Dial "V" for Mur­der

by Christoph

If you read this blog fre­quently and, per­haps, be­cause we "em­pha­size the un­usual and the un­ex­pec­ted phe­nom­ena", you may feel a lit­tle, Um.., bored to see yet an­other post on hor­i­zon­tal gene trans­fer (HGT). But wait, we've ac­tu­ally touched this topic in fewer than 30 of our 1,200+ posts (to be pre­cise: 1,247 as of De­cem­ber 31th, 2017 ). That's less than the es­ti­mated av­er­age num­ber of genes ac­quired via HGT over time by a small bac­te­r­ial genome of, say, 1,200 genes! So, I should not worry too much about re­dun­dancy when talk­ing about it again (at length, a fair warn­ing! ). We pre­vi­ously dis­cussed HGT by con­ju­ga­tion as it hap­pens, and by phage trans­duc­tion. Now I will dis­cuss HGT by trans­for­ma­tion, that is, by DNA up­take from the en­vi­ron­ment. This has been stud­ied in great de­tail in sev­eral sys­tems, and I could have stopped di­al­ing at "B" for Bacil­lus sub­tilis, at "H" for Haemophilus in­fluen­zae, at "N" for Neis­se­ria gon­or­rhoeae, or at "S" for Strep­to­coc­cus pneu­mo­ni­ae (think of Griffith's 'trans­form­ing prin­ci­ple' from 1928 ). I skipped di­al­ing "M" for Al­fred Hitch­cock, and ul­ti­mately chose "V" for Vib­rio cholerae. Be pre­pared for a crime movie par ex­cel­lence, one that out­per­forms Hitch­cock for thrill and sus­pense.

Keep in mind that the Ten Com­mand­ments don't ap­ply for bac­te­ria, and any men­tion of mur­der, killers, vic­tims, weaponry, etcetera is, there­fore, mere an­thro­po­mor­phiz­ing. That said, jump for­ward to 'Take #3: Street Ac­tion!' if you're keen to wit­ness the ac­tual mur­der scene. Oth­er­wise, con­tinue read­ing about nat­ural com­pe­tence and trans­for­ma­tion in 'Take #1: In the Base­ment', and about Type VI Se­cre­tion Sys­tems (T6SS) in 'Take #2: In­side the Ar­mory'. Watch the screen­play un­fold as the plot thick­ens...

Take #1: In the Base­ment

As sim­ple as it sounds, 'DNA up­take' is mech­a­nis­ti­cally as com­plex as con­ju­ga­tion or phage trans­duc­tion. In both Gram-pos­i­tive and Gram-neg­a­tive bac­te­ria, 'com­pe­tence' for DNA up­take re­quires the co­or­di­nated ac­tion of rough­ly two dozen genes (in­clud­ing some with reg­u­la­tory func­tions ) or­ga­nized within sev­eral oper­ons. The pic­ture is si­milar in Vib­rio – 'sim­i­lar' refers to a great deal of pro­tein ho­mol­ogy among the nu­mer­ous com­po­nents in var­i­ous species, de­spite vastly dif­fer­ent reg­u­la­tory cir­cuits – for which the present model for com­pe­tence in­volves 19+ genes, and, as core com­po­nents: 1. a type IV pilus (Tfp) (not to be con­fused with T4SS as Elio had pointed out here ) that does not at­tach to a sur­face for twitch­ing mo­tility but may rather act like a plug for the outer-mem­bra­ne pore, PilQ; 2. a periplas­mic pro­tein, ComEA, that binds non­specif­i­cally to dou­ble-stranded DNA; and 3. the in­ner mem­brane-bound ComEC com­plex (see here for a sche­ma­tic di­a­gram (Source )). Ac­cord­ing to this model, 'DNA up­take' starts with re­trac­tion of the Tfp pilus and pulling of the DNA through the outer mem­brane pore, PilQ, by a sup­posed 'Brown­ian ratchet' mech­a­nism (de­spite Richard Feyn­man hav­ing debunk­ed the pos­si­bil­ity of a 'Brown­ian ratchet' ). Upon en­try into the periplasm, the in­com­ing DNA is tight­ly bound and com­pacted by ComEA (so tightly that ComEA over­ex­pressed in the cy­to­plasm of  E. coli kills the cell by se­ques­ter­ing the nu­cleoid ), and then shut­tled to the ComEC in­ner mem­brane-pore com­plex. As in Bacil­lus sub­tilis, the in­com­ing DNA en­ters the cy­to­plasm of V. cholerae probab­ly sin­gle-stranded with the 3'-end head-on, but the re­spon­si­ble periplas­mic (mem­brane bound? ) ex­onu­cle­ase is still elu­sive. Upon en­try into the cy­to­plasm, the in­com­ing DNA is bound by sin­gle-strand bind­ing pro­tein (Ssb), by DprA (like Ssb an ss­DNA-pro­tect­ing pro­tein ), and by RecA, thus prim­ing it for re­com­bi­na­tion. While that's Vib­rio cholerae trans­for­ma­tion in a nut­shell, two miss­ing de­tails are still needed for the de­vel­op­ing 'screen­play' of the crime story.

Fig­ure 1. V. cholerae ComEA binds to DNA in vi­vo. Com­EA-mCherry aggrega­tion and foci for­ma­­tion af­ter the ad­dition of ex­ternal DNA. Com­pe­­­tence-in­duced cells with­out (no DNA) or with ex­ternal DNA (PCR frag­ment, 10.3 kb; λDNA, 48.5 kb; gDNA, ge­nomic DNA of var­i­ous length) were im­aged in the red (mCherry; lower row) or the phase con­trast chan­nel (phase; up­per row) to vi­su­al­ize ComEA-mCherry lo­cal­iza­tion. Ex­ter­nal DNA did not lead to foci forma­tion of peri­plas­mic mCherry alone (pre­ceded by the ComEA sig­nal se­quence; ss[ComEA]-mCherry). Scale bars: 2 µm. Mod­i­fied from Source

For one, syn­the­sis of the trans­for­ma­tion ap­pa­ra­tus is en­ergetically costly. It would make sense, there­fore, to trig­ger its syn­the­sis only un­der con­di­tions where ex­oge­nous DNA was po­ten­tially avail­able, for ex­am­ple, in biofilms. In their nat­ural aque­ous habi­tats, the vib­rios pref­er­en­tially col­o­nize the cara­paces (=ex­oskele­tons ) of crus­taceans. When they sense chitin, the ma­jor 'sugar‑y' macromole­cu­lar com­po­nents of the crus­taceans' shells which they can de­grade to earn their liv­ing, they at­tach to it and make a biofilm (see here: vib­rios on chitin par­ti­cle (Source )). As was found early on, com­pe­tence fol­lows. The in­te­gra­tion of sig­nals from chitin sens­ing, nu­tri­ent lim­i­ta­tion (catabo­li­te re­pres­sion ), and cell den­sity sens­ing in/for biofilm for­mation (quo­rum sens­ing ) trig­gers the ex­pres­sion of TfoX, a key reg­u­la­tor of com­pe­tence ex­pres­sion. De­spite this com­plex­ity, TfoX alone can elicit full com­pe­tence in high cell den­sity when ex­pressed from an in­ducible pro­moter. This is for­tu­nate for the de­vel­op­ment of this crime story among vib­rios be­cause it sim­pli­fies their mis-en-scène ex­perimentally.

Fig­ure 2.  Flu­o­res­cence mi­croscopy of chitin bead-co­lo­niz­ing cells of V. cholerae en­gi­neered to carry trans­lational fu­sions for the competen­ce pro­tein ComEA-mCherry (red) and the T6SS sheath pro­tein VipA-sfGFP (green), both encod­ed at their na­tive chro­mo­so­mal lo­cus. Merged sig­nals from the two flu­o­res­cence chan­nels (GFP+mCherry). Three as­sem­bled and/or con­trac­ted T6SS (VipA‑sfGFP sheath) struc­tures are in­di­cated by white ar­rows. Scale bars: 2 µm. Source

For the sec­ond key de­tail, in com­pe­tent Vib­rio cholerae  cells, ComEA translo­cates to the periplasm. Luck­ily, so does a func­tional ComEA‑mCherry fu­sion which can be read­ily vi­su­al­ized as seen in Fi­gures 1 + 2 (slightly brighter flu­o­res­cence in the cell pe­riph­ery is nor­mal when look­ing at a 2D pic­ture of a rod-shaped 3D ob­ject ). Upon adding pu­ri­fied DNA to such cells, the 'periplas­mic stain­ing' of the cells by ComEA‑mCherry co­a­lesces into one or a few bright spots that in­di­cate tight bind­ing to and com­pact­ing of the in­com­ing DNA (Fig­ure 1). Borgeaud et al. as­sume that these ComEA ag­gregates rep­re­sent the 'pawl' of the 'Brown­ian ratchet' men­tioned above. Any­ways, ComEA aggre­gation upon bind­ing to DNA is of prime im­por­tance for the de­vel­op­ing crime story as this 'be­hav­ior' pro­vides clear vi­sual  ev­i­dence for on­go­ing DNA up­take.

Take #2: In­side the Ar­mory

When you peek into the ar­mory of Pro­teobac­te­ria and Bac­teroidetes (=not all Gram-neg­a­tives, and also not all species of the Pro­teos ) you will im­me­di­ately spot their mas­ter­piece, the con­spic­u­ous Type VI Se­cre­tion Sys­tem (Fig­ure 3). As Zoued et al. char­ac­ter­ize it: "The T6SS is con­sti­tuted of an en­ve­lope-span­ning com­plex an­chor­ing a cy­to­plas­mic tubu­lar ed­i­fice. This tubu­lar struc­ture is evo­lu­tion­ar­ily, func­tion­ally and struc­turally re­lated to the tail of con­trac­tile phages. It is com­posed of an in­ner tube tipped by a spike com­plex, and en­gulfed within a sheath-like struc­ture. This struc­ture as­sem­bles onto a plat­form called "base­plate" that is con­nected to the mem­brane sub-com­plex." We have men­tioned the T6SS ear­lier in STC here and here, and in both posts fo­cused on its use as a weapon. Zoued et al. con­tinue:"The T6SS func­tions as a nano‑crossbow: upon con­trac­tion of the sheath, the in­ner tube is pro­pelled to­wards the tar­get cell, al­low­ing ef­fec­tor de­liv­ery." They should have said "pro­pelled into the tar­get cell", be­cause this is what ac­tu­ally hap­pens. T6SS is not a long-range weapon like a "cross­bow" : it works only at short dis­tances, when cells sit side-by-side. And in con­trast to the tails of phage par­ti­cles, the "in­ner tube" does not de­liver DNA through it but, like a poi­son ar­row, through "ef­fec­tor" pro­teins at­tached to its "spike", the arrow­head. Yet Zoued et al. point out rightly by us­ing a more neu­tral term that "ef­fec­tor de­liv­ery" is not nec­es­sar­ily bad news for the re­cip­i­ent. Al­though it mostly is, since most known ef­fec­tors are nuc­leic-acid-de­grad­ing, cell wall-de­grad­ing, or mem­brane-de­grad­ing en­zymes. Think of T6SS as a sort of forced, near-field com­mu­ni­ca­tion that the re­cip­i­ent cell can­not de­cline ac­cept­ing and that can serve var­i­ous pur­poses in­clud­ing in­va­sion or de­fense of a com­mu­nity, biofilm re­mod­el­ing, killing of phage-in­fected bac­te­ria, and killing of non-co­op­er­a­tors (the scuf­fle be­tween Pseudomonas and Vib­rio was fea­tured by Spencer Scott & John De Friel here in STC ). How­ever, a T6SS 'no­ti­fi­ca­tion' can turn into a mere 'sig­nal­ing' if the re­cip­i­ent cell has cog­nate im­mu­nity fac­tors at its dis­posal. Note that the "ef­fec­tor" cholera toxin (CT) en­coded by the CTXφ prophage of many V. cholerae strains is not de­liv­ered to hu­man in­testi­nal cells via T6SS – that's a com­pletely dif­fer­ent crime story.

Fig­ure 3. Overview of the T6SS.  A Elec­tron cryo-to­­mo­­graph of V. cholerae cells pro­duc­ing T6SS. The ar­row points the long cy­to­plas­mic tu­bular struc­ture cor­res­ponding to the T6SS sheath. A mag­ni­fi­ca­tion of the up­per part of the tomo­graph is shown in panel B, em­pha­siz­ing the pre­sence of dis­tinct com­plexes (green, tail sheath; pur­ple, base­plate; yel­low, pu­ta­tive mem­brane com­plex). Pa­n­­els A and B: Source. C Schematic re­presentation of the T6SS based on the elec­tron cryo-to­mo­graphs and on gene­tic, bioche­mical, mi­croscopy and struc­tural data (IM, in­ner mem­brane; PG, pep­ti­doglycan layer; OM, outer mem­brane). Scale bar is 50 nm. Source

The T6SS "nano-cross­bow" of Vib­rio cholerae  is a multi-pro­tein com­plex of >20 com­po­nents that in­cludes not only the me­chan­i­cal parts (sheath, in­ner tube, mem­brane an­chors ), but also the "ef­fector" pro­teins and the re­spec­tive (auto)immunity fac­tors needed for in­sur­ance against harm­ing one­self or one's sib­lings. The T6SS genes are or­ga­nized in three clus­ters, two on chro­mo­some II and one on the main chro­mo­some I. Re­cur­ring to the ar­gu­ment made for the com­pe­tence machi­nery (see 'Take #1: In the Base­ment' ): syn­the­sis of the T6SS ap­pa­ra­tus is en­er­get­i­cally costly, and it would make sense to trig­ger its syn­the­sis only un­der con­di­tions of high cell den­sity. And in­deed, T6SS ex­pres­sion is co-reg­u­lated by quo­rum sens­ing.  Borgeaud et al. iden­ti­fied, in ad­di­tion, TfoX as the key reg­u­la­tor, ex­actly the same fac­tor that trig­gers com­pe­tence ex­pres­sion (see their di­a­gram here ). Thus, ex­pres­sion of T6SS and com­pe­tence are tightly cou­pled, and the idea that T6SS leads to killing of neigh­bor­ing cells fol­lowed by hor­i­zon­tal gene trans­fer is, well, straight­for­ward. But how to demon­strate T6SS ex­pres­sion vi­su­ally ? Basler et al. had found ear­lier in their study of the T6SS sheath of Vib­rio cholerae  that one of its two com­po­nents, VipA, can be cou­pled to GFP with­out grossly af­fect­ing its prop­er­ties (ac­tu­ally they used a par­tic­u­larly well 'fold­able' ver­sion of GFP called 'su­per­folder GFP' (sfGFP )). This en­abled them to vi­su­al­ize sheath for­ma­tion (=as­sem­bly ), con­trac­tion, and re-as­sem­bly af­ter 'fir­ing' in live cells (see their stun­ning movie here ). To test their hy­poth­e­sis, Borgeaud et al. made a V. cholerae  O1 El Tor A1552 strain de­riv­a­tive with the genoty­pe TntfoX vipA-sfgfp comEA-mcherry. Upon in­duc­tion, this 'preda­tor' strain ex­presses T6SS and com­pe­tence, and the cells be­come de­tectable in the mi­cro­scope by green flu­o­res­cence (VipA sheaths ) and red flu­o­res­cence (periplas­mic ComEA ) (Fig­ure 2). To com­plete the cast­ing for the 'mur­der scene', they chose as 'prey' V. cholerae strain Sa5Y ΔvipA gfp, which is T6SS-de­fi­cient, lacks im­mu­nity to the tox­ins of A1552, and ex­presses GFP that is then dis­trib­uted through­out its cyto­plasm.

Take #3: Street Ac­tion!

I in­vite you now to the shoot­ing lo­ca­tion where film­ing of the key mur­der scene is about to be­gin, and will con­tinue for the next two hours. Cam­era is in po­si­tion; the ac­tors have re­hearsed their script, have been made up pro­fes­sion­ally, and are ea­ger to act. They have al­ready taken their po­sitions, gath­ered in close con­tact to each other: one 'blue' Vib­rio  preda­tor and three 'grey' Vib­rio  prey cells (t=0; shown schemat­i­cally in Fig. 4, bot­tom row ). Take your seat and keep an eye on the ac­tion in Fig­ure 4. Also keep in mind that the play­ers will grow (=elon­gate ) and di­vide over the next 120 min unless/until they're killed. Rec­og­nize the killer dressed for at­tack by the faint green GFP flu­o­res­cence of one of its T6SS-sheath build­ing blocks, VipA‑sfGFP, and pre­pared for DNA up­take by the red flu­o­res­cence of ComEA-mCherry dis­trib­uted in its periplasm (com­pare with Fig­ure 2 ). The three pre­sump­tive vic­tims ex­press GFP in their cy­to­plasm and are thus rec­og­niz­able by their cell-wide bright green flu­o­res­cence.

Fig­ure 4.  In­duced T6SS leads to the killing of neigh­boring cells fol­lowed by hor­i­zon­tal gene trans­fer. This fig­ure shows a time-lapse micro­scopy im­age se­ries of a T6SS-de­pen­dent at­tack of non­im­mune prey cells (prey, bot­tom row) fol­lowed by their round­ing and ly­sis and the upta­ke of their DNA by a neighbor­ing T6SS-com­pe­­tent preda­tory cells (pre­dator; bot­tom row). A vipA-sfgfp and comEA-mcherry Vib­rio cho­lerae strain (preda­tor) was grown to high cell den­sity un­der T6SS-in­duc­ing con­di­tions, mixed with the T6SS-neg­a­tive and gfp-la­beled V. cholerae strain Sa5Y (ΔvipA gfp) (prey), and spot­ted onto agaro­se pads for ima­ging. White ar­rows and arrow­heads de­note prey cells be­fore and af­ter at­tack (seen as cell round­ing), re­spec­tively, and black ar­rows in­di­cate cell de­bris af­ter ly­sis as visuali­zed in the phase con­trast im­ages (phase; top row). Accumu­lations of the VipA-sfGFP pro­tein and as­sembled + con­tracted T6SS sheath struc­tures are vis­i­ble in the green chan­nel (GFP; se­cond row). DNA re­leased by the prey ly­sis is ta­ken up by the com­pe­tent preda­tor, as in­di­cated by the gray ar­row­heads high­light­ing the re­dis­tri­bu­tion of periplas­mic ComEA-mCherry (mCherry; sec­ond last row). Scale bar: 2 µm in­di­cated in the up­per left ima­ge here and in the orig­i­nal pic­ture, and valid for all ima­ges). Mod­i­fied from Source

The ac­tion starts im­me­di­ately, at t=0 min, when the killer at­tacks vic­tim #1 (white ar­rows ). At t=30 min, vic­tim #1 has al­ready been stabbed and lysed (black ar­row ). Its GFP flu­o­res­cence di­lutes into the surround­ing medium and thus be­comes in­vis­i­ble. Vic­tim #2 is about to be stabbed (white ar­row ) and by t=60 min (white tri­an­gle ) has blown-up to a bleb (=pro­to­plast for­ma­tion due to pep­ti­do­gly­can dis­in­te­gra­tion, prece­ding ly­sis ). The killer is now busy stab­bing vic­tim #3 as seen by bright green streaks and dots (as­sem­bled and con­tracted T6SS sheaths ), while also tak­ing up the for­mer victim's DNA: the red flu­o­res­cence co­a­lesces into dots (gray ar­rows ). Al­though by t=90 and t=120 min there are no more po­ten­tial vic­tims within reach, the rag­ing killer keeps wield­ing its T6SS cross­bow while simultane­ously tak­ing up DNA from the de­ceased. Ac­tually, the killer has al­ready dou­bled once dur­ing the film­ing, and is about to di­vide again shortly af­ter t=120 min. From the ci­ne­matographic point of view: no stunts or spe­cial ef­fects re­quired here to make the mur­der scene scarier! Note that al­though vic­tim #3 got stabbed be­fore t=90 min and has promptly blebbed, it has as­ton­ish­ingly re­mained in this stage for >30 min. Al­though the screen­play does not de­mand the imme­diate death of vic­tim #3 dur­ing the take, we know that the cell is doomed (one could ea­sily spec­u­late that the burst of vic­tim #3 is de­layed due to com­plete cov­er­age of the 'zom­bie' pro­to­plast by all the DNA goo of vic­tims #1 and #2 ).

Fun aside, the imag­i­nary 'mur­der scene' I guided you through was just an ex­am­ple of the ef­forts Borgeaud et al. made to show the T6SS-killing of neigh­bor­ing cells fol­lowed by HGT. Now that you're well versed in the de­tails, you could peek at their 'crowd scene', from which the 'mur­der scene' is a cutout (the hi-res ver­sion was kindly sup­plied by Melanie Blokesch ). The ob­served kil­ling of neigh­bor cells and the tak­ing-up their re­leased DNA are, taken alone, not con­clu­sive proof of hor­i­zon­tal gene trans­fer (HGT). One must also demon­strate in­te­gra­tion of trans­ferred genes into the killer's chromosome(s). RecA‑dependent ho­mol­o­gous re­com­bi­na­tion would be the most ef­fi­cient route for such in­te­gra­tion, and Seitz and Blokesch had shown in an ear­lier study that in­deed RecA-GFP ac­cu­mu­lates (=forms foci ) close to the cy­to­plas­mic site where sin­gle-stranded DNA slips from the periplasm into the cy­to­plasm. To prove HGT, one needs to also demon­strate the re­com­bi­na­tion of some of this in­com­ing DNA into one of the two V. cholerae  chro­mo­somes and its sub­se­quent ex­pres­sion. To this end, the au­thors co-cul­tured a T6SS‑proficent V. cholerae  O1 El Tor A1552 (Rif^R) 'preda­tor' strain with a T6SS‑deficient and non‑immune V. cholerae  Sa5Y 'prey' strain tagged with a lacZ::aph (Kan^R) marker on chro­mo­some II un­der T6SS- and com­pe­ten­ce-in­duc­ing con­di­tions on LB-plates (=keep­ing the cells in in­ti­mate con­tact ). Af­ter colony forma­tion, they se­lected trans­for­mants on LB‑plates con­tain­ing kanamycin and ri­fampicin (note that se­lec­tion for an­tibi­otic re­sis­tance was not ap­plied dur­ing the HGT-pro­mot­ing step, a sub­tle but im­por­tant de­tail rem­i­nis­cent of the clas­si­cal se­lec­tion pro­ce­dure for transcon­ju­gants in Hfr-cross­es ). Dou­ble-re­sis­tant trans­for­mants (Rif^R+Kan^R) were then checked by PCR for the pres­ence of the aph  gene (Kan^R), and for the pres­ence of two V. cholerae  O1 El Tor-spe­cific mark­ers – not present in Sa5Y – on chro­mo­some I and chro­mo­some II, re­spec­tively. All trans­for­mants turned out to be V. cholerae  O1 El Tor A1552 (Rif^R) hav­ing ac­quired the aph  gene (Kan^R) via ho­mol­o­gous recombina­tion from the prey strain (Fig­ure 5). Con­clu­sion: al­though the au­thors did not catch the 'preda­tor' af­ter re­com­bi­na­tion in fla­grante delicto on cam­era, this trans­for­ma­tion ex­per­i­ment con­sti­tutes a legally in­con­testable con­fes­sion. They can thus right­fully claim to have shown HGT of a chromoso­mal an­tibi­otic re­sis­tance marker.

Take #4: A Side Story (Dial "A" for Mur­der)

T6SS‑mediated neigh­bor killing and gene ac­qui­si­tion by HGT is by no means con­fined to the vib­rios. Cooper et al. demon­strated in a very re­cent study that T6SS-pro­fi­cient and trans­for­ma­tion-com­pe­tent Acine­to­bac­ter bay­lyi  kill co-cul­tured E. coli  cells (to watch one of their videos, scroll to the end of this post for the video and the leg­end ) and ac­quire their an­tibi­otic re­sis­tance genes. HGT oc­curred at high fre­quency within a few hours un­der the cho­sen experi­mental con­di­tions (in mi­croflu­idic de­vices or on plate spots ) and newly re­sis­tant A. bay­lyi  cells rapidly displa­ced an­tibi­otic-sen­si­tive kin un­der se­lec­tive growth condi­tions. These find­ings may help to solve the rid­dle of how so many clin­i­cal iso­lates of the hu­man pathogen Acine­to­bacter bau­man­nii  ac­cu­mu­late mul­ti­ple an­tibi­otic resist­ance genes from other hu­man pathogens so rapidly (cau­tionary note: A. bay­lyi is a soil bac­terium and the regu­la­tion of com­pe­tence and T6SS ex­pres­sion might dif­fer con­si­de­rab­ly from that of the hu­man pathogen A. baumann­ii). Like­wise, the stud­ies per­formed over sev­eral years by the lab of Melanie Blokesch at the EPFL in Lau­sanne, Switzer­land, do not aim merely for a re­duc­tion­ist ex­pla­na­tion of HGT via trans­for­ma­tion in Vib­rio cholerae  (see 'Take #3: Street Ac­tion!' ), but also at us­ing this knowl­edge to bet­ter un­der­stand the mech­a­nisms that drive the re­peated re-emer­gence of this hu­man pathogen from the pool of its ex­tant (sub)spe­cies in their nor­mal ma­rine habi­tats (think of the re­cent cholera epi­demic af­ter the 2010 earth­quake in Haiti with up to now >700,000 in­fec­tions and ~9,000 deaths; but In this case, the V. cho­le­rae strain caus­ing the epi­demic was ap­par­ently im­ported from Asia by UN peace­keep­ing forces). That's for the fac­tual back­ground of my fic­tive 'mur­der story'.

Fig­ure 5. Dou­ble-re­sis­tant colonies are trans­formants of the preda­tor strain. PCR ampli­fi­ca­tion prod­ucts con­firming the trans­for­ma­tion of the preda­tor strain (lower two im­ages) by the prey-de­rived aph gene (kanamycin re­sis­tance cas­sette; up­per im­age). The schemes next to each im­age de­pict the prim­ing sites of the oli­go­nuc­leotides. Strains used for the colony PCR were the preda­tor (lane 1; WT, A1552-TntfoX Rif^R), the prey (lane 2; Sa5Y ΔvipA Kan^R), and ran­domly cho­sen trans­formants (lanes 3 to 8). L, DNA lad­der. chrI and chrII ab­bre­vi­ates V. cholerae chro­mo­some I and chromo­so­me II, re­spec­tively. Source

Take #5: in the Post­pro­duc­tion Stu­dio

It's a long-stand­ing no­tion (hotly de­bated, of course, be­cause it's sci­ence! ) that 'nat­u­rally compe­tent' bac­te­ria may take up chunks of DNA to sup­ple­ment their food stocks, or more pre­cisely, their pre­cious phos­phate and nu­cle­o­side pools. This might well hold true for en­vi­ron­men­tal DNA, which is of­ten rather de­graded (think of the DNAses that are present in any bac­te­ria-laden en­vi­ron­ment, that is, every­where ), but less so for DNA 'freshly' re­leased from lysing cells in close prox­im­ity to com­pe­tent ones. In this case, in­gested DNA, es­pe­cially when com­ing from close rel­a­tives, might in­stead be most use­ful for genome re­pair processes and gene ac­qui­si­tion via HGT by a com­pe­tent bac­terium, thus in­creas­ing its fit­ness. Maybe the above men­tioned vib­rios found a clever way to 'have their cake and eat it, too' ? Any in­com­ing DNA could eas­ily first serve for recombina­tion – re­pair or gene ac­qui­si­tion – and then the 'loose ends' of dou­ble-crossovers, along with non‑recom­bined pieces, used to re­plen­ish their phos­phate and nu­cle­o­side pools. Just a thought. Yet it re­mains a fact that the up­take of ex­oge­nous DNA is a vi­able path for HGT that can be vi­su­ally de­mon­­strated, just like con­ju­ga­tion or trans­duc­tion by phage. And that was the point of this post, a hap­py end­ing  for 'Dial "V" for mur­der'.

 

Ref­er­ences

Basler M, Pil­hofer M, Hen­der­son GP, Jensen GJ, Mekalanos JJ. 2012. Type VI se­cre­tion re­quires a dy­namic con­trac­tile phage tail-like struc­ture. Na­ture, 483  (7388), 182 – 186. PMID 22367545 (Open Ac­cess PDF here)

Borgeaud S, Met­zger LC, Scrig­nari T, Blokesch M. 2015. The type VI se­cre­tion sys­tem of Vib­rio cho­lerae fos­ters hor­i­zon­tal gene trans­fer. Sci­ence, 347  (6217), 63 – 67. PMID 25554784 (Open Ac­cess PDF via free reg­is­tra­tion)

Cooper RM, Tsim­ring L, Hasty J. 2017. In­ter-species pop­u­la­tion dy­nam­ics en­hance mi­cro­bial ho­ri­zontal gene trans­fer and spread of an­tibi­otic re­sis­tance. Elife, 6, e25950. PMID 29091031 (Open Ac­cess PDF here)

Lo Scrudato M, Blokesch M. 2012. The reg­u­la­tory net­work of nat­ural com­pe­tence and trans­for­ma­tion of Vib­rio cholerae. PLoS Genet, 8  (6), e1002778. PMID 22737089 (Ope­nAc­cess PDF here)

Seitz P, Blokesch M. 2014. DNA trans­port across the outer and in­ner mem­branes of nat­u­rally trans­formable Vib­rio cholerae is spa­tially but not tem­po­rally cou­pled. Mbio, 5  (4), e01409-14. PMID 25139903 (Open Ac­cess PDF here)

Seitz P, Pezeshgi Modar­res H, Borgeaud S, Bu­lu­shev RD, Stein­bock LJ, Rade­n­ovic A, Dal Per­aro M, Blokesch M. 2014. ComEA is es­sen­tial for the trans­fer of ex­ter­nal DNA into the periplasm in na­tu­rally trans­formable Vib­rio cholerae cells. PLoS Genet, 10  (1), e1004066. PMID 24391524 (Open Ac­cess PDF here)

 

Leg­end to linked video (Source): rep­re­sen­ta­tive mi­cro-colonies to high­light the en­rich­ment of E. coli  cell ly­sis at the bound­aries of mi­cro-colonies, where the cells are in con­tact with Acine­to­bac­ter  cells. E. coli  are shown in green (GFP) and Acine­to­bac­ter  in red (mCherry); in some cases the E. coli  mi­cro-colony is elim­i­nated (as in this video ), and in oth­ers it is heav­ily re­duced but sur­vives. Left pan­els: dif­fer­ence im­ages were cal­cu­lated for the green (E. coli ) chan­nel by sub­tract­ing the sub­se­quent time point from the cur­rent one. This shows where the GFP sig­nal dra­mat­i­cally de­creases, in­di­cat­ing cell ly­sis be­tween the two time points. These re­gions of puta­tive cell ly­sis are shown in yel­low. Right pan­els: the same movies, but with­out the yel­low dif­fer­ence sig­nal and in­clud­ing the trans­mit­ted light chan­nel, for com­par­i­son. Af­ter the mCherry sig­nal from Acine­to­bac­ter  fades (due to pro­longed laser ex­po­si­tion ), the lo­ca­tion of Acine­to­bac­ter  sur­round­ing the E. coli  can be seen in the trans­mit­ted light chan­nel. Note the timer in the up­per left cor­ner.