One If by Land and Two If by Sea

by S. Mar­vin Fried­man

Is the cur­rent pre­oc­cu­pa­tion of the sci­en­tific com­mu­nity with se­quenc­ing every genome ex­tant merely an ex­er­cise in pro­vid­ing data­bases for Gen­Bank? Cer­tainly not, for this in­for­ma­tion also re­veals very im­por­tant char­ac­ter­is­tics of the or­gan­ism un­der analy­sis. For ex­am­ple, we are at the early stages of a med­ical rev­o­lu­tion that will see af­ford­able in­di­vid­ual hu­man genome se­quences made avail­able to physi­cians as part of rou­tine an­nual check-ups. As the role of var­i­ous genes in­volved in health is­sues be­comes even more fully elu­ci­dated, such a re­port will al­low the physi­cian to com­pile a pro­file of the patient's sus­cep­ti­bil­i­ties to a broad range of dis­eases. In the case of mi­cro­bi­ol­ogy, many fresh in­sights have been de­rived from genome se­quences, in­clud­ing one that I have pre­vi­ously blogged in STC, namely genome re­duc­tion as one pos­si­ble con­se­quence of a bacterium's com­mit­ment to an ob­lig­ate par­a­sitic lifestyle. An­other of my blogs showed that the genome of the 14^th cen­tury plague bacil­lus isn't so dif­fer­ent from that of the much less vir­u­lent mod­ern strains, in­di­cat­ing that other fac­tors dur­ing the me­dieval era were prob­a­bly in play. In the pa­per I am about to dis­cuss, we will see that bac­te­r­ial genome se­quences tell us when and how some bac­te­ria tran­si­tioned from aquatic to ter­res­trial ecosys­tems.

Col­o­niza­tion of wheat roots with Azospir­il­lum lipoferum la­beled with the green flu­o­res­cent pro­tein (gfp) re­porter gene. Source

Ac­cord­ing to fos­sil records, life orig­i­nated in ma­rine en­vi­ron­ments 3.5–3.8 bil­lion years ago (Gyr) and then tran­si­tioned to ter­res­trial habi­tats about 2.6 Gyr. The rel­a­tive ab­sence of fos­sil records for bac­te­ria makes it dif­fi­cult to as­sign these time points for them, al­though se­quence data in­di­cates that a large clade of prokary­otic phyla, called terrabac­te­ria, might have evolved on land as early as 3 Gyr. In­ter­est­ingly, it ap­pears that some lin­eages sub­se­quently re­turned to their an­ces­tral home. Thus, cyanobac­te­ria are mem­bers of the terrabac­te­ria, al­though the genus Prochloro­coc­cus is now the dom­i­nant pri­mary pro­ducer in the oceans.

Azospir­il­lum brasilense. Source

Terrabac­te­ria have good sto­ries to tell. For ex­am­ple, Azospir­il­lum col­o­nizes the roots of ce­re­als (rice, wheat, and maize) and other grasses where it en­hances plant growth by car­ry­ing out ni­tro­gen fix­a­tion and se­cret­ing phy­to­hor­mones, such as in­dole-3-acetic acid. The genus Azospir­il­lum be­longs to the α‑proteobacteria, one of the ma­jor groups of hy­drobac­te­ria (a clade of prokary­otes that first ap­peared in ma­rine habi­tats). Within its fam­ily, the Rho­dospir­il­laceae, al­most all of its known rel­a­tives are found in aquatic en­vi­ron­ments, sug­gest­ing that Azospir­il­lum may have tran­si­tioned to life on land later, af­ter the Pre­cam­brian di­ver­gence of hy­drobac­te­ria and terrabac­te­ria. To gain in­sights into how these bac­te­ria tran­si­tioned from ma­rine to ter­res­trial ecosys­tems, the au­thors se­quenced the genomes of Azospir­il­lum brasilense and Azospir­il­lum lipoferum. A third genome from an uniden­ti­fied Azospir­il­lum species was de­scribed while their work was in progress.

In all three species, the Azospir­il­lum genome con­sists of seven repli­cons. The largest repli­con can be con­sid­ered a chro­mo­some, the small­est a plas­mid.  Ge­netic el­e­ments in­ter­me­di­ate  be­tween chro­mo­somes and plas­mids are re­ferred to as chro­mids. Dif­fer­ent com­bi­na­tions of these three repli­con types oc­curred in the three species of Azospir­il­lum. A. lipoferum con­tains five chro­mids, the largest num­ber among all prokary­otes se­quenced to date, sug­gest­ing great ge­nomic plas­tic­ity. In fact, there are more ge­nomic re­arrange­ment in Azospir­il­lum than in Rhi­zo­bium, the multi-repli­con α‑proteobacterium that pre­vi­ously had the dis­tinc­tion of hav­ing the most plas­tic genome known among the prokary­otes.

Func­tional cat­e­gories for A. lipoferum 4B genes en­riched in an­ces­tral (top) and hor­i­zon­tally trans­ferred (bot­tom) genes. Only genes that were pre­dicted with high con­fi­dence are shown. Source

Genome nu­cleotide se­quences were an­a­lyzed us­ing a BLAST al­go­rithm mod­i­fied by the au­thors to dis­tin­guish an­ces­tral pro­tein cod­ing genes from those that had been ac­quired by hor­i­zon­tal trans­fer (HGT). Gene as­sign­ments to each of these two cat­e­gories could be as­signed a high, medium, or low level of con­fi­dence. The as­ton­ish­ing re­sults in­di­cated that about half of the Azospir­il­lum genes seem to be hor­i­zon­tally trans­ferred. Con­trol analy­ses of other Rho­dospir­il­laceae genomes re­vealed much lower lev­els of HGT in aquatic species, whereas an­ces­tral gene fre­quen­cies were about the same in all or­gan­isms tested. The an­ces­tral genes were mostly in­volved in house­keep­ing roles, such as trans­la­tion, post-trans­la­tional mod­i­fi­ca­tions, cell di­vi­sion, and me­tab­o­lism of both nu­cleotides and coen­zymes. The hor­i­zon­tally trans­ferred genes con­trolled po­ten­tially dis­pens­able func­tions, such as de­fense mech­a­nisms, cell wall bio­gen­e­sis, and the trans­port and me­tab­o­lism of var­i­ous fac­tors in­clud­ing amino acids, car­bo­hy­drates, in­or­ganic ions, and sec­ondary metabo­lites. These newly ac­quired ca­pa­bil­i­ties would have proved use­ful in adapt­ing to the rhi­zos­phere, an ecosys­tem that is rich in amino acids, car­bo­hy­drates, in­or­ganic ions, and sec­ondary metabo­lites se­creted by plant roots. One might also ex­pect to see the other side of the coin, namely that an­ces­tral genes that were ben­e­fi­cial in aquatic habi­tats would be lost when the azospir­illa col­o­nized the ter­res­trial rhi­zos­phere. In­deed, genes in­volved in pho­to­syn­the­sis are a ba­sic fea­ture of the aquatic Rho­dospir­il­laceae, but are ab­sent from Azospir­il­lum.

Pro­por­tion of an­ces­tral (red) and hor­i­zon­tally trans­ferred (blue) genes in­volved in adap­ta­tion of Azospir­il­lum to the rhi­zos­phere and its in­ter­ac­tion with host plants. Color in­ten­sity in­di­cates high (dark), medium (medium) and low (light) con­fi­dence lev­els for pre­dic­tion. Source

Let me now take a look at some spe­cific HGT genes in Azospir­il­lum that most likely en­abled it to spe­cial­ize for life in close as­so­ci­a­tion with plants. Azospir­illa har­bor a large num­ber of chemo­taxis genes, ac­quired in a step­wise fash­ion. As a mat­ter of fact, the un­known Azospir­il­lum species con­tains a whop­ping 126 chemo­taxis genes, more than any other prokary­ote se­quenced to date. It is of in­ter­est here that a re­cent study  showed that the rhi­zos­phere abounds with chemo­taxis genes, in­di­cat­ing their im­por­tance in root-mi­crobe in­ter­ac­tions. The Azospir­il­lum genomes also en­coded many cel­lu­lases, hemi­cel­lu­lases, and glycosylhydrolases—all be­ing en­zymes cru­cial for the de­com­po­si­tion of plant cell walls and sub­se­quent mi­cro­bial pen­e­tra­tion. The A. brasilense genome en­coded a set of genes for TAD (tight ad­he­sion) pili and their im­por­tance was em­pha­sized by the find­ing that a mu­tant de­fi­cient in TAD pili was se­verely im­paired in plant at­tach­ment and biofilm for­ma­tion. The ex­tra­or­di­nar­ily high level of HGT in the azospir­illa was prob­a­bly ac­com­plished through con­ju­ga­tion and trans­duc­tion as these bac­te­ria host phages and con­tain a Gene Trans­fer Agent.

Tax­o­nomic dis­tri­b­u­tion, based on best BLAST hits, of the sources of genes Azospir­il­lum likely ac­quired by HGT. Source

The next ques­tion the au­thors ad­dressed was: Where did the HGT genes come from? The best BLAST hits In­di­cated that ter­res­trial pro­teobac­te­ria, es­pe­cially mem­bers of the Rhi­zo­biales (dis­tantly re­lated, ni­tro­gen-fix­ing α‑proteobacteria ) and of the Burk­hold­erales (β‑proteobacteria that col­o­nize soil and plants) were the most promi­nent donors. Fail­ing fos­sil data, the is­sue of when the azospir­illa di­verged from their aquatic Rho­dospir­il­laceae an­ces­tors can only be roughly ap­prox­i­mated us­ing the es­tab­lished clock by which a 1–2% di­ver­gence in the 16S rRNA gene rep­re­sents 30 mil­lion years (Myr). Us­ing this cal­cu­la­tion, they found that the azospir­illa would have tran­si­tioned to land about 400 Myr. In­ter­est­ingly, that is about the time of the ini­tial ma­jor ra­di­a­tion of vas­cu­lar land plants and the evo­lu­tion of plant roots. Grasses, the pre­dom­i­nant plant host for Azospir­il­lum, ap­peared much later, about 65–80 Myr.

This pa­per once again demon­strates that ge­nomics is a pow­er­ful tool for open­ing doors to mi­cro­bial evo­lu­tion that can­not be ac­cessed by other ap­proaches. So, the next time you en­counter that lengthy string of let­ters for a newly se­quenced bac­te­r­ial genome, re­mem­ber that some very im­por­tant and unique in­fer­ences can be drawn from that seem­ingly ab­stract data.

Ref­er­ence

Wis­niewski-Dyé F, Borziak K, Khalsa-Moy­ers G, Alexan­dre G, Sukharnikov LO, Wuichet K, Hurst GB, Mc­Don­ald WH, Robert­son JS, Barbe V, Cal­teau A, Rouy Z, Mangenot S, Pri­gent-Com­baret C, Nor­mand P, Boyer M, Sigu­ier P, Dessaux Y, Elmerich C, Con­dem­ine G, Kr­ish­nen G, Kennedy I, Pa­ter­son AH, González V, Mavin­gui P, Zhulin IB (2011). Azospir­il­lum genomes re­veal tran­si­tion of bac­te­ria from aquatic to ter­res­trial en­vi­ron­ments. PLoS Ge­net­ics, 7 (12). PMID 22216014

Mar­vin is Pro­fes­sor Emer­i­tus in the De­part­ment of Bi­o­log­i­cal Sci­ences at Hunter Col­lege of CUNY in New York City, and an As­so­ciate Blog­ger for Small Things Con­sid­ered.