We just published a paper in BMC Genomics on the occurrence of genes coding for alkane hydroxylases in the genomes of psychrophilic (cold loving) bacteria. This paper does a couple of interesting things and fails to do several other interesting things. A little background before we get into that.
For several decades investigators have been interested in the ability of psychrophilic (cold loving) microbial communities to degrade the various components of crude oil, a process known as bioremediation. Because temperature has a direct impact on enzyme activity there is a concern that, at very low temperatures, the microbial community might be inhibited in its ability to undertake bioremediation. The benefits of bioremediation were seen in the response of the microbial community during the infamous Deepwater Horizon event in the Gulf of Mexico. Indigenous bacteria deep in the water column consumed much of the escaping oil before it had the opportunity to reach the surface, lowering (but not eliminating) the ecological impact of that event.
Despite the subtropical location of the Deepwater Horizon, the water temperature where much of this bioremediation took place is only 5 degrees C. This is not much warmer than summertime Arctic surface waters, and close to the point at which temperature is observed to start inhibiting microbial activity (see figure). What would happen if an Arctic drill rig had a similar accident? Would the microbial community be able to respond as effectively?
Taken from Kirchman et al. 2009. Bacterial growth rate, here a useful proxy for the activity of any system of bacterial enzymes, doesn’t change much with temperature until about 4 degrees. Below 4 degrees C enzyme activity seems to be inhibited most of the time. The devil is in the detail and we don’t understand the exceptions. Whether a marine microbial community in -1 degree C water could respond to a crude oil spill as effectively as a community at 5 degrees C is an open question.
To shed a small bit of light on this question we compared a population of psychrophile genomes to a taxonomically similar population of mesophiles (bacteria that live at approximately room temperature), searching both populations for genes coding for alkane hydroxylases. Alkane hydroxylases are enzymes involved in the degradation of alkanes, a major component of crude oil. We used a novel strategy to find putative alkane hydroxylase genes in these genomes. First, we used the excellent hmmscan tool in HMMER 3.0 to identify conserved domains in gene products that are also found in alkane hydroxylases. Then we used non-metric multidimensional scaling (NMDS) to find clusters of similar sequence (see figure).
Taken from Bowman and Deming, 2014. Each point in this NMDS plot of the FA_desaturase protein family is a protein from either the PFAM database (orange), the Uniprot database (black), a mesophile (red), or a psychrophile (blue). The distance between any two points on the plot is related to their sequence similarity. The Uniprot proteins are known alkane hydroxylases. Thus any red and blue points that fall close to black points are hypothetical alkane hydroxylases. You: What’s close? Me: How blue is the ocean? These questions are not without merit but we will not answer them today. The PFAM proteins are included only to provide “landscape” (proteinscape?), which helps to visualize clusters of proteins.
Surprisingly we found many putative alkane hydroxylase genes in both mesophile and psychrophile genomes that were not annotated as alkane hydroxylase. Typically these genes were annotated simply as “hydroxylase”, or something similarly nondescript. But here’s where our analysis falls a bit short. The critical next step is to synthesize these putative alkane hydroxylase genes, clone them into a vector, and express them in a model bacterium (probably E. coli, as I’m not aware of a transformable model psychrophile. Anyone know one?) to see if they are what we think they are. Unfortunately that’s a whole project by itself and one can be a PhD student for only so long (I didn’t start this project until pretty late in the game). It’s on the to do list and, in the meantime, perhaps another lab will pick it up and run with it…
Many thanks to the EPA STAR (Science To Achieve Results) Fellowship program for funding me while I worked on this and related projects. The EPA STAR program is sadly now defunct, a victim of congressional cost-cutting. As I understand it the program was perceived as duplicative of the larger NSF GRFP (Graduate Research Fellowship Program). Setting aside for a moment the vastly different missions of the EPA and NSF (the former being mission driven and the latter supporting basic research), suggesting that the programs are redundant is a bit like saying a thirsty person doesn’t need a second sip of water, because it duplicates their first. The NSF GSRFP funds a very small fraction of the graduate researchers who are in need of support.
