JXB Advance Access published online on February 16, 2008
Journal of Experimental Botany, doi:10.1093/jxb/erm264
SPECIAL ISSUE PAPER |
Molecular diversity of bacterial production of the climate-changing gas, dimethyl sulphide, a molecule that impinges on local and global symbioses
School of Biological Sciences, University of East Anglia, Norwich NR4 7TJ, UK
* To whom correspondence should be addressed. E-mail: a.johnston{at}uea.ac.uk
Received 30 May 2007; Revised 27 September 2007 Accepted 1 October 2007
 |
Abstract |
|---|
 Top Abstract IntroductionDistribution of the DddD...'Pick 'n Mix' arrangement...Bacterial DMSP lyase: myth...DMS production and signalling...ConclusionsReferences |
|---|
This paper describes the ddd genes that are involved in the production of the gas dimethyl sulphide from the substrate dimethylsulphoniopropionate (DMSP), an abundant molecule that is a stress protectant in many marine algae and a few genera of angiosperms. What is known of the arrangement of the ddd genes in different bacteria that can undertake this reaction is reviewed here, stressing the fact that these genes are probably subject to horizontal gene transfer and that the same functions (e.g. DMSP transport) may be accomplished by very different mechanisms. A surprising number of DMS-emitting bacteria are associated with the roots of higher plants, these including strains of Rhizobium and some rhizosphere bacteria in the genus Burkholderia. One newly identified strain that is predicted to make DMS is B. phymatum which is a highly unusual β-proteobacterium that forms N2-fixing nodules on some tropical legumes, in this case, the tree Machaerium lunatum, which inhabits mangroves. The importance of DMSP catabolism and DMS production is discussed, not only in terms of nutritional acquisition by the bacteria but also in a speculative scheme (the ‘messy eater’ model) in which the bacteria may make DMS as an info-chemical to attract other organisms, including invertebrates and other plankton.
Key words: Acyl CoA transferase, Burkholderia, CLAW hypothesis, dimethyl sulphide, dimethylsulphoniopropionate, Marinomonas, nitrogen fixation, Rhizobium, rhizosphere, root nodules, Spartina
|
Introduction |
|---|
|
TopAbstractIntroductionDistribution of the DddD...'Pick 'n Mix' arrangement...Bacterial DMSP lyase: myth...DMS production and signalling...ConclusionsReferences |
|---|
Perhaps the greatest symbiosis of all was enunciated in the form of James Lovelock's ‘Gaia hypothesis’ in which mutual benefits occur between communities of organisms and their respective collective environments (Lovelock, 1979). And, depending on one's scale of ambition, those communities may stretch to the entire biosphere, and the environment can encompass planet Earth. One definition of Gaia, from Lovelock (1999) himself is
‘:...a complex entity involving the Earth's biosphere, atmosphere, oceans, and soil, the totality constituting a feedback or cybernetic system which seeks an optimal physical and chemical environment for life on this planet’.
Although contentious, and sometimes extrapolated and interpolated to breaking point and beyond, the Gaia hypothesis has been, and continues to be, a focus of much debate on how some forms of biological homeostasis are maintained – or not – on this planet.
This all-pervading theory required at least some hard data as a means of supporting, though not necessarily proving, its import. And that came in 1987, stemming from another of Lovelock's incisive discoveries, namely that the gas dimethyl sulphide (DMS) was the predominant form of sulphur that was emitted from the seas to the atmosphere and thence to the land (Lovelock et al., 1972). In what is known as the ‘CLAW hypothesis’ (Charlson et al., 1987), it was proposed that the rates of DMS emissions had a homeostatic effect on global cloud cover, and hence on climate. The line of argument was: the greater the temperature, the higher the level of biogenically made DMS that would be liberated from the oceans to the air, and, since oxidation products of DMS can act as nuclei for cloud formation, this would cut the levels of solar radiation. This, in turn, would drop the temperature, and thus the DMS emission rates, completing a biosphere-mediated feedback loop.
Remarkably, DMS has other, very different biological effects, since it is a chemo-attractant for animals as diverse as the hugely abundant copepods, which are zooplanktonic crustaceans (Steinke et al., 2006) and some oceanic seabirds, including petrels and shearwaters (Nevitt and Bonadonna, 2005).
The source of DMS in the sea and its margins is dimethylsulphoniopropionate (DMSP, or dimethylpropiothetin), a molecule that occurs worldwide in prodigious amounts (109 tonnes or more) in marine algae, where it acts as a compatible solute to counter various stresses, including osmotic, oxidative, and potential damage by UV light (Sunda et al., 2002). When released from marine algae, following grazing or viral attack, the DMSP becomes available for subsequent microbial catabolic conversions, some of which release DMS. In addition to marine algae (including microscopic and some seaweed macroalgae), at least three different types of land angiosperms can synthesize DMSP. But, the roles of DMSP in these unrelated species, the beach sunflower Wollastonia biflora, the salt marsh grass Spartina, and some lines of Saccharum (sugar cane), are unknown (Otte et al., 2004).
The generally held view is that DMS is produced by the enzyme ‘DMSP lyase’, which cleaves the substrate to release DMS, a proton and acrylate (Fig. 1). As far back as 1956, extracts of the red alga Polysiphonia lanosa were found to have such enzymatic activity (Anderson and Cantoni, 1956), and DMSP lyases have been described, in various levels of detail, in a range of marine eukaryotes (Steinke et al., 1998; Bentley and Chasteen, 2004).
|
In addition, different types of marine bacteria can catabolize DMSP; in fact, bacteria are responsible for most of the DMSP catabolism in the seas (Kiene and Taylor, 1988; Zubkov et al., 2001). Certainly, huge numbers of marine bacteria, can import DMSP (Gonzalez et al., 2000; Malmstrom et al., 2004; Vila et al., 2004; Mou et al., 2005; Pinhassi et al., 2005).
Tracer experiments indicate that, globally, the major route of bacterially-mediated DMSP catabolism is by a demethylation process (Kiene et al., 1999). Although this predominates in terms of the degradation of DMSP and hence of marine carbon and sulphur biotransformations, it does not liberate DMS, so does not contribute to the atmospheric changes alluded to above.
Several bacterial strains grow well on DMSP as the sole carbon source in free-living culture (Kiene, 1990; Visscher and Taylor, 1994; de Souza and Yoch, 1995a, b; van der Maarel et al., 1996; Ansede et al., 1999, 2001a; Gonzalez et al., 1999, 2000, 2003) and in some cases, this is accompanied by DMS release. This has been described for several strains and species of the 
-proteobacterial Roseobacter family (Moran et al., 2003; Malmstrom et al., 2004) and of the 
-proteobacteria Marinomonas, Psychrobacter, and Vibrio (Ansede et al., 2001a). This phenotype is termed ‘Ddd+’ (DMSP-dependent DMS).
There have been a few reports on bacterial enzyme(s) that result in the emission of DMS through DMSP catabolism. Thus, DMS-emitting strains of an Alcaligenes-like bacterium and of Pseudomonas doudoroffii yielded enzymes with DMS lyase-like activity (de Souza and Yoch, 1995a, b; Yoch et al.,1997), which appeared to be at the cell surface in the former species (Yoch, 2002). Further, van der Maarel et al. (1996) isolated a strain of Desulfovibrio acrylicus that grew on DMSP and on acrylate as carbon sources and which had an enzyme that cleaved DMSP, releasing DMS.
Given the importance of DMSP catabolism, in terms of the huge flux through the marine carbon and sulphur cycles, and the effects of DMS on climate and as an info-chemical, it is surprising that so little is known of the genetics of the process, especially since several DMSP-degrading bacteria are amenable to laboratory-based culture and the genome sequences of some of them are known (Moran et al., 2004; http://www.roseobase.org/).
It seemed likely that a way into the genes for DMSP lyase might be achieved by reverse genetics, since de Souza and Yoch (1996) described the N-terminal sequences of two bacterial DMSP lyases, these being NH3: AQFQHQDDVKPAAISAEEGKGKLVDEQFQEAQKNNEAL in Alcaligenes and NH3: AQFQSQDDVKPASIDAWSGL in Pseudomonas doudoroffii. However, they did not exploit these data and so no mutants in the corresponding genes have been made (see also below). Recently, however, a gene (dmdA) that is required for the demethylation pathway was described by Howard et al. (2006) as has a series of ddd (DMSP-dependent DMS) genes involved in a DMS-emitting pathway (Todd et al., 2007) in different bacterial genera.
In the marine Roseobacter species Silicibacter pomeroyi DSS-3 (Moran et al., 2004), dmdA encodes a demethylase, the first dedicated step in the demethylation pathway (Howard et al., 2006), yielding the initial catabolite methylmercaptopropionate (Fig. 1). Homologues of DmdA are found in other bacteria with DMSP-catabolizing ability and occur frequently among the gene sequences in metagenomic clone libraries made directly from marine bacteria, harvested from their natural environments (Venter et al., 2004; Tringe et al., 2005). Of particular interest, Pelagibacter ubique (SAR11), the most populous bacteria on the planet, has a functional DMSP demethylase (Howard et al., 2006).
Turning to the ddd genes, these were identified in a strain of Marinomonas that was isolated from the rhizosphere of the DMSP-producing angiosperm Spartina anglica (Todd et al., 2007). The results obtained in that study were surprising, on several counts. First, a single gene, termed dddD, was sufficient to confer a Ddd+ phenotype to Escherichia coli, providing that it was expressed from a vector promoter. However, the deduced DddD gene product was not a lyase, as might have been expected. Rather, its closest homologue with known function is CaiB of E. coli, a Class III acyl CoA transferase that donates acetyl CoA to carnitine when that amino acid is used as a terminal electron acceptor in anaerobic conditions. CaiB is a homo-dimer (Rangarajan et al., 2005), but DddD is 
twice the size of CaiB, and has a tandemly repeated domain structure that likely forms an ‘intramolecular dimer’. Nevertheless, by analogy with the function of CaiB, it is predicted that DddD adds an acyl CoA moiety to DMSP, prior to its subsequent cleavage and release of DMS (Todd et al., 2007; Fig. 1).
As expected, other strains of Marinomonas and Marinomonas-like strains had proteins that were >90% identical to DddD in ‘our’ strain, MWYL1. Furthermore, DddD homologues occurred in Silicibacter pomeroyi DSS-3 and Sagittula stellata E37. These are both in the more distantly related -proteobacteria, but these observations are consistent with the known Ddd+ phenotypes of these bacteria (see below).
What was more unexpected was the presence of DddD homologues in two ‘terrestrial’ strains of bacteria that were unsuspected of DMSP catabolism. Perhaps significantly, both of these strains interact with higher plants. One of these was the -proteobacterial N2-fixing symbiont Rhizobium sp. NGR234. Isolated by Trinick (1973), NGR234 was already known to be highly unusual since it nodulates a very wide range of taxonomically distinct legumes (most rhizobia have narrow host-ranges, some of them exquisitely so). More striking yet, it is still the only known rhizobial strain that forms nodules on non-legume plants; indeed NGR234 was isolated from a member of the Elm family, Parasponium, in New Guinea. The second unexpected strain was the β-proteobacterium Burkholderia cepacia AMMD, a very widespread member of the rhizosphere community of angiosperms, which is sold as a biocontrol agent to augment plant vigour (Coenye and Vandamme, 2003).
It was shown directly that NGR234 and AMMD (but not other strains of Rhizobium or Burkholderia that lacked dddD) emitted DMS when exposed to DMSP, though, puzzlingly, neither strain could fully catabolize this substrate and so could not use it as sole carbon source. Nevertheless, their individual dddD genes, when cloned in an expression vector plasmid, conferred a Ddd+ phenotype to E. coli (Todd et al., 2007).
It has been known for some time that the Ddd+ phenotype can be induced by pregrowth of bacteria on the DMSP substrate, and, at least in some cases, by acrylate, acrylamide or β-hydroxypropionate (De Souza and Yoch, 1995a, b; Yoch et al., 1997; Ansede et al., 1999, 2001b; Jonkers et al., 1999). It was noted that the Ddd+ phenotypes of the Marinomonas, Rhizobium, and Burkholderia strains described above were also induced by DMSP, but not by acrylate (Todd et al., 2007 and see below). By making a transcriptional fusion between dddD of Marinomonas and lacZ in wide host-range promoter-probe plasmids, it was shown that this induction was at the initiation of transcription and that this was mediated by the regulatory gene dddR, which is transcribed divergently from dddD (see below). Further, DddR– mutants of Marinomonas were defective in dddD expression and did not make DMS. Consistent with this regulatory role, DddR is in the LysR family of transcriptional regulators (Todd et al., 2007).
Here, further observations are presented on the molecular basis of DMSP-dependent DMS production in bacteria. The properties of the ‘model’ strain Marinomonas MWYL1 are compared and contrasted with those in other bacterial lineages that we have begun to examine experimentally and in silico. In keeping with the theme of this volume, bacterial DMS production is related to its possible role in the interactions between prokaryotes and higher organisms.
|
Distribution of the DddD protein in different bacterial lineages |
|---|
|
TopAbstractIntroductionDistribution of the DddD...'Pick 'n Mix' arrangement...Bacterial DMSP lyase: myth...DMS production and signalling...ConclusionsReferences |
|---|
Figure 2 shows the relatedness tree of all the known homologues (as of March, 2007) of the DddD proteins, all of which had the characteristic intragenic duplication of the CaiB-like acyl CoA transferases (see above). DddD of Marinomonas MWYL1 is (not surprisingly) very closely related to the corresponding proteins of Marinomonas MED121, Marinobacter ELB17, and the marine
-proteobacterium HTCC2207 (Group ‘A’ in Fig. 2). However, as mentioned above, the taxonomic positions of several other bacteria with DddD homologues that are 60% identical to that of Marinomonas (group B in Fig. 2) are surprising. One of them, S. stellata E37 is a known Ddd+ strain (Gonzalez et al., 1997), but the DddD+ phenotypes of the other two, Rhizobium NGR234 and B. cepacia AMMD, were wholly unanticipated. Close Ddd homologues have now been observed in two other Burkholderia strains (Fig. 2). These were B. ambifaria MC40-6 (locus tag BamMC406DRAFT_4307) which, like B. cepacia AMMD is a biocontrol agent against plant diseases, although (disconcertingly) it also occurs in the lungs of cystic fibrosis patients (Coenye et al., 2001). The other strain is B. phymatum STM815 (BphyDRAFT_5496), which was obtained from a N2-fixing root nodule on the tropical legume shrub, Machaerium lunatum (Moulin et al., 2001; http://zipcodezoo.com/Plants/M/Machaerium_lunatum.asp). Thus, STM815 is one of the recently found root nodule ‘rhizobia’ that are not in the conventional -proteobacteria lineage that contains the familiar genera Bradyrhizobium, Mesorhizobium, Rhizobium, and Sinorhizobium.
|
In addition to S. stellata, two other Roseobacters, namely S. pomeroyi DSS-3 and Dinoroseobacter shibae DFL 12 (Biebl et al., 2005) have DddD homologues, but these (in Group C in Fig. 2) are only
40% identical to Marinomonas DddD. S. pomeroyi DSS-3 is known to be Ddd+ (Moran et al., 2004) and it can also catabolize DMSP via demethylation, using the DmdA demethylase (Howard et al., 2006; see above). As well as the DddD homologues in known bacterial species, there is one close match (protein EAK4524.1) in the large Sargasso Sea metagenomic database (Venter et al., 2004).
The distribution of DddD homologues in taxonomically distinct lineages (, β, , proteobacteria) strongly suggests that dddD is part of the auxiliary gene pool, being transferred between distantly related strains by horizontal gene transfer (HGT). Further circumstantial evidence for this notion is that dddD of Rhizobium sp. NGR234 is on a large resident plasmid, not the chromosome. Also, dddD occurs sporadically, being present in some bacteria, but not in very closely related strains; for example, in the two sequenced genomes of S. pomeroyi, DSS-3 has dddD, but strain TM1040 (see http://www.roseobase.org/roseo/tm1040.html) does not.
|
‘Pick ’n Mix' arrangement of genes near dddD in different DddD+ bacteria |
|---|
|
TopAbstractIntroductionDistribution of the DddD...'Pick 'n Mix' arrangement...Bacterial DMSP lyase: myth...DMS production and signalling...ConclusionsReferences |
|---|
The names and deduced functions of the various ddd genes described below are listed in Table 1 and their locations in the ddd gene clusters of different bacteria are shown in Fig. 3.
|
|
In Marinomonas MWYL1, dddD is transcribed divergently from a predicted 4-gene operon, dddTBCR (Todd et al., 2007; Fig. 3). The predicted DddT protein is a transporter of the BCCT (betaine, choline, carnitine transporter) type, an inner membrane protein that imports molecules that resemble DMSP. It is likely, therefore, that DddT is a DMSP importer (Todd et al., 2007). The functions of the DddB and DddC proteins are unknown, but their sequences suggest that they are involved in oxido-reductive functions, so may modify DMSP either before or after the addition of the acyl CoA moiety (Todd et al., 2007).
This gene arrangement of dddD-dddTBCR also occurs in Marinomonas MED121 and in Marinobacter. However, in the Marinomonas-like strain HTCC2207, there is an extra gene between dddC and dddR, whose product is in the IclR family of transcriptional regulators (Cortay et al., 1991). Given that DddR responds to DMSP, perhaps this ‘extra’ regulator in strain HTCC2207 allows it to respond to other (unknown) co-inducers or co-repressors.
The genes that abut dddD in the more distantly related B. cepacia AMMD, Rhizobium NGR234 and Sagittula stellata reveal a mix of similarities and differences compared with those in the Marinomonas group (Fig. 3).
Todd et al. (2007) noted that dddD of B. cepacia AMMD is adjacent to a gene with striking similarity to Marinomonas dddR; indeed, DddR proteins of both strains could each activate the expression of the dddD gene of the other in response to the co-inducer DMSP (Todd et al., 2007). There are no genes predicted to encode DMSP transporters near dddD of B. cepacia AMMD (or in the other DddD-containing strains of Burkholderia), nor are there any genes that correspond, in sequence and/or deduced function, to Marinomonas dddB or dddC.
By contrast, dddD of Rhizobium NGR234 is located between two sets of genes whose deduced functions resemble those near dddD of Marinomonas. Thus, 3’ of Rhizobium NGR234 dddD is a predicted operon, dddEFG, that specifies a transporter whose nearest homologues are involved in importing betaine-like molecules. The DddEFG proteins form an ABC type transporter, which is mechanistically totally different from the BCCT-type of transporters, even though they may import similar types of molecules. The product of dddA, which is upstream of dddD in Rhizobium NGR234, is a predicted oxido-reductase with some relatedness to alcohol dehydrogenases, the same overall function as DddB in Marinomonas (see above). However, these ‘alcohol dehydrogenases’ are in very different families – Marinomonas DddB is a predicted Fe-containing enzyme (Pfam PF00465), but DddA of Rhizobium NGR234 is a GMC-type oxido-reductase (Pfam PF00732). Finally, in this comparison of Marinomonas and Rhizobium NGR234, the latter has a gene, which we term dddZ, whose product, like that of Marinomonas dddR is in the LysR family of transcriptional regulators. However, the DddR and DddZ proteins are only 25% identical to each other.
Both DddZ and DddA of Rhizobium NGR234 are very similar (49% and 73% identical, respectively) to the products of the two genes that are next to dddD of S. stellata (Fig. 3). S. stellata and Rhizobium NGR234 may therefore both regulate their dddD genes via the DddZ regulator, rather than by the DddR that is employed by Marinomonas and B. cepacia AMMD and may conduct an oxido-reductive step by the DddA-type, rather than the DddB-type oxido-reductase (see above). Finally, next to dddD of S. stellata is a gene whose product is 36% identical to DddT of Marinomonas, suggesting that these two species may import the DMSP by similar mechanisms, involving a BCCT-type transporter.
Overall then, at least in the four strains considered here, there is something of a ‘pick ’n mix' arrangement of the genes near their dddD genes. In some cases, the corresponding genes are predicted to have the same function and to have very similar sequences (e.g. the regulatory dddZ in S. stellata and Rhizobium NGR234). In others, the overall functions may resemble each other, but the mechanisms, and hence the corresponding genes are wholly different (e.g. the putative BCCT-type and ABC-type of DMSP transporters).
Clearly, further experimental work is required to ratify the roles of these various ddd genes and to establish if there is functional significance in the differences between the genes in different lineages. For example, do the different types of transporter have different affinities for DMSP and, if so, does this relate to the ecological status of the corresponding bacteria?
One further unexpected finding in the in silico searches of the genomes of strains that are phenotypically Ddd+ is that some of them lack dddD. For example, the marine Sulfitobacter sp. EE-36, (Gonzalez et al., 1999; http://www.roseobase.org/roseo/ee36.html) has no DddD homologue, so must use a rather different method to generate DMS from DMSP. These strains were searched for the presence of genes that would encode the N-terminal sequences of the DMSP lyases of Alcaligenes and Pseudomonas (see above), but no matches were seen. Indeed, it was noted that those ‘DMSP lyase’ sequences described by de Souza and Yoch (1996) had no matches in any of the NCBI databases (including all the available microbial genomes and environmental sequences).Therefore, a challenge is to identify these ‘alternative’ mechanisms of DMS production and to establish if strains such as SulfitobacterEE-36 accomplish their DMS emission with the DMSP lyase.
|
Bacterial DMSP lyase: myth or missing link? |
|---|
|
TopAbstractIntroductionDistribution of the DddD...'Pick 'n Mix' arrangement...Bacterial DMSP lyase: myth...DMS production and signalling...ConclusionsReferences |
|---|
As mentioned above, there have been some studies on enzyme activities that have the characteristics expected of DMSP lyase, whereby DMS is directly cleaved from the DMSP, generating acrylate. Further, in some such cases, it has been shown that the bacteria can use acrylate as the sole carbon source and that acrylate can act as a co-inducer for the production of the DMSP lyase, consistent with such a pathway (Yoch, 2002).
By contrast, it was found that Marinomonas MWYL1 did not grow on acrylate nor was the production of DMS enhanced following pregrowth of the cells in the presence of acrylate. These findings are in keeping with the ‘alternative’ pathway of DMSP catabolism that are proposed for this species (Todd et al., 2007).
Bacteria were therefore sought that can grow on both acrylate and on DMSP as the sole carbon source and which emitted DMS when grown with the latter. Such a strain was obtained, identified as a member of the Halomonas genus, from the surface of the DMSP-containing macroalga Ulva (de Souza et al., 1996) isolated from Caister, Norfolk, England.
Significantly, the ability to produce DMS from DMSP was much-enhanced when the Halomonas was pregrown in either DMSP or acrylate. Further, proteomics revealed several proteins that were induced by both these co-inducers. Perhaps significantly, one protein that was induced by pregrowth of Halomonas in DMSP was a predicted zinc-dependent alcohol dehydrogenase, specified by a gene that was very closely linked to genes that encoded acyl CoA transferases. One of these was similar to the ‘conventional’ Class III type, exemplified by E. coli CaiB and the other was strikingly similar to the DddD proteins. Therefore, growth at the expense of both acrylate and of DMSP may be due to various forms of CoA transferases, and not DMSP lyase (AWB Johnston, unpublished observations).
It was recently found that one species, Sulfitobacter EE-36, which lacks DddD from its deduced proteome, has the long-predicted DMSP lyase (Curson et al., 2007). A single gene termed dddL was cloned and it was shown that E. coli expressing dddL could make DMS from DMSP and that the other product was acrylate, the defining feature of a DMSP lyase. Furthermore, close DddL homologues occur in other marine -proteobacteria (Dinoroseobacter, Fulvimarina, Loktanella, Oceanicola) and in some strains of Rhodobacter sphaeroides, which had not been suspected of making DMS hitherto, even though this species has been studied intensively regarding its other characteristics. The DddL polypeptide appears to be a representative of a new protein family, since there are no homologues with known function (Curson et al., 2007).
Although we succeeded in indentifying a DMSP lyase, the search for mechanisms of DMSP-dependent DMS production is not over. This is because another DMS-producing strain, Roseovarius nubinhibens has neither DddD nor DddL homologues, so must use yet other enzymatic methods.
|
DMS production and signalling between bacteria and eukaryotes |
|---|
|
TopAbstractIntroductionDistribution of the DddD...'Pick 'n Mix' arrangement...Bacterial DMSP lyase: myth...DMS production and signalling...ConclusionsReferences |
|---|
It is noteworthy that several of the bacteria with DddD protein associate with higher organisms. The Marinomonas strain MWYL1 was obtained from the rhizosphere of Spartina, a known species of DMSP-producing angiosperm and a plant on whose roots Ddd+ strains of Marinomonas have been isolated in North America (Ansede et al., 2001a).
Analogously, at least two strains of rhizosphere-dwelling Burkholderia (strains B. cepacia AMMD and B. ambifaria MC40-6) have a Ddd+ phenotype. Even more complex and intimate symbioses involving Ddd+ bacteria and higher plants were seen in the case of Rhizobium sp. NGR234, which forms N2-fixing nodules on diverse legumes, and at least one non-legume host, Parasponium (Todd et al., 2007). Although not experimentally ratified, it now seems that the β-proteobacterium B. phymatum STM815, an unrelated legume root nodule bacterium, is also strongly predicted to be a Ddd+ strain (see above). It may be no coincidence that Machaerium lunatum, the host of this unusual rhizobial strain inhabits brackish marshes. It seems conceivable that some of the hosts of B. phymatum and of Rhizobium NGR234 (and other, as yet unidentified, related bacteria) may contain and exude DMSP and that this is exploited by bacteria that associate with these plants. If this is the case, it represents an interesting case in which the finding of a bacterial gene in silico leads to the prediction of unusual biochemical features in plants, even though the particular host species have not been identified yet.
There have been reports of interactions between Ddd+ bacteria with other types of eukaryotes, as well as with plants. For example Silicibacter associates with Pfiesteria piscicida, to which is it is chemo-attracted, via the DMSP made by these dinoflagellates (Miller and Belas, 2006). In addition, there are organisms that contain very high levels of DMS, pointing to the presence of active populations of micro-organisms that make this molecule, although, at present, their identitities and the mechanisms that are used are unknown. Such ‘high-intensity’ DMS sources include the mantle and gills of the tridacnid clams, the DMSP starting material almost certainly being located in zooxanthellae, which occur in high numbers in these tissues (Hill et al., 2000). Also, many species of coral, for example, Acropora formosa, are covered with mucus, which contains extremely high levels of DMSP, produced by abundant zooxanthellae dinoflagellates, which form symbiotic relationships with many coral species (Broadbent et al., 2002).
It is not known what species, let alone enzymes and genes are responsible for the emission of large amounts of DMS from these abundant sources of DMSP, but it is intriguing that so many Ddd+ systems are likely to involve associations between pro- and eukaryotes. It remains to be seen if this represents a form of nutritional opportunism on the part of the former in which they simply catabolize an available carbon and sulphur source. Certainly, the remarkable metabolic flexibility of bacteria does not preclude this as a possibility.
It is already known that DMS is a remarkably versatile and potent info-chemical, attracting organisms that are as diverse as bacteria, crustaceans, and birds. So, it is not impossible that there are other, perhaps more subtle roles for DMSP catabolism and DMS release. One thing that puzzles us is that, on the face of it, the catabolism of DMSP in any process that involves the release of DMS may be seen as inherently wasteful, since all the sulphur and much of the carbon is lost from the catabolizing organism. Given that the demethylation pathway has the potential to recoup all the C and S, why use such an apparently profligate system? This is especially pertinent for species, for example, S. pomeroyi, which have a demethylase and can release DMS, likely by the DddD system (see above).
Perhaps the Ddd pathways are only used when there is a surfeit of DMSP, so the bacteria can afford to be ‘wasteful’. Or, perhaps, this system allows carbon assimilation at a time when the cells are fully replete with sulphur. However, some bacteria, such as Rhizobium NGR234 and B. cepacia AMMD, can liberate DMS from DMSP but cannot grow on DMSP as the sole carbon source. This might be a laboratory artefact, but perhaps nutrition is not always the driving force, and the Ddd phenotype is geared more to info-chemical signalling. One truly speculative thought, the ‘messy eater’ model, combines the nutritional and the signalling properties of DMS and is as follows.
The DMSP is available to bacteria only when it is released into the environment, as would occur when the producing organisms are being lysed or damaged by pathogens or grazers. Since it is known that at least some zooplankton that eat phytoplankton are attracted by DMS, it might make sense if the Ddd+ bacteria could catabolize some of the background levels of DMSP, releasing the DMS as an attractant for more enthusiastic grazing by other planktonic herbivores. Such animals would not eat all the phytoplanktonic material and so the bacteria might be able to live off the substantial ‘scraps’ left behind. Such materials need not be confined to DMSP but could include sugars, vitamins, and amino acids, which would become available to these microbial scavengers.
|
Conclusions |
|---|
|
TopAbstractIntroductionDistribution of the DddD...'Pick 'n Mix' arrangement...Bacterial DMSP lyase: myth...DMS production and signalling...ConclusionsReferences |
|---|
It is clear from the very few molecular studies on DMSP catabolism by bacteria that there is a great deal to do—even more than might have been anticipated, given the diversity that has already been revealed. There is now an urgent need to understand the detailed mechanisms, at a molecular genetic level, of the entire pathways of DMSP catabolism in a sample of bacteria that accomplish this process in different ways (demethylation versus DMS emission and, in the latter, the different ways in which this is done). Further, the ‘specific activity’ of DMS production is far higher near eukaryotes than in the open seas. This presents a golden opportunity to study the interactions between the DMSP consumers and the DMSP producers and to determine any signalling roles of the DMS in any interactions, whether they be symbiotic, associative or pathogenic.
|
Acknowledgements |
|---|
Funding came primarily from the Biotechnological and Biological Research Council, with further support from the Natural Environmental Research Council and the European Union Framework V project, ‘Gemini’. SL was supported by a David Bryan Scholarship of the University of East Anglia. We are grateful to Michael Steinke, Nick Watmough, Marg Wexler, and Charles Brearley for useful discussions and to Gill Malin and Ron Kiene for technical advice and the provision of materials.
|
References |
|---|
|
TopAbstractIntroductionDistribution of the DddD...'Pick 'n Mix' arrangement...Bacterial DMSP lyase: myth...DMS production and signalling...ConclusionsReferences |
|---|
Anderson DG, Cantoni GL. Enzymatic cleavage of dimethylpropiothetin by Polysiphonia lanosa. Journal of Biological Chemistry (1956) 222:171–177.
Ansede JH, Robert Friedman R, Duane C, Yoch DC. Phylogenetic analysis of culturable dimethyl sulfide-producing bacteria from a Spartina-dominated salt marsh and estuarine water. Applied Environmental Microbiology (2001a) 67:1210–1217.
Ansede JH, Pellechia PJ, Yoch DC. Metabolism of acrylate to beta-hydroxypropionate and its role in dimethylsulfoniopropionate lyase induction by a salt marsh sediment bacterium, Alcaligenes faecalis M3A. Applied Environmental Microbiology (1999) 65:5075–5081.
Ansede JH, Pellechia PJ, Yoch DC. Nuclear magnetic resonance analysis of [1-C-13]dimethylsulfoniopropionate (DMSP) and [1-C-13]acrylate metabolism by a DMSP lyase-producing marine isolate of the alpha-subclass of Proteobacteria. Applied Environmental Microbiology (2001b) 67:3134–3139.
Bentley R, Chasteen TG. Environmental VOSCs: formation and degradation of dimethyl sulfide, methanethiol and related materials. Chemosphere (2004) 55:291–317.[Medline]
Biebl H, Allgaier M, Tindall BJ, Koblizek M, Lunsdorf H, Pukall R, Wagner-Dobler I. Dinoroseobacter shibae gen. nov. sp. nov. a new aerobic phototrophic bacterium isolated from dinoflagellates. International Journal of Systematics and Evolutionary Microbiology (2005) 55:1089–1096.[CrossRef]
Broadbemt AD, Jones GB, Jones RJ. DMSP in corals and benthic algae from the Great Barrier Reef. Estuarine. Coastal Shelf Science (2002) 55:547–555.[CrossRef]
Charlson R, Lovelock J, Andreae M, Warren S. Oceanic phytoplankton, atmospheric sulphur, cloud albedo and climate. Nature (1987) 326:655–661.[CrossRef]
Coenye T, Mahenthiralingam E, Henry D, LiPuma JJ, Laevens S, Gillis M, Speert DP, Vandamme P. Burkholderia ambifaria sp. nov. a novel member of the Burkholderia cepacia complex including biocontrol and cystic fibrosis-related isolates. International Journal of Systematics and Evolutionary Microbiology (2001) 51:1481–1490.
Coenye T, Vandamme P. Diversity and significance of Burkholderia species occupying diverse ecological niches. Environmental Microbiology (2003) 5:719–729.[CrossRef][Medline]
Cortay JC, Nègre D, Galinier A, Duclos B, Perrière G, Cozzone AJ. Regulation of the acetate operon in Escherichia coli: purification and functional characterization of the IclR repressor. EMBO Journal (1991) 10:675–679.[Web of Science][Medline]
Curson ARJ, Rogers R, Todd JD, Brearley CA, Johnston AWB. Moleular genetic analysis of a dimethylsulphoniopropionate lyase that liberates the climate-changing gas dimethyl sulphide in several marine -proteobacteria and Rhodobacter sphaeroides. Environmental Microbiology (2007) (in press).
de Souza MP, Chen YP, Yoch DC. Dimethylsulfoniopropionate lyase from the marine macroalga Ulva curvata: purification and characterization of the enzyme. Planta (1996) 199:433–438.[CrossRef][Web of Science]
de Souza MP, Yoch DC. Purification and characterization of dimethylsulfoniopropionate lyase from an Alcaligenes-like dimethyl sulfide-producing marine isolate. Applied Environmental Microbiology (1995a) 61:21–26.[Abstract]
de Souza MP, Yoch DC. Comparative physiology of dimethyl sulfide production by dimethylsulfoniopropionate lyase in Pseudomonas doudoroffii and Alcaligenes sp. strain M3A. Applied Environmental Microbiology (1995b) 61:3986–3991.[Abstract]
de Souza MP, Yoch DC. N-terminal amino acid sequences and comparison of DMSP lyases from Pseudomonas doudoroffii and Alcaligenes strain M3A. In: Environmental and biological chemistry on dimethylsulfoniopropionate and related sulfonium compounds—Kiene RP, Visscher PT, Keller MD, Kirst GO, eds. (1996) New York, NY: Plenum Press. 293–304.
Gonzalez JM, Covert JS, Whitman WB, et al. Silicibacter pomeroyi sp nov and Roseovarius nubinhibens sp nov. dimethylsulfoniopropionate-demethylating bacteria from marine environments. International Journal of Systematic Evolutionary Microbiology (2003) 53:1261–1269.[CrossRef]
Gonzalez JM, Kiene RP, Moran MA. Transformation of sulfur compounds by an abundant lineage of marine bacteria in the alpha-subclass of the class Proteobacteria. Applied Environmental Microbiology (1999) 65:3810–3819.
Gonzalez JM, Mayer F, Moran MA, Hodson RE, Whitman WB. Sagittula stellata gen. nov. sp. nov. a lignin-transforming bacterium from a coastal environment. International Journal of Systematic Bacteriology (1997) 47:773–780.
Gonzalez JM, Simo R, Massana R, Covert JS, Casamayor EO, Pedros-Alio C, Moran MA. Bacterial community structure associated with a dimethylsulfoniopropionate-producing North Atlantic algal bloom. Applied Environmental Microbiology (2000) 66:4237–4246.
Hill RW, Dacey JW, Edward A. Dimethylsulfoniopropionate in giant clams (Tridacnidae). Biological Bulletin (2000) 199:108–115.[Abstract]
Howard EC, Henriksen JR, Buchan A, et al. Bacterial taxa that limit sulfur flux from the ocean. Science (2006) 314:649–652.
Jonkers HM, Jansen M, Van der Maarel MJ, Van Gemerden H. Aerobic turnover of dimethyl sulfide by the anoxygenic phototrophic bacterium Thiocapsa roseopersicina. Archives of Microbiology (1999) 172:150–156.[CrossRef][Web of Science][Medline]
Kiene RP. Dimethyl sulfide production from dimethylsulfoniopropionate in coastal seawater samples and bacterial cultures. Applied Environmental Microbiology (1990) 56:3292–3297.
Kiene RP, Linn LJ, Gonzalez J, Moran MA, Bruton JA. Dimethylsulfoniopropionate and methanethiol are important precursors of methionine and protein-sulfur in marine bacterioplankton. Applied Environmental Microbiology (1999) 65:4549–4558.
Kiene RP, Taylor BF. Demethylation of dimethylsulfoniopropionate and production of thiols in anoxic sediments. Applied Environmental Microbiology (1988) 54:2208–2212.
Lovelock JE. Gaia. A new look at life on Earth (1979) Oxford University Press.
Lovelock JE. Travels with an electron capture detector. Resurgence (1999) 187.
Lovelock JE, Maggs RJ, Rasmussen RA. Dimethyl sulphide and the natural sulphur cycle. Nature (1972) 237:452–453.[CrossRef]
Malmstrom RR, Kiene RP, Cottrell MT, Kirchman DL. Contribution of SAR11 bacteria to dissolved dimethylsulfoniopropionate and amino acid uptake in the North Atlantic ocean. Applied Environmental Microbiology (2004) 70:4129–4135.
Miller TR, Belas R. Motility is involved in Silicibacter sp. TM1040 interaction with dinoflagellates. Environmental Microbiology (2006) 8:1648–1659.[CrossRef][Medline]
Moran MA, Buchan A, Gonzalez JM, et al. Genome sequence of Silicibacter pomeroyi reveals adaptations to the marine environment. Nature (2004) 432:910–913.[CrossRef][Medline]
Moran MA, González JM, Kiene RP. Linking a bacterial taxon to sulfur cycling in the sea: studies of the marine Roseobacter group. Geomicrobiology Journal (2003) 20:375–388.[CrossRef][Web of Science]
Mou X, Moran MA, Stepanauskas R, Gonzalez JM, Hodson RE. Flow-cytometric cell sorting and subsequent molecular analyses for culture-independent identification of bacterioplankton involved in dimethylsulfoniopropionate transformations. Applied Environmental Microbiology (2005) 71:1405–1416.
Moulin L, Munive A, Dreyfus B, Boivin-Masson C. Nodulation of legumes by members of the β-subclass of proteobacteria. Nature (2001) 411:948–950.[CrossRef][Medline]
Nevitt GA, Bonadonna F. Sensitivity to dimethyl sulphide suggests a mechanism for olfactory navigation by seabirds. Biology Letters (2005) 1:303–305.[CrossRef][Web of Science][Medline]
Otte ML, Wilson G, Morris JT, Moran BM. Dimethylsulphoniopropionate (DMSP) and related compounds in higher plants. Journal of Experimental Botany (2004) 55:1919–1925.
Pinhassi J, Simo R, Gonzalez JM, Vila M, Alonso-Saez L, Kiene RP, Moran MA, Pedros-Alio C. Dimethylsulfoniopropionate turnover is linked to the composition and dynamics of the bacterioplankton assemblage during a microcosm phytoplankton bloom. Applied Environmental Microbiology (2005) 71:7650–7660.
Rangarajan ES, Li Y, Iannuzzi P, Cygler M, Matte A. Crystal structure of Escherichia coli crotonobetainyl-CoA: carnitine CoA-transferase (CaiB) and its complexes with CoA and carnitinyl-CoA. Biochemistry (2005) 44:5728–5738.[CrossRef][Web of Science][Medline]
Steinke M, Stefels J, Stamhuis E. Dimethyl sulfide triggers search behavior in copepods. Limnology and Oceanography (2006) 51:1925–1930.[Web of Science]
Steinke M, Wolfe GV, Kirst GO. Partial characterization of dimethyl-sulfoniopropionate (DMSP) lyase isozymes in 6 strains of Emiliania huxleyi. Marine Ecology: Progress Series (1998) 175:215–225.[CrossRef][Web of Science]
Sunda W, Kieber D, Kiene RP, Huntsman S. An antioxidant function for DMSP and DMS in marine algae. Nature (2002) 418:317–320.[CrossRef][Medline]
Todd JD, Rogers R, Li YG, Wexler M, Bond PL, Sun L, Curson ARJ, Malin G, Steinke M, Johnston AWB. Structural and regulatory genes required to make the gas dimethyl sulfide in bacteria. Science (2007) 315:666–669.
Todd JD, Sawers G, Johnston AWB. Proteomic analysis reveals the wide-ranging effects of the novel, iron-responsive regulator RirA in Rhizobium leguminosarum bv. viciae. Molecular Genetics Genomics (2005) 273:197–206.[CrossRef]
Tringe SG, von Mering C, Kobayashi A, et al. Comparative metagenomics of microbial communities. Science (2005) 308:554–557.
Trinick MJ. Symbiosis between Rhizobium and non-legume Trema aspera. Nature (1973) 244:459–460.[CrossRef]
van der Maarel MJEC, van Bergeijk S, van Werkhoven AF, Laverman AM, Meijer WG, Stam WT, Hansen TA. Cleavage of dimethylsulfoniopropionate and reduction of acrylate by Desulfovibrio acrylicus sp. nov. Archives of Microbiology (1996) 166:109–115.[CrossRef][Web of Science]
Venter JC, Remington K, Heidelberg JF, et al. Environmental genome shotgun sequencing of the Sargasso Sea. Science (2004) 304:66–74.
Vila M, Simo R, Kiene RP, Pinhassi J, Gonzalez JM, Moran MA, Pedros-Alio C. Use of microautoradiography combined with fluorescence in situ hybridization to determine dimethylsulfoniopropionate incorporation by marine bacterioplankton taxa. Applied Environmental Microbiology (2004) 70:4648–4657.
Visscher PT, Taylor BF. Demthylation of dimethylsulfoniopropionate to 3-mercaptopropionate by an aerobic marine bacterium. Applied Environmental Microbiology (1994) 60:4617–4619.
Yoch DC. Dimethylsulfoniopropionate: its sources, role in the marine food web, and biological degradation to dimethylsulphide. Applied Environmental Microbiology (2002) 68:5804–5815.
Yoch DC, Ansede JH, Rabinowitz KS. Evidence for intracellular and extracellular dimethylsulfoniopropionate (DMSP) lyases and DMSP uptake sites in two species of marine bacteria. Applied Environmental Microbiology (1997) 63:3182–3188.[Abstract]
Zubkov MV, Fuchs BM, Archer SD, Kiene RP, Amann R, Burkill PH. Linking the composition of bacterioplankton to rapid turnover of dissolved dimethylsulphoniopropionate in an algal bloom in the North Sea. Environmental Microbiology (2001) 3:304–311.[CrossRef][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J.-B. Raina, D. Tapiolas, B. L. Willis, and D. G. Bourne Coral-Associated Bacteria and Their Role in the Biogeochemical Cycling of Sulfur Appl. Envir. Microbiol., June 1, 2009; 75(11): 3492 - 3501. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||