the evolution of these bacterial pathogens.
Lieberman et al looked at the genetic adaptation of a single bacterial strain in multiple human hosts during the spread of an epidemic. Specifically, Burkholderia dolosa, which is a bacterial pathogen that is associated with cystic fibrosis. In the 1990s, 39 individuals with cystic fibrosis were infected, and bacteria was isolated from each, and routinely frozen. The researchers then followed each patient over the course of 16 years, isolating samples of B. dolosa along the way.
What scientists can do with sequencing data
Scientists sequenced the whole genome of every bacteria isolated. They focused their analysis on SNPs (single nucleotide polymorphisms), since other genomic variations (structural and mobile elements) are a little harder to interperet. They were able to identify that mutations were accumulating at about 2 SNPs per year, which is consistent with bacterial mutation-fixation rates (the rate that a mutation becomes "fixed"--or stable within that genome). Due to the nature of the study (at how many time points scientists had isolated bacteria) they were able to document enough genetic diversity within that single strain that they were able to look at the evolutionary relationships among the isolates. Because this is such an infectious bacteria, the amount of data collected allows them to trace the path of infection from person to person. How cool is that? Unless you’re the person that started it all, and then, oops, should have stayed home from work that day.
They were also able to look at the evolution at a gene level. They looked at two specific phenotypes: resistance to ciprofloxacin (which is frequently prescribed to individuals with cystic fibrosis), as well as the presentation of O-antigen repeats in the lipopolysaccharide of the bacterial outer membrane, which plays a key role in virulence.
What they found was that in each of these phenotypes, a single gene was implicated, and using the phylogenetic data, they were able to determine that multiple mutations present in an individual were accrued independent of each other.
Here’s where statistics comes in. Since all evolution is about is random mutations, in order to demonstrate that a selective pressure is at work, you need to show that what you found was happening more than just by chance. They found 561 independent mutational events in 304 genes. Assuming a neutral evolution model (everything happening by chance, no selective pressure), they expected to see these mutations randomly distributed among all of the 5,014 B. dolosa genes, and it would be rare for a gene to acquire more than a single mutation. Instead, a lot of genes contained multiple mutations; four in particular had more than ten.
There are of course sites of mutational bias. In humans, these are genetic loci including olfactory receptors, and genes contributing the immune system. Genes that just happen to acquire more mutations as a response to the environment. They ended up identifying that 17 of the genes with mutations were not mutational hotspots, but underwent adaptive evolution under the pressure of natural selection.
What we can learn from this
High throughput sequencing and computational analysis are finally tools that can allow us to delve into the world of ever-evolving disease-causing bacteria. In a clinical setting this has a powerful impact; being able to tailor antibiotics to a particular genotype of bacteria would be a more efficient way of fighting disease. Not to mention that we’re now able to look at evolution happening in real time. How's that for catching up?
 |
| click for source |
This post has been submitted for the NESCent 2011 ScienceOnline Blog Travel Award Contest.
