Cell Di­vi­sion Through DNA Cur­tains

by Gemma Reguera

De­spite the ap­par­ent sim­plic­ity of bac­te­r­ial cells, their cell di­vi­sion cy­cle is a com­plex de­vel­op­men­tal pro­gram that cou­ples cel­lu­lar growth to the repli­ca­tion and seg­re­ga­tion of chro­mo­somes and the di­vi­sion of the cell's cy­to­plasm (aka cy­toki­ne­sis) (Fig. 1). The bac­te­r­ial cell di­vi­sion cy­cle starts with the com­mit­ment of the cell to re­pro­duce. This is the step in which, for­give the pun, size truly mat­ters. Dur­ing ac­tive growth, the cell's size changes to ac­com­mo­date in­creases in mass, vol­ume, and biosyn­thetic ca­pac­ity. When the cell reaches a crit­i­cal size, es­sen­tial cel­lu­lar func­tions such as in­tra­cel­lu­lar trans­port and nu­tri­ent up­take are con­strained and op­ti­mal growth can no longer be sup­ported. Pro­vided con­di­tions are ad­e­quate for growth, cell di­vi­sion is un­der­taken to al­low the bac­terium to re­gain its in­di­vid­ual cel­lu­lar fit­ness.

Fig. 1: As the B. sub­tilis cell grows in size, the chro­mo­some is repli­cated and seg­re­gated. A con­strict­ing sep­tum forms at mid­cell to di­vide the cell's cy­to­plasm and pro­duce two daugh­ter cells. Source

In Es­cherichia coli,as in most bac­te­ria, cell di­vi­sion starts with the for­ma­tion of the FtsZ or Z‑ring at mid­cell. FtsZ binds GTP and poly­mer­izes into fil­a­ments, which then as­sem­ble to form the Z‑ring. FtsZ is less likely to poly­mer­ize in re­gions of the cell with DNA, a process known as 'nu­cleoid oc­clu­sion.' In a grow­ing cell, the re­gions with lit­tle or no DNA are the mid­cell and near the poles. Pro­teins, such as the MinCDE sys­tem, os­cil­late lon­gi­tu­di­nally and co­op­er­ate to desta­bi­lize FtsZ poly­mer­iza­tion in the cell poles. As a re­sult, FtsZ can only sta­bly poly­mer­ize and form a ring at mid­cell. In ad­di­tion to mark­ing the mid­point of the cell, FtsZ also func­tions as a scaf­fold and re­cruit­ment site for the as­sem­bly of the cell di­vi­sion pro­teins (of­ten re­ferred to as the 'di­vi­some'). In E. coli, FtsZ re­cruits at least 10 pro­teins. Stud­ies of their lo­cal­iza­tion in vivo point at a se­ries of tem­po­rally or­dered re­cruit­ment events in which one pro­tein can­not as­sem­ble un­til all the up­stream pro­teins have done so (Fig. 2). How­ever, in vitro stud­ies ar­gue against a tem­po­ral hi­er­ar­chy and hint at a more flex­i­ble as­sem­bly process fa­cil­i­tated by the promis­cu­ous in­ter­ac­tions among com­po­nents of the di­vi­some.

Fig. 2: In E. coli, at least 10 pro­teins are se­quen­tially re­cruited to the FtsZ-ring to form the di­vi­some. The first pro­teins to be re­cruited (ZipA and FtsA) sta­bi­lize the FtsZ ring and re­cruit ZapA which fur­ther sta­bi­lizes the ring (see bot­tom pan­els C and D). These 'early' pro­teins also par­tic­i­pate in the re­cruit­ment of the down­stream di­vi­sion pro­teins, the so-called 'late' pro­teins (FtsK, FtsQ, Fts B and FtsL, FtsW, FtsI, and FtsN). Source

Also un­cer­tain is the bi­o­log­i­cal func­tion of many com­po­nents of the di­vi­some. The first pro­teins to be re­cruited to the Z‑ring (FtsA and ZipA) sta­bi­lize it and tether it to the mem­brane. This early as­sem­bly is fur­ther sta­bi­lized by pro­teins such as ZapA, which re­cruits ad­di­tional pro­teins to sta­bi­lize and bun­dle the Z‑ring (Fig. 2). These 'early' pro­teins also par­tic­i­pate in the re­cruit­ment of the down­stream di­vi­sion pro­teins, the so-called 'late' pro­teins (Fig. 2). The first of these, FtsK, is a mem­brane bound DNA pump. The func­tions of the oth­ers re­main largely un­known, ex­cept for the fact that one has transpep­ti­dase ac­tiv­ity and an­other con­tains a murein-bind­ing site, sug­gest­ing pos­si­ble roles in cell wall syn­the­sis dur­ing sep­ta­tion.

Fig. 3: Schematic view of the FtsK translo­case of the di­vi­some dur­ing en­dospore for­ma­tion in B. sub­tilis. FtsK cou­ples chromo-some repli­ca­tion and seg­re­ga­tion to sep­ta­tion and in­ter­acts with the XerCD pro­teins in or­der to re­solve chro­mo­some dimers. Modi-fied from source

As the DNA translo­case of the di­vi­some, Ftsk plays the crit­i­cal role in cou­pling chro­mo­some repli­ca­tion and seg­re­ga­tion to sep­ta­tion. Stud­ies of en­dospore for­ma­tion in the Gram-pos­i­tive bac­terium Bacil­lus sub­tilis have pro­vided some crit­i­cal in­sights into FtsK func­tion in vivo. The B. sub­tilis di­vi­some as­sem­bles asym­met­ri­cally to di­vide the parental cell into a daugh­ter cell (aka the mother cell) and a smaller cell, which will even­tu­ally dif­fer­en­ti­ate into a dor­mant struc­ture or spore for long-term preser­va­tion (Fig. 3). FtsK is a hexa­m­eric pro­tein that is teth­ered to the mem­brane and binds to the sep­tum through its N‑terminal re­gion (Fig. 3). A flex­i­ble linker con­nects the sep­tum-bound do­main to a pro­tein ring struc­ture that en­cir­cles and translo­cates the newly repli­cated DNA to the small cell us­ing en­ergy gained from hy­drol­y­sis of ATP. As the sep­tum has two dou­ble Z‑rings, there are two FtsK com­plexes at the mid­cell hold­ing one of the chro­mo­somes and pump­ing it into the small cell as DNA repli­ca­tion pro­ceeds (Fig. 3). Every so of­ten, about once in six di­vi­sion cy­cles, the newly repli­cated chro­mo­somes be­come in­ter­twined like the links of a chain. The chro­mo­so­mal ter­mi­na­tion sites ap­proach the FtsK pump but the chro­mo­somes are in­ter­locked. FtsK comes to the res­cue. It in­ter­acts with the XerCD pro­teins that are bound to spe­cific se­quences (dif sites) in the ter­mi­na­tion re­gion of the chro­mo­some. This in­ter­ac­tion ac­ti­vates the re­com­bi­nase ac­tiv­ity of XerCD, which re­solves the chro­mo­some dimer. The newly repli­cated chro­mo­somes are now un­tan­gled and can be seg­re­gated.

Fig. 4A: As­sem­bly of dou­ble-teth­ered DNA cur­tains in a mi­croflu­idic cham­ber. Source

Al­though we know a lot about FtsK func­tion com­pared to the other di­vi­sion pro­teins, some key ques­tions still re­main unan­swered. One such ques­tion is how FtsK in­ter­acts with DNA and translo­cates it in pre­cisely the right direction—towards the ter­mi­na­tion re­gion. It has been known for some time that FtsK binds spe­cific 8‑nucleotide se­quences in DNA termed KOPS (FtsK Ori­ent­ing Polar Sequences). These con­served KOPS se­quences abound in the chro­mo­somes of not only E. coli but also many other bac­te­ria. Since they are pref­er­en­tially ori­ented to­wards the ter­mi­na­tion re­gion of the chro­mo­some, they could pro­vide both a load­ing site for FtsK and a mech­a­nism for di­rec­tion­al­ity. Other stud­ies have sug­gested that FtsK can also load onto and translo­cate DNA in a KOPS-in­de­pen­dent man­ner, but still rec­og­nize KOPS se­quences and use their di­rec­tion­al­ity along the way to re­ori­ent it­self, if needed.

In a re­cent pa­per, Eric Greene's and David Sherratt's group­sused dou­ble-teth­ered "DNA cur­tains" to mon­i­tor FtsK's DNA-bind­ing and translo­ca­tion ac­tiv­i­ties in real time and thus an­swer some of the out­stand­ing ques­tions about its func­tion. DNA cur­tains are made in­side a flu­idic de­vice con­structed onto a sil­ica slide that con­tains nanofab­ri­cated chromium bar­ri­ers and an­chors, and is coated with a syn­thetic lipid bi­layer (Fig. 4A). The lipid bi­layer is treated with strep­ta­vidin, a pro­tein of bac­te­r­ial ori­gin that binds bi­otin (aka vi­t­a­min B7) with high affin­ity.

Fig. 4B: The first step in the as­sem­bly of DNA cur­tains in­volves the tether-ing of the DNA mol­e­cules to the lipid bi­layer via a streptavidin–biotin link­age and its dif­fu­sion un­der flow un­til they reach the dif­fu­sion bar­rier. Source

The re­searchers then pre­pared DNA mol­e­cules with a bi­otin-tagged end so it can bind the strep­ta­vidin. This teth­ered the DNA mol­e­cules to the lipid bi­layer, pretty much like hairs in their fol­li­cles (Fig. 4B). When flow is ap­plied, the teth­ered DNA mol­e­cules dif­fuse in the bi­layer un­til they reach the chromium bar­ri­ers. They then stretch with the flow of fluid and their free ends reach the nanofab­ri­cated an­chors. By la­bel­ing the free end of the DNA mol­e­cules with digox­i­genin (DIG) and coat­ing the an­chors with anti-DIG an­ti­bod­ies, the re­searchers were able to im­mo­bi­lize the free end of the DNA mol­e­cules to the an­chors as well. The re­sult is a cur­tain of hun­dreds-to-thou­sands of threads of DNA ap­prox­i­mately 10 µm long that can be im­aged with com­mer­cially-avail­able flu­o­res­cent dyes that have high affin­ity for DNA, such as YOYO1 (Fig. 5). Bi­otiny­lated trimeric forms of FtsK are then pro­duced via re­com­bi­nant tech­niques and la­beled with strep­ta­vidin quan­tum dots to en­able vi­su­al­iza­tion of their flu­o­res­cence. When added to the mi­croflu­idic cham­ber, one can see sin­gle FtsK mol­e­cules bind­ing and translo­cat­ing along the DNA threads in real time (Fig. 5).

Not only are these im­ages beau­ti­ful to look at, they also al­low us to peek into a process never be­fore vi­su­al­ized with such res­o­lu­tion and el­e­gance. Most im­por­tantly, this tech­nique pro­vided an­swers to many ques­tions still lin­ger­ing about FtsK func­tion. The re­searchers first demon­strated that FtsK pref­er­en­tially binds to KOPS se­quences in the DNA cur­tains, which is con­sis­tent with KOPS act­ing as load­ing sites. The data also sup­ported a model in which FtsK lo­cated KOPS through ran­dom col­li­sions, and ruled out a mech­a­nism based on bind­ing to non-spe­cific se­quences fol­lowed by long-dis­tance dif­fu­sion along the DNA strands, at least within the res­o­lu­tion lim­its of the tech­nique used. FtsK was pref­er­en­tially loaded onto the KOPS sites in its ADP-bound form. KOPS bind­ing was also stim­u­lated when ATP hy­drol­y­sis was pre­vented by added in­hibitors, lack of the Mg²⁺ co­fac­tor, or use of an AT­Pase-de­fi­cient FtsK mu­tant pro­tein. In­ter­est­ingly, this also caused FtsK to bind to non­spe­cific re­gions of DNA. Be­cause ATP hy­drol­y­sis pow­ers FtsK's translo­case ac­tiv­ity, it makes sense that it also sup­presses KOPS recog­ni­tion, as oth­er­wise FtsK would re­main stuck at its load­ing site. The au­thors at­trib­uted this feed­back mech­a­nism to an al­losteric process trig­gered by ATP hy­drol­y­sis that en­ables the mo­tor re­gion of FtsK to com­mu­ni­cate with the KOPS-bind­ing do­main.

Fig. 5: DNA mol­e­cules con­tain­ing KOPS sites arranged in var­i­ous ori­en­ta­tions (top) were aligned as dou­ble-teth­ered DNA cur­tains and vi­su­al­ized af­ter stain­ing with the flu­o­res­cent dye YOYO1. Ad­di­tion of FtsK tagged with quan­tum dots (ma­genta) en­abled the vi­su­al­iza­tion and quan­tifi­ca­tion of FtsK bind­ing and translo­ca­tion activi­ties. Source

The re­searchers also mon­i­tored the translo­ca­tion of FtsK from the KOPS load­ing site. Upon ad­di­tion of ATP to the cham­ber, the KOPS-bound FtsK be­gan to translo­cate along the DNA and it did so fol­low­ing the di­rec­tion dic­tated by the KOPS ori­en­ta­tion. Once in mo­tion, FtsK did not rec­og­nize KOPS se­quences en­coun­tered along the way, nor was it af­fected by their po­lar­ity. Yet it was able to change di­rec­tions abruptly while translo­cat­ing in­de­pen­dently of KOPS, sug­gest­ing that FtsK re­ori­en­ta­tion was not re­spon­sive to spe­cific se­quences in the DNA mol­e­cule but, rather, was a spon­ta­neous phe­nom­e­non. The ori­gin of the spon­ta­neous re­ver­sals is not known, and the au­thors rightly cau­tion that FtsK re­ori­en­ta­tion may not be bi­o­log­i­cally rel­e­vant. In vivo, FtsK is as­so­ci­ated with the di­vi­siome pro­tein com­plex, which likely re­stricts its re­ori­en­ta­tion dur­ing DNA pump­ing.

The moral of the story is that we know a lot, but not enough, about the es­sen­tial process of bac­te­r­ial cell di­vi­sion. The good news, how­ever, is that nan­otech­nol­ogy is pro­vid­ing us with tools that can help an­swer some (many!) of the long-stand­ing ques­tions, tools that ob­serve at tem­po­ral and spa­tial scales that have been tra­di­tion­ally viewed as un­at­tain­able. In a re­cent post, Elio spoke beau­ti­fully about the in­flux of physi­cists, math­e­mati­cians, and en­gi­neers into bi­ol­ogy in re­cent years, and the im­pact they are hav­ing in ad­vanc­ing our un­der­stand­ing of clas­si­cal bi­o­log­i­cal prob­lems. Let this story serve as an ex­am­ple of how the old and the new can come to­gether and com­bine in novel ways to ad­vance knowl­edge. The show has just started. Cur­tains up!

 

Ref­er­ence

Lee JY, Finkel­stein IJ, Crozat E, Sher­ratt DJ, Greene EC (2012). Sin­gle-mol­e­cule imag­ing of DNA cur­tains re­veals mech­a­nisms of KOPS se­quence tar­get­ing by the DNA translo­case FtsK. PNAS 109 (17), 6531–6536. PMID 22493241

 

Gemma is as­sis­tant pro­fes­sor in the De­part­ment of Mi­cro­bi­ol­ogy and Mol­e­c­u­lar Ge­net­ics, Michi­gan State Uni­ver­sity.