Wednesday, March 24, 2010

Nod you later

Lupin, a nodulating legume -- Creative Commons photo by Flickr user aussiegall

Nitrogen seems like a strange element to have problems acquiring. After all, it’s by far the most abundant element in the atmosphere -- about two thousand times as abundant as CO2 -- and, on average, plants only need one nitrogen atom per 30 carbon atoms. Yet, as any gardener will tell you, nitrogen is also one of the most important limiting resources for plants, and is the sole reason we have a lucrative international trade in bird crap. Despite being bathed in nitrogen for their entire lives, plants lack the metabolic machinery to break the strong triple bond between the two nitrogen atoms. They need someone else do the hard work for them. 

Enter the microbes.
Unlike the Blochmannia symbionts of carpenter ants I discussed last time, the function of this symbiosis had been figured out long before the bacterial partner was identified. Farmers have known for thousands of years that certain varieties of legumes could grow in soil depleted by other crops. In 1888, the Dutch agricultural microbiologist Martinus Beijerinck figured out that the funny little pink nodules on legume roots were actually packed full of bacteria, and that those bacteria could fix atmospheric nitrogen. It was one of the seminal discoveries in microbiology, and its extraordinary economic and ecological significance has led legume nodulation to become one of the most studied mutualisms in nature. (ed. note: the copious literature has something to do with why this post took a while!) However, the story is far from complete: despite over a century of biochemical study, the system’s fascinating evolutionary questions have come under investigation only recently. 
Collectively, the bacteria that inhabit nitrogen-fixing nodules in legumes are known as the rhizobia. Although they were originally considered to form a single lineage of bacteria, molecular and other methods have made it clear that the rhizobia are really a paraphyletic collection of many different groups within the alpha- and beta-proteobacteria (Masson-Boivin et al. 2009). These lineages themselves diverged long before the evolution of mutualistic legume nodulation and, as a rule, contain a number of non-nodulating lineages, indicating that the capacity to nodulate legumes has evolved many times independently in the proteobacteria, probably as a result of horizontal gene transfer (Amadou et al. 2008). The “true” rhizobia only nodulate legumes, which form the monophyletic family Fabaceae; this suggests that, among plants, rhizobial nodulation has only evolved once (Gualtieri and Bisseling 2000). 
The legumes are not the only plants to form nitrogen-fixing nodules with bacterial mutualists, however. A number of other plant families are known to form nitrogen-fixing root nodules with the actinomycete genus Frankia, which is wholly unrelated to the rhizobia. Although originally placed in separate clades, the “actinorhizal” plants have more recently been shown to belong to a single lineage within the Rosid group of core eudicots. Differences in the ontogeny of nodules among the actinorhizal and rhizobial plants suggest that much of the morphological similarities are a result of evolutionary convergence, and it is uncertain whether the actinorhizal plants had a nodulating ancestor (Gualtieri and Bisseling suggest it actinorhizal nodulation evolved several times independently). In contrast to the rhizobia, though, the actinorhizal symbionts do seem to form a monophyletic group. 
This sets up an interesting contrast, to which I’ll return in a bit: the rhizobia are a diverse group of bacteria that form a nitrogen-fixing mutualism with very restricted group of Rosid plants, while Frankia is a single lineage of bacteria that forms mutualisms with a much broader range of the Rosids. The overwhelming economic importance of the legumes, combined with the comparative amenability of most rhizobia to culture, means that tremendously more work has been done on the the former system (just check the number of hits on Google Scholar for rhizobium vs frankia), but as we learn more about the actinorhizal mutualism, the comparisons should prove extremely interesting.
For now, though, let’s return to the rhizobia. Unlike the intracellular symbionts of carpenter ants and aphids, rhizobia don’t spread through the seeds of their hosts, and must be acquired anew by each generation. As one might expect, the soils are teeming with rhizobial bacteria ready to team up with legume seedlings. The flipside to this is that the soils are also teeming with related bacteria that don’t possess the genes necessary to make good mutualists, or may even be pathogenic. To help pick the wheat from the chaff (or, I suppose, the soy from the pea), legumes and rhizobia have evolved a complex chemical signaling system to recognize potential partners. Rhizobia secrete these “nodulation (nod) factors” when they are exposed to flavonoids produced by the legume roots. The nod factors in turn induce the roots to develop structures to facilitate infection, often in the form of a specialized root hair. Individual rhizobia species produce specific cocktails of nod factors, which contribute to varying ranges of host specificity; although these factors may be transferred within rhizobial lineages, there do seem to be some barriers to natural transfer of specific nod factors between widely divergent groups (Wernegreen and Riley 1999). 
Once the rhizobia successfully form an intracellular infection, they generally differentiate from their typical rod-shaped free-living morphology to become enlarged, nitrogen-fixing bacteroids. These cells may or may not be terminally differentiated, meaning that they do not continue to divide. Although there are exceptions, bacteroids in determinate nodules (root nodules that stop growing at a certain size, and in which all the cells are approximately the same age) are not terminally differentiated, while those in indeterminate nodules (nodules that retain a dividing meristem at one end) are terminally differentiated, and new bacteroids at the apical end of the nodule differentiate from a small population of rod-morph bacteria that remain inside the infection thread Denison 2000).  
This leads to an interesting conundrum: if you’re a bacterium that must get out of the nodule and survive away from your host each generation, your fitness isn’t linked to that of the plant; why not cheat, and hoard host-supplied nutrients without performing the extremely expensive nitrogen fixing in return?
Despite lots of work done on rhizobia, and lots of discussion of the theory of cooperation in biology and evolution since John Maynard-Smith in the 60‘s, not many people have looked into this until relatively recently. This is partly because the the functional focus in the field came partly at the expense of natural history; as Denison notes in his 2000 review on this topic, “Sutton’s complaint [in 1983] that ‘modern studies on... [the] microbiology of nodule senescence... [are] now overdue’ remains largely true today.”
Recent research suggests that, as predicted by theory, a lot of biological mechanisms are in place to ‘stack the deck’ towards maintenance of cooperation. The first pass of recognition factors in early infection may produce one boundary to cheating, by forcing cheating to evolve among the population of bacteria already containing the proper signaling genes; and the smaller the population size, the less likely the necessary cheater mutation. Even then, you would expect cheating to evolve occasionally; and indeed, a number of studies have showed the presence of natural strains of rhizobial bacteria that can nodulate some plants with decreased or absent nitrogen fixing ability. 
So it is perhaps unsurprising that Denison and his colleagues have demonstrated that host plants seem to have an ability to sanction bacteria that stop cooperating. While the specifics are too in depth for this already monstrous post, the neat trick seems to be that the host ‘punishes’ underperforming nodules not by withholding photosynthate ‘food,’ which might inspire the cheaters to start consuming the plant’s carbon, but by limiting the supply of oxygen, limiting the cheater’s ability to reproduce (Kiers et al. 2003; Oono et al. 2009). 
The presence of host sanctions may also explain quandary of differentiation, or why some bacteria seem to give up the ability to reproduce altogether. As has been demonstrated in social insects such as ants and bees, the important term is not necessarily individual fitness, but inclusive fitness  -- the number of offspring related to you in the next generation. Hamilton’s rule states that evolution will favor a mutation when rb > c  -- in other words, when the relatedness times the benefit of cooperation is greater than the cost. If bacteria in a nodule are mostly clones of each other, r ≈ 1; any benefit derived (or cost of host sanctions avoided) is thus more likely to exceed the cost of the alternative, leading to greater reproduction of your genes. Thus terminally differentiated bacteria in indeterminate nodules, by focusing their energies on nitrogen fixation, can increase the fitness of their undifferentiated clone-mates enough to compensate for their own sterility. By contrast, non-terminally differentiated bacteroids in determinate nodules do seem to reproduce, and these have been shown to sequester energy-dense molecules that may serve to provision them after the nodule senesces (cheating in the sense that they could be focusing that energy on nitrogen fixation).
All of which brings me to a final question. Are there fundamental differences in the evolutionary dynamics of rhizobial and actinorhizal symbioses? Rhizobia have the large, complex genomes that you would expect for soil bacteria, which must face a multitude of challenges (Amadou et al. 2008); they seem well-provisioned for independent life, which is potentially dangerous in a partner -- you’d prefer they be completely dependent! Comparative analysis of Frankia genomes has shown that actinorhizal bacteria also have the large genomes typical of soil-dwelling actinomycetes, but that Frankia species with narrower host ranges have smaller genome sizes. Rhizobial bacteria are extraordinarily diverse, but but form partnerships with only one lineage of plants; actinorhizal bacteria are much less diverse, but may have evolved nodulation multiple times across the Rosid I clade of plants. Detailed recognition systems, like the Nod factors present in almost all rhizobia, haven’t yet been described in Frankia. Are they, or something like them, hidden in the genome? Or has the actinorhizal symbiosis evolved in such a way that this particular kind of signaling is unnecessary? 
Of the nodulating symbioses, I tend to be more attracted to the actinorhizal systems -- but then, I seem to be drawn irresistibly to less-studied and less-tractable study organisms. Both are pretty fascinating, and provide ready test-beds for the latest advances in genomic sequencing and analysis. I’m sure the coming years will provide an explosion of new work on both, and provide an extraordinary comparative view into the evolution and maintenance of cooperation.
I just feel bad for the person who updates this blog post ten years from now...
Annotated Bibliography
Willems. The taxonomy of rhizobia: an overview. Plant Soil (2006) vol. 287 (1-2) pp. 3-14

Review focused on the taxonomy of traditionally established rhizobial strains, naming a total of 53 species across 7 genera. Doesn’t deal with the evolutionary history of these lineages, except insofar as it relates to the 16S tree published as figure 1.  
Gualtieri and Bisseling. The evolution of nodulation. Plant Mol Biol (2000) vol. 42 (1) pp. 181-194

Nice, broad review focusing on the morphological and developmental indicators of nodulation evolution in actinorhizal and rhizobial plants. Also includes a brief overview of the then-current molecular relationships among symbionts, as well as biochemical foundations of specificity (nod factors). 
Wernegreen and Riley. Comparison of the evolutionary dynamics of symbiotic and housekeeping loci: A case for the genetic coherence of rhizobial lineages. Molecular Biology and Evolution (1999) vol. 16 (1) pp. 98-113

sequenced a number of Rhizobium, Sinorhizobium, and Mezorhizobium strains at NodB, NodC, Glutamine Synthetase II, and ITS. Found general congruence of phylogenies within genera, but evidence of transfer among congeneric bacteria -- implication is that host range restricts transfer. (Could also be sensitive to things like sequence identity in homologous recombination, host range restriction in plasmids, etc.) Suggested broader transfer events may be artifacts of agricultural pressures. 
Masson-Boivin et al. Establishing nitrogen-fixing symbiosis with legumes: how many rhizobium recipes?. Trends in Microbiology (2009) vol. 17 (10) pp. 458-466

Fantastic review (I love Trends) focusing on the evolution and comparative genomics of the rhizobial bacteria, updating the picture to include the dramatic advances in molecular biology (sequencing, genomics) and microbial taxonomy undertaken since Gualtieri and Bisseling (2000). Provides a good perspective on the functional elements (nod factors, nif genes, infection pathways) necessary for establishment of rhizobial symbiosis in legumes, and a convincing general hypothesis for why it appears to be so widespread in the proteobacteria. 
Amadou et al. Genome sequence of the beta-rhizobium Cupriavidus taiwanensis and comparative genomics of rhizobia. Genome Research (2008) vol. 18 (9) pp. 1472-1483

First genome published for a non-alpha rhizobium. Two chromosomes surprisingly syntenous with free-living saprotrophic congener; lots of stereotypical symbiosis genes on the plasmid, though. Authors compared existing rhizobia genomes for genes that are universally present and specific (not in free-living genomes) and found none; performed clustering analysis on 214 genes in S. meliloti that have been described experimentally as important to symbiosis, and found that a bunch were pretty much universally distributed among sequenced prokaryotes (housekeeping genes, except in obligate vertical endosymbionts), a bunch were pretty specific to rhizobia or even S. meliloti (nod factors, etc), and a handful that had somewhat more patchy distribution. Also looked for genes that appeared enriched in rhizobia vs. nonrhizobia; found that most were undescribed, but many appeared to be involved in nutrient (Phosphorous and nitrogen) metabolism, as might be expected. Also noted that laboratory screens for mutants may very well miss a bunch of genes that are really important in nature, which may help explain why lots of those genes are unannotated. 
Normand et al. Genome characteristics of facultatively symbiotic Frankia sp strains reflect host range and host plant biogeography. Genome Research (2007) vol. 17 (1) pp. 7-15

Fascinating genome paper examining the comparative genomics of three Frankia strains, each with different host ranges and distributions. The strain with the narrowest host range also had the smallest genome size, with an apparent bias towards loss of genes implicated in metabolite uptake and processing. Also some DNA repair genes. Really cool.
Gibson et al. Molecular Determinants of a Symbiotic Chronic Infection. Annu. Rev. Genet. (2008) vol. 42 pp. 413-441

Nice review, focusing especially on molecular underpinnings of symbiosis in rhizobia (nod factors, bacterial and host biochemical products), but with some attention paid to the role of these products in the evolution of the interaction. Considers also the cellular processes of division and differentiation involved in nodulation, both for the plant and bacteria.
Szczyglowski and Amyot. Symbiosis, inventiveness by recruitment?. Plant Physiol (2003) vol. 131 (3) pp. 935-940

Short review focused on the biochemical evidence that rhizobial nodulation evolved by taking advantage of existing pathways utilized in arbuscular mycorrhizal symbioses. 
Sprent. 60Ma of legume nodulation. What's new? What's changing?. J Exp Bot (2008) vol. 59 (5) pp. 1081-1084

Short review explicitly considering the likely evolutionary history leading to nodulation of the legumes. Useful starting point for links to references that sort out the functional elements of relevance to questions of evolution, focusing on plant and microbial lineages in turn. 
Sprent. Evolving ideas of legume evolution and diversity: a taxonomic perspective on the occurrence of nodulation. New Phytologist (2007) vol. 174 (1) pp. 11-25

Neat, very readable review examining the evolutionary history of the nodulating legumes, and drawing attention to the most relevant empirical studies of the system. Literature indicates that nodulating symbiosis evolved shortly after evolution of the legumes, ~60mya. Bulk of paper is devoted to specific examination of each of the three main nodulating legume lineages; the Caesalpinioideae, the Mimosoideae, and the Papilionoidea.

Intersting factoids include observation that, in general, interactions are more specific in temperate than tropical ecosystems. Temperate colonization was more recent, and Sprent attributes greater specificity in that region to greater dependence on N-fixation and possibly lower existing diversity. 
Kiers et al. Host sanctions and the legume-rhizobium mutualism. Nature (2003) vol. 425 (6953) pp. 78-81

Experiment testing whether hosts maintain cooperation in symbionts through sanctions. Kiers et al grew soybeans with a normally cooperative Bradyrhizobium strain in a variety of different media, ‘enforcing’ cheating in half of the conditions by limiting the plants to a nitrogen-free atmosphere; they were able to detect increased nodule size, bacterial density, and numbers of rhizobia in the media for treatments with nitrogen. Decreases in oxygenated leghaemoglobin in the Ar:O
2 treatments suggested that these differences in bacterial fitness were the result of host sanctioning through decreased supply of oxygen. 
Denison. Legume sanctions and the evolution of symbiotic cooperation by rhizobia. Am Nat (2000) vol. 156 (6) pp. 567-576

Really well thought out review examining the evolutionary problems of cooperation in the rhizobia system, and considering the various ways in which cooperation might be stabilized. Calls attention to the glaring omissions in the literature (e.g. what is the complete life history of cooperative bacteria) that have direct relevance to the question. Does an admirable job of linking standard considerations in the cooperation theory (e.g. importance and mechanism of kin selection, germ-soma differentiation, etc) to the natural history of the rhizobia system. Makes a compelling case (demonstrated in later papers by the group) for the importance of host sanctions. 
Oono et al. Controlling the reproductive fate of rhizobia: how universal are legume sanctions?. New Phytologist (2009) vol. 183 (4) pp. 967-979

Another interesting review updating the concepts explored in Denison 2000, but focusing primarily on 1) the implications of different bacterial life-history strategies (do I differentiate into nonreproductive bacteroids, and why?) and 2) how these might interact with host sanctioning. Propose the hypothesis that rhizobial dimorphism evolved as a way to promote factors that enhance the availability of host-supplied incentives (O
2 and photosynthate) while maintaining a within-nodule competitive advantage; e.g., non-reproductive differentiated bacteroids work to enhance the fitness of their reproductive clone-mates by adopting a morphology that inherently increases the efficiency of N2 fixation, and since the relatedness within the nodule is very high, the increased benefit is adequate to maintain cooperation through kin selection. Conversely, in bacteroids that retain the capacity to reproduce, investment in things like PHBs makes sense. 


Anonymous said...

Good dispatch and this post helped me alot in my college assignement. Gratefulness you for your information.

Anonymous said...

Sorry for my bad english. Thank you so much for your good post. Your post helped me in my college assignment, If you can provide me more details please email me.

Anonymous said...

Sorry for my bad english. Thank you so much for your good post. Your post helped me in my college assignment, If you can provide me more details please email me.

Anonymous said...

[url=]online payday loans[/url]
This is the best way to get all your health products online like green coffee, african mango, phen375 and others. Visit now

[url=]Zeta Clear Reviews[/url]