The Genes, The Whole Genes, & Noth­ing But The Genes

by Merry

We have come to ex­pect the un­ex­pected of cil­i­ates, and Oxytricha tri­fal­lax, with its ge­nomic ca­pers, does not dis­appoint. Like many of its more fa­mous cil­i­ate rel­a­tives (e.g., para­me­cium, tetrahy­mena, sten­tor), Oxytricha is a com­plex uni­cel­lu­lar or­gan­ism with many spe­cial­ized cel­lular struc­tures. Of course, they have the req­ui­site cilia for lo­co­mo­tion, feed­ing, and sens­ing their en­vi­ron­ment. But they also have a "mouth," food vac­uoles where di­gestion takes place, kid­ney-like con­trac­tile vac­uoles for os­moreg­u­la­tion, and a pseudo-anus – the cy­to­proct. Not sur­pris­ingly, cil­i­ates tend to be large cells, some mea­su­red in mil­lime­ters. Com­bined with a fast rate of growth, this may re­quire more gene tran­scrip­tion than a diploid genome can sup­port. To cope with this de­mand, cil­i­ates em­ploy a con­vo­luted strat­egy, one unique so far to this group.

SEM of Oxytricha in the process of re­pro­ducing. Source

Cil­i­ates have two types of nu­clei in each cell, mi­cronu­clei (Mics) and macronu­clei (Macs), each of which per­forms dif­fer­ent func­tions. Mics are small, but con­tain the en­tire genome and pass it on to the next gen­er­a­tion; Macs are huge and con­tain only a sub­set of the genome, du­pli­cated many times. That sub­set is suf­fi­cient to sup­port veg­e­ta­tive growth. The num­ber of these nu­clei varies among the cil­i­ates; Oxytricha has two iden­ti­cal Mics plus two to four Macs. Dur­ing con­ju­ga­tion, one of the Mics un­der­goes meio­sis and fer­til­iza­tion (see Fig­ure to the right). Af­ter­wards, the old Macs are dis­carded and a Mic gives rise to the new Macs. To cre­ate a new Mac, the Mic per­forms an in­tri­cate – one could say mind-bog­gling – spec­ta­cle of DNA dif­fer­en­ti­a­tion.

The 'Mic-Mac' cy­cle in cil­i­ates. Source

Oxytricha's germline genome housed in the Mic is a hefty 1 Gb. Of that to­tal, only 5% is used to en­code its 30,000 genes; the other 95% is "junk" DNA in­clud­ing transpo­sons and other non-cod­ing se­quences. The Mac, in con­trast, con­tains only the cod­ing 5%. The other 95% of the genome is elim­i­nated when a Mic dif­fer­en­ti­ates into a Mac. But the process is even more com­plex be­cause in the Mic many of those 30,000 genes are bro­ken into many seg­ments each – some­times more than 40 from one gene. The seg­ments are sep­a­rated by re­gions of non-cod­ing DNA. They aren't even arranged in the right or­der along the chro­mo­some, and some of them are aligned in the wrong ori­en­ta­tion. In that dis­rupted form, they could not be tran­scribed. But no big deal, be­cause gene tran­scrip­tion is the work of the Mac. How­ever, this does mean that when a Mic is trans­formed into a Mac, those genes have to be put to­gether from the frag­ments; the seg­ments must be joined to­gether in the cor­rect se­quence and ori­en­ta­tion to recre­ate the com­plete cod­ing se­quence for each gene.

TEM of Mac DNA iso­lated from an Oxy­tri­cha species. Bar = 1 μm. Source

Stranger still, in the Mac the genes are not on long chro­mosomes, but are dis­trib­uted among 25,000 dis­tinct minichro­mo­somes, mostly just one gene each (plus telom­eres). Each minichro­mo­some is then am­pli­fied a thou­sand-fold mak­ing for 25 mil­lion minichro­mo­somes in each Mac. That's a lot of chro­mo­somes to par­ti­tion to the daugh­ter Macs formed each time the cells di­vide. Al­though the Mic un­der­goes mi­to­sis, thus pre­cisely dis­trib­ut­ing one copy of each chro­mo­some to each daugh­ter cell, the Mac just divvies up the mil­lions of minichro­mo­somes by a some­what sloppy but un­known mech­a­nism.

When think­ing about how the Mic dif­fer­en­ti­ates into a Mac, two ques­tions pop up im­me­di­ately. How are the cuts made in the Mic genome? How is the use­ful 5% sep­a­rated from the junk 95% and the gene seg­ments then cor­rectly joined to­gether into func­tional genes? Two pa­pers from Nowacki and col­leagues give us some hints.

In a re­cent pa­per, they re­port that the cuts are made by the trans­posons in the germline genome. DNA trans­posons, first known as jump­ing genes, are mo­bile ge­netic el­e­ments that can move within a cell from one site in the genome to an­other. The type of trans­po­son found in Oxytricha does this us­ing the en­zyme trans­posase that is en­coded by the trans­po­son. The trans­posase ex­cises the trans­po­son out of the Mic DNA. In this case, the trans­po­son is dis­carded rather than be­coming rein­serted else­where in the genome. The trans­posases are made only when a Mic is in the process of be­com­ing a Mac. If you ex­per­i­men­tally block their ac­tiv­ity, the DNA is not cut nor­mally, larger chro­mo­somes are found in the Mac, and few of the cells sur­vive.

Now let's jump to the other ques­tion: how are the gene seg­ments as­sem­bled into mini-chro­mo­­somes in the cor­rect se­quence? In an ear­lier pa­per, the au­thors hy­poth­e­sized that Oxytricha might make a sin­gle-stranded RNA copy of each minichro­mo­some be­fore the old Mac is de­graded. One could imag­ine a mech­a­nism whereby these RNA copies could serve as a tem­plate. Gene seg­ments could base-pair with the com­ple­men­tary re­gion of the tem­plate and then be joined to­gether. When they went look­ing, these re­searchers found that such RNAs do ex­ist and, tellingly, only du­ring the hours when the DNA is be­ing processed for the new Mac. If these RNAs are in­deed guid­ing the cor­rect as­sem­bly of frag­mented genes, then in­tro­duc­ing RNA for such a gene with the seg­ments arranged in a dif­fer­ent or­der should re­sult in minichro­mo­somes that matched the in­jected RNA. They did this ex­per­i­ment, syn­the­siz­ing the RNAs in the lab and mi­croin­ject­ing them into the Oxytricha. The genes in the prog­eny Macs had been as­sem­bled in the man­ner spec­i­fied by the in­jected RNA, thus con­firm­ing a tem­plate role for the RNA.

In every or­gan­ism, DNA serves as both the in­for­ma­tion archive to be passed on to suc­ceed­ing gen­er­a­tions and as the tem­plate for RNA tran­scrip­tion. Leave it to the cil­i­ates to han­dle this in their own way – us­ing two dif­fer­ent genomes lo­cated in two dif­fer­ent nu­clei. The chore­og­ra­phy is in­tri­cate, but nev­er­the­less it all seems to work.

Ref­er­ences

Nowacki, M., Vi­jayan, V., Zhou, Y., Schotanus, K., Doak, T., & Landwe­ber, L. (2007). RNA-me­di­ated epi­ge­netic pro­gram­ming of a genome-re­arrange­ment path­way Na­ture, 451 (7175), 153–158 DOI: 10.1038/nature06452

Nowacki, M., Hig­gins, B., Maquilan, G., Swart, E., Doak, T., & Landwe­ber, L. (2009). A Func­tional Role for Trans­posases in a Large Eu­kary­otic Genome Sci­ence, 324 (5929), 935–938 DOI: 10.1126/science.1170023