Some hypotheses you cannot plan

As I was preparing for my big leap from the relative safety of postdochood into assistant professorship, I'd worked to construct myself some safety lines. I'd listened to people's advice and talked with my postdoc advisor and senior colleagues early on about which projects I could take with me when I left, and wrote a few small grant proposals to bridge the time until I got my new research program off the ground. My main export would be some bacterial isolates from the tropical forest soils of Puerto Rico, which had the ability to grow on lignin anaerobically as sole carbon source. One of the proposals I'd written got funded, with myself as the PI (yay!), though this was an in kind award, meaning no "real" money. But this award seemed just as good when paired with my start-up funds: proteomics, transcriptomics and metabolomics of my trusty isolates to determine the mechanism of lignin degradation during growth, courtesy of the Environmental Molecular Sciences Laboratory, a DOE user facility. 

These isolates were special not just because they were able to degrade lignin anaerobically, a function that is attributed to most fungi when grown aerobically but only a handful of known bacteria. In my efforts to contribute positively to the development of sustainable lignocellulosic biofuels, these isolates would be important in discovering enzymes to remove lignin from cellulose, making fuel production from glucose more efficient, as well as contributing mechanisms for our host biofuel-generating organism to withstand the harsh conditions of lignocellulose pretreatment. 

We had decided to look for anaerobic lignin degrading bacteria in the tropical soils because plant litter decomposition is exceedingly fast in these soils despite the frequent soil anoxia brought on by daily rain showers totaling more than four meters of precipitation per year. Our field site was located in the Luquillo Experimental Forest, which is also an NSF-sponsored Long-Term Ecological Research Station (LTER). The LTER support makes this forest a wonderful place to do research, where we were able to have access to clean dorms and good research facilities within a few kilometers of our field sites. All the travel and time dedicated to obtaining these isolates, considering that the initial colonies took months to grow, made them that much more special to me as my first real research project.

However, it was not the lignin-degrading abilities that were such a surprise to me for this project. We'd worked to construct growth conditions under defined minimal media to isolate the transcripts and proteins associated specifically with lignin degradation, and were lucky to get very clean data. I'd set out to look for assimilatory degradation pathways, that is, enzymes that would break the lignin into pieces, shuttle the pieces through central metabolic pathways, and result in increased biomass. We had good evidence that our organism grew to higher cell density with lignin added, and had assumed that it was the lignin that was contributing to the increased growth. 

One of the best features about omics data are the post-hoc tests, or hypothesis generating analyses. While we had our a priori tests and protein targets, I began to see in the data that there was dissimilatory as well as assimilatory lignin degradation. That is, my bug was eating and breathing lignin, making it possible to grow more efficiently to the tune of doubling the biomass under the same growth conditions! I think this was the first time for me that a completely new hypothesis was introduced from the data. This was also a satisfying research experience because it was the first independent research to come out of my new lab. Though if I had any regrets, it is that we did not go on the junkets in Puerto Rico that everyone assumed we did anyways. 

Copyright: © 2013 Kristen DeAngelis. The above content is licensed under the Creative Commons Attribution License (CC-BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.