DMSP synthesis in phytoplankton unravelled
Dimethylsulfoniopropionate (helpfully abbreviated to DMSP) is produced in huge quantities (estimated over 1 billion tonnes per year) by marine phytoplankton, seaweeds, corals and, as we recently found, bacteria as well. The precise function of DMSP in the organisms that produce it is not known although roles including osmoprotection, cryoprotection, oxidative stress protection, predator deterrence and removal of excess sulfur have been suggested. Once outside the cell, the sheer amount of carbon and sulfur available from this molecule make it a key nutrient, and many microorganisms in the marine environment have taken full advantage of this. Many of these microorganisms break down DMSP into the gas dimethylsulfide (DMS). DMS is a key player in the global sulfur cycle and has potential links to climate through its role in cloud formation over the oceans. DMS is sometimes referred to as the ‘smell of the seaside’ for the fact that this highly odorous gas is a major part of the smell we encounter when we visit the beach, but importantly it is also detectable by animals, such as seabirds and seals, which may use it to locate their food.
Figure. Roles of DMSP and DMS in the environment
The paper in Nature Microbiology is here: http://go.nature.com/2EY1Xu9
Following our noses, this is how we first got into the DMSP field - identifying genes for the DMSP lyase enzymes that bacteria use to break down DMSP to DMS. More recently, we have moved on to the synthesis of DMSP rather than its breakdown. Since the discovery of DMSP in 1948, studies on this compound have focussed on how much DMSP was produced in phytoplankton and under what conditions, and its abundance in the environment and links to the sulfur cycle and climate. However, in all those studies there was always something missing: genetic information on how the DMSP is made, much like the lack of genetic information on the DMSP lyases (prior to our work), mentioned above. Our finding that bacteria could also produce DMSP greatly aided this genetic work, and led to the identification of the first gene for DMSP synthesis, termed dsyB. This work was published (in Nature Microbiology) in 2017, and in turn directly led on to the work reported here on eukaryotic DMSP-producers.
By using the bacterial dsyB gene sequence, we were able to identify homologues, termed DSYB, in many phytoplankton. These DSYB genes encode functional DMSP synthesis enzymes catalysing, as with the bacterial DsyB enzyme, the key step in the DMSP synthesis pathway in these organisms. We believe that regulation of the DSYB gene may also be the point of control in the pathway that determines how much DMSP an organism makes, with regulation of DSYB expression by environmental factors, such as salinity and nitrogen availability, closely mirroring the DMSP content in our model lab strains.
Our knowledge of which organisms out there in the oceans are producing DMSP is limited, as is the amount of DMSP that each species/strain produces. There have been extensive studies that reveal the DMSP content of multiple members of different taxa but these are obviously a drop in the ocean (excuse the pun) compared to the number of species not tested or grown in the laboratory. Such studies also reveal significant variability in DMSP content, even within very closely related organisms at the genus/species level and certainly at the class level, so it is very difficult to predict how much DMSP a particular organism would produce or even if it produces DMSP at all. Having the sequences of key genes in bacterial and eukaryotic DMSP synthesis pathways will allow us to better predict: (i) which organisms are producing DMSP based on the presence of dsyB/DSYB sequences in their genomes, (ii) how many organisms are producing it in the environment based on dsyB/DSYB gene numbers in metagenome data, and (iii) how much they are producing based on the numbers of dsyB/DSYB transcripts in metatranscriptome data. We think that this is a very exciting step forward and has huge potential for improving estimates of DMSP production, and more accurately informing current sulfur cycling models.
The function(s) of DMSP in the diverse range of organisms that produce it (see above) is also something that has long been a question we would like to address. While DMSP can undoubtedly act as an osmoprotectant, its contribution to the cellular pool of compatible solutes is unlikely to be very significant in organisms that produce low concentrations. This suggests that DMSP might have another functions in such organisms. To our great frustration, we have been unable to identify a phenotype for the bacterial dsyB- mutant so we do not know what the function of DMSP is in bacteria, which are generally low DMSP producers. We have not, so far, obtained a DSYB mutant in any of our eukaryotic strains, but we have attempted to gain more clues about function by looking at the location of the DSYB protein and DMSP itself within the cell. Microscopy, coupled to both immunogold labelling and NanoSIMS, were used to identify the intracellular locations of DSYB and DMSP in the haptophyte Prymnesium parvum and this suggests a chloroplast, and possibly mitochondrial, location. These are sites with high amounts of reactive oxygen species so oxidative stress protection is a potential function for DMSP in this organism. Sadly, a definitive answer to the question of DMSP function still eludes us at this time.
Despite DMSP research being focussed on the eukaryotic producers, since bacterial DMSP production was only recently identified, it was perhaps a small victory for us, as long-term bacterial microbiologists, to discover that the DSYB gene, and therefore DMSP synthesis itself, likely originated in marine bacteria. Evolutionary analysis suggests that the dsyB gene may have passed from an alphaproteobacterium into the eukaryote phytoplankton, possibly by endosymbiosis or a more recent horizontal transfer event.
As for the well-travelled DSYB gene, this was a great journey for the people involved. Many skills were required from collaborators both old and new, including bacterial/algal microbiology, evolutionary biology, biochemistry, microscopy and bioinformatics. We think these various skills came together successfully, and we hope to continue these collaborations in the future.
The paper in Nature Microbiology is published here: http://rdcu.be/HQoD
The following authors of the paper contributed to writing this blog: Dr Andrew Curson, Ana Bermejo Martínez and Dr Jonathan Todd