Phage DNA: Go­ing with the Flow

by Merry Youle

We've heard it said so of­ten, it must be true. Af­ter tailed phages ad­sorb to their host and bind se­curely to their spe­cific re­cep­tor on the cell sur­face, they in­ject their DNA into the cell and the in­fec­tion is off and run­ning. This in­jec­tion no­tion arose nat­u­rally from the clas­sic im­age of a T4 phage with its con­trac­tile tail, poised sy­ringe-like on the sur­face of its soon-to-be host E. coli. The tail con­tracts, the DNA squirts into the cell. Sim­ple, but is it true? In par­tic­u­lar, it leaves us ask­ing where the force that moves the DNA comes from.

Phage T4 viri­ons poised on the sur­face of an E. coli cell. Cour­tesy of Cor­nell In­te­grated Mi­croscopy Cen­ter. Source

A pop­u­lar an­swer in­vokes the pres­sure in­side the phage cap­sid, and that pres­sure is not neg­li­gi­ble. The DNA within a T4 virion is es­ti­mated from in­di­rect mea­sure­ments to ex­ert a pres­sure on the cap­sid of about 60 Atm, the re­sult of be­ing pack­aged at very high den­sity (on the or­der of 500 mg ml^–1) and con­fined in tight quar­ters against its will. The dou­ble-stranded DNA (ds­DNA) of phage T4, for ex­am­ple, is about 50 μm long, whereas its cap­sid di­am­e­ter is only about 1/1000 of that (85 nm). Tightly coiled in­side the cap­sid, the DNA dou­ble he­lix pushes back due to its in­her­ent bend­ing re­sis­tance and the mu­tual re­pul­sion of its neg­a­tively-charged phos­phate back­bone. Stuff­ing all the DNA into the cap­sid in the first place, against ever-in­creas­ing re­sis­tance takes work. The DNA 'pack­ag­ing mo­tor' of phage φ29 can move its DNA into a cap­sid against a force of at least 100 pN (whereas mol­e­c­u­lar mo­tors such as dynein and ki­nesin ex­ert a force of only 5−7 pN).

A clas­sic dark field im­age of the DNA mol­e­cule re­leased by rup­ture of a sin­gle T4 cap­sid. (Two other in­tact viri­ons are also shown.) Source

When the T4 phage tail in­ter­acts with the re­cep­tor on the cell sur­face, the phage pops its cork;  a pro­tein plug that was pre­vi­ously block­ing the tail tube is re­moved. Now the DNA—one end al­ready within the tail tube—is pushed by the in­ter­nal pres­sure into the cell where it sets to work. This pro­posed ex­pla­na­tion, and its vari­ants, is known as a me­chan­ics model.

But there are a few prob­lems with this kind of ex­pla­na­tion. A ma­jor one is that as more and more of the DNA passes into the cell, the pres­sure within the cap­sid drops. As a re­sult, the rate of DNA en­try, ini­tially high, would be pre­dicted to de­crease con­tin­u­ously. Dur­ing the en­tire trans­fer, the force mov­ing the DNA is op­posed by the tur­gor pres­sure of the cy­to­plasm on the re­ceiv­ing end. This pres­sure, es­sen­tial for a bac­terium to grow, is al­ways present and is not neg­li­gi­ble: for Gram-neg­a­tive E. coli, it av­er­ages around 3.5–5 Atm; for Gram-pos­i­tive B. sub­tilis, it is ~19 atm. In ei­ther case, the last por­tion of the DNA would be go­ing up an os­motic hill. For the me­chan­ics model, this re­quires in­vok­ing some other mech­a­nism to get the DNA to the fin­ish line. If only we could sta­tion a ge­nie with a me­ter at the por­tal and have her mea­sure the rate at which DNA base pairs zip past through­out the trans­fer in vivo! But we're short of ge­nies and so far in vivo mea­sure­ment of these ki­net­ics has not been made (ex­cept for the stubby-tailed phage T7 and kin that trans­fer their genomes by an en­tirely dif­fer­ent and sig­nif­i­cantly slower mechanism…a story for a fu­ture post).

To take a vi­sual trip through the phage T4 tail tube, click here, and then dou­ble-click the im­age dis­played. This an­i­ma­tion is one of a col­lec­tion made by Steven Mc­Quinn, in­de­pen­dent sci­ence artist, based on re­con­struc­tions of T4 tail struc­tures.

More com­pli­ca­tions. Me­chan­ics mod­els can't ex­plain the trans­fer of sin­gle-stranded RNA or DNA phage genomes that are pack­aged at much lower den­sity and thought to not be un­der pres­sure. Nor can they ex­plain the trans­fer of phage in­ter­nal proteins…and all phages trans­fer at least some. At a min­i­mum, tail plug pro­teins have to go, and for the long-tailed phages, there is also a tape mea­sure pro­tein that has to get out of the way be­fore the DNA can exit via the tail tube. Some phages pack­age es­sen­tial pro­teins in the virion, pro­teins needed in the cell im­me­di­ately in or­der to get past host de­fenses and/or ini­ti­ate repli­ca­tion.

Ian Mo­lineux and col­leagues have pro­posed an­other model that  makes one think about this process in a dif­fer­ent fash­ion: a hy­dro­dy­nam­ics model. The ba­sic idea is re­ally quite in­tu­itive and re­calls prin­ci­ples we learned in high school sci­ence. In an aquatic en­vi­ron­ment the os­motic pres­sure will be higher in­side than out­side the cap­sid, so wa­ter will want to en­ter. It does so to a small de­gree, caus­ing a slight ex­pan­sion of the cap­sid, and then is coun­tered by the hy­dro­sta­tic pres­sure in­side. Af­ter the phage tail at­taches to the host re­cep­tors, the tail pen­e­trates the cell mem­brane. Now the tail pro­vides an open chan­nel through which wa­ter and small solutes can move from the en­vi­ron­ment through the virion and into the cell cy­to­plasm where the os­motic pres­sure is greater. (The os­motic pres­sure within the cap­sid does not mat­ter here, as the path taken does not al­ter the un­der­ly­ing process ef­fect­ing the wa­ter flow.) At the same time, other solutes in­clud­ing K⁺ leak from the cell and the mem­brane po­ten­tial is tem­porar­ily re­duced. As the wa­ter rushes down the tail tube into the cell, it brings the DNA—ultimately all the DNA—along with it by hy­dro­dy­namic drag. Pic­ture slurp­ing a strand of spaghetti through a straw. The amount of wa­ter com­ing in this way does not al­ter the cell's os­motic pres­sure sig­nif­i­cantly, so this model pre­dicts that all the DNA would trans­fer at the same rate.

Schematic of DNA ejec­tion in vivo ac­cord­ing to the hy­dro­dy­nam­ics model. (a) A phage when first ad­sorbed to its re­cep­tors on a host cell. (b) The plug block­ing the tail is opened or re­moved as the cell en­ve­lope is breached. Wa­ter (and small solutes) en­ters the phage cap­sid from the cul­ture medium and moves through the tail tube into the host cy­to­plasm, drag­ging the DNA along with it. Source

The leak­age of K⁺ pre­dicted by this model has been ob­served dur­ing DNA en­try. This was par­tic­u­larly ob­vi­ous for phage T5 that trans­fers its DNA in two steps with a pause in be­tween. The K⁺ leak­age oc­curs only dur­ing the trans­fers, not dur­ing the pause. But how come the cell doesn't keep on leak­ing K⁺ and other solutes? No one knows how the chan­nel is closed af­ter DNA en­try, but it surely must be. Oth­er­wise, with­out its mem­brane po­ten­tial the cell would lack the meta­bolic ca­pa­bil­i­ties for mak­ing more phage. Maybe our ge­nie at the por­tal can check on this for us, too.

Struc­tural or­ga­ni­za­tion of a PM2 virion. The out­er­most layer is an icosa­he­dral pro­tein shell com­posed of the ma­jor cap­sid pro­tein P2, or­ga­nized on a T=21 lat­tice, and the pen­tameric re­cep­tor-bind­ing pro­tein P1 oc­cu­py­ing the ver­tices. The lipid layer lies be­tween the cap­sid and the DNA. Mod­i­fied from this source.

Ac­tu­ally, I'd like to have a few more ge­nies avail­able, as not all phages rely on the same mech­a­nism for DNA translo­ca­tion. To il­lus­trate just how dif­fer­ent these mech­a­nisms can be, I men­tion phage PM2. It has a ds­DNA genome en­closed within a lipid mem­brane, both nested within the icosa­he­dral pro­tein shell. The DNA is a highly su­per­coiled cir­cu­lar mol­e­cule that is pack­aged at lower den­sity. This con­fig­u­ra­tion does not re­sult in the same pres­sure as ex­ists in­side some phage cap­sids. To de­liver that DNA, the phage takes ad­van­tage of its host's mech­a­nisms for DNA up­take from the en­vi­ron­ment. Those hosts (mem­bers of the genus Pseudoal­teromonas) are Gram-neg­a­tive ma­rine bac­te­ria that can grow with DNA as their sole nu­tri­ent source. Af­ter PM2 con­tacts its re­cep­tors on the host's sur­face, its pro­tein cap­sid falls apart ex­pos­ing the lipid mem­brane. That mem­brane fuses with the outer mem­brane, thereby de­posit­ing the DNA into the periplasm where it faces the pep­ti­do­gly­can layer. That trig­gers the cell's DNA up­take re­sponse. A pore forms in the cy­to­plas­mic mem­brane, al­low­ing the con­trolled re­lease of "cel­lu­lar lytic fac­tor" that de­grades the pep­ti­do­gly­can layer lo­cally. The phage DNA then en­ters the cy­to­plasm through that same pore. Other lipid-con­tain­ing phages do it dif­fer­ently. Phages with sin­gle-stranded RNA or DNA genomes are likely to use other strate­gies, and I'd guar­an­tee that those odd-ball hy­per­ther­mophilic cre­nar­chaeal viruses will have a few in­no­va­tions of their own.

Ref­er­ence

Panja D, Mo­lineux IJ. (2010). Dy­nam­ics of bac­te­rio­phage genome ejec­tion in vitro and in vivo. Phys­i­cal Bi­ol­ogy, 7 (4). PMID 21149974