A Tale of Two Strate­gies

by S. Mar­vin Fried­man

A Love-Hate Re­la­tion­ship

Bac­te­ria are for the most part gre­gar­i­ous or­gan­isms, liv­ing pre­dom­i­nantly in dense com­mu­ni­ties con­sist­ing of mul­ti­ple strains. In fact, the ma­jor­ity of in­fec­tious bac­te­ria oc­cur as multi-lay­ered struc­tures called biofilms, many of which are com­posed of mul­ti­ple types of bac­te­ria. As one can imag­ine, neigh­bor re­la­tions in such com­plex pop­u­la­tions can be tricky. How­ever, since we mostly study pure cul­tures of in­di­vid­ual bac­te­r­ial strains in the lab, knowl­edge re­lat­ing to be­hav­ior within a com­plex pop­u­la­tion is very scarce. Now Stacy and co-work­ers elu­ci­dated the some­what thorny re­la­tions be­tween two mi­cro­bial neigh­bors – one com­men­sal and one path­o­genic – in a biofilm com­mu­nity that may be in­hab­it­ing your mouth.

Ag­gre­gat­i­bac­ter ac­tion­o­mycetem­comi­tans (Aa) is a non­motile, Gram-neg­a­tive fac­ul­ta­tive anaer­obe that con­tributes to pe­ri­odon­ti­tis, or gum dis­ease. In a polymi­cro­bial biofilm, Aa is a neigh­bor of the com­men­sal Strep­to­coc­cus gor­donii (Sg). These two have a love-hate re­la­tion­ship sort of. On the pos­i­tive side, Aa ben­e­fits from Sg by feed­ing on L‑lactate that Sg pro­duces, and has been shown to thrive in the pres­ence of Sg in a murine ab­scess in­fec­tion model. But here's where it gets prickly. An­other meta­bolic end prod­uct ex­creted by Sg is hy­dro­gen per­ox­ide (H₂O₂), which ac­cu­mu­lates to highly toxic lev­els. In or­der for Aa to hang around Sg, lap­ping up the L‑lactate it pro­vides, it also has to find a way to avoid its neighbor's more nox­ious out­put, H₂O₂. How does Aa ac­com­plish this feat?

Biofilms, For a Change

The re­searchers rea­soned that the an­swer lies in genes that are up­reg­u­lated dur­ing aer­o­bic res­pi­ra­tion, since this is the con­di­tion un­der which H₂O₂ is gen­er­ated. So, us­ing DNA mi­croar­ray analy­sis, they com­pared the Aa tran­scrip­tome un­der both oxic and anoxic growth. As pre­dicted, genes known to be in­volved in re­sponses to H₂O₂, in­clud­ing katA (en­codes a cata­lase that detox­i­fies H₂O₂) and apiA (en­codes an ad­hesin), were in­duced by oxy­gen. To their sur­prise, they also found that dspB, a gene en­cod­ing the biofilm dis­pers­ing en­zyme Dis­persin B (DspB), was coreg­u­lated with kat A and apiA. DspB helps de­grade the biofilm ma­trix, al­low­ing cells to de­tach and dis­perse from the colony. They also ob­served that a se­quence near the dspB pro­moter is sim­i­lar to the bind­ing site for OxyR, which is the tran­scrip­tional reg­u­la­tor of katA and apiA in re­sponse to H₂O₂. When they in­ac­ti­vated this site, dspB was no longer in­duced un­der high oxy­gen con­di­tions, in­di­cat­ing that dspB is reg­u­lated by the same fac­tor as katA and apiA. This find­ing added fur­ther sup­port to their hunch that DspB-me­di­ated dis­per­sal plays a role in al­low­ing Aa to deal with H₂O₂ se­creted by its neigh­bor, Sg. But how would this work?

Fig­ure 1. dspB me­di­ates dis­per­sal in re­sponse to oxy­gen. A Qual­i­ta­tive test tube as­say for mon­i­tor­ing biofilm dis­per­sal. (1) Aa is ini­tially grown as an anoxic shak­ing cul­ture to form a ring biofilm (brown), (2) the me­dia is re­placed with a larger vol­ume, (3) the cul­ture is fur­ther in­cu­bated un­der anoxic (‒O₂) con­di­tions, and (4) de­tach­ment is as­sessed af­ter crys­tal vi­o­let above the ini­tial biofilm (#1). B Rep­re­sen­ta­tive re­sult of the test tube as­say.

To com­pare what hap­pens to an Aa biofilm in the pres­ence and ab­sence of oxy­gen, the re­searchers came up with a sim­ple yet in­ge­nious test tube as­say (Fig. 1). Here's how it works: a shak­ing liq­uid cul­ture of Aa grown un­der anoxic con­di­tions forms a biofilm ring on the test tube just where the liq­uid reaches its max­i­mum height. The re­searchers rea­soned that if they now added liq­uid to this tube and shook it un­der oxic con­di­tions, DspB would be in­duced, al­low­ing the cells to de­tach from the biofilm and form a new ring, higher in the tube. This is in­deed the be­hav­ior they saw. Fo­cus­ing now on the role of H₂O₂ in dis­per­sal, the team com­pared biofilm mass formed un­der oxic con­di­tions in the pres­ence or ab­sence of H₂O₂ (Fig. 2). They saw that ad­di­tion of H₂O₂ in­duced Aa biofilm dis­per­sal in a DspB-de­pen­dent man­ner, since dis­per­sal oc­curred in a strain that car­ried the dspB gene, but not in a strain lack­ing dspB.

Fig­ure 2. dspB me­di­ates dis­per­sal in re­sponse to H₂O₂. (1) Aa is in­cu­bated sta­t­i­cally un­der anoxic con­di­tions to pro­mote at­tach­ment to the bot­tom of mi­crotiter dish wells, (2) biofilms are in­cu­bated un­der oxic shak­ing con­di­tions ei­ther with (+H₂O₂) or with­out H₂O₂ ad­di­tion, and (3) biofilm mass is quan­ti­fied us­ing crys­tal vi­o­let.

Back to the is­sue of neigh­bor re­la­tions. The re­searchers found that when Aa and Sg were grown in mixed-species biofilms un­der oxic con­di­tions, tran­scrip­tion of Aa dspB (just like that of katA) was en­hanced com­pared to mono­cul­ture growth. Fur­ther­more, this en­hance­ment was mostly due to Sg-gen­er­ated H₂O₂, since in­ac­ti­va­tion of the Sg gene re­spon­si­ble for H₂O₂ pro­duc­tion re­sulted in lev­els of Aa dspB tran­scrip­tion sim­i­lar to those found in Aa mono­cul­tures. These var­i­ous re­sults led the re­searchers to hy­poth­e­size that Aa uses the com­bined forces of the H₂O₂-de­stroy­ing cata­lase (from katA) and Dis­persin B (from dspB) to fend off, and flee from, Sg-ex­creted H₂O₂. In this way, Aa avoids be­ing killed by one prod­uct of Sg while pros­per­ing from an­other (L‑lactate).

Of Mice and Syn­ergy

In or­der to test their the­ory at an ac­tual in­fec­tion site, in this case a mouse ab­scess in­fec­tion, the re­searchers counted the num­ber of bac­te­ria in ab­scesses formed with Aa WT, Aa katA-, or Aa dspB-, alone or in the pres­ence of Sg, af­ter three days of growth (Fig. 3). As ex­pected, Aa num­bers were higher when Sg was present, but only in the WT strain. In con­trast, the katA- and the dspB- strains showed no en­hanced per­sis­tence in the pres­ence of Sg, al­though when grown alone, the mu­tant strains grew as well as or bet­ter than the WT strain. Sur­pris­ingly, they also found that Sg it­self grew more poorly in coin­fec­tion with ei­ther Aa katA- or dspB-, lead­ing them to hy­poth­e­size that per­haps Aa cross-pro­tects Sg, which does not pro­duce its own cata­lase, from self-in­flicted ox­ida­tive dam­age. Thus, the syn­ergy be­tween Aa and Sg ap­pears to be of a mu­tu­al­is­tic na­ture.

Fig­ure 3. dspB and katA are re­quired for polymi­cro­bial syn­ergy with Sg. Murine ab­scesses were formed us­ing Aa WT (Aa), katA−, dspB−, and Sg. At 3 d postin­fec­tion, ab­scesses were har­vested for enu­mer­at­ing vi­able bac­te­ria on se­lec­tive me­dia.

What role does DspB play in this sce­nario? Re­mem­ber that the re­searchers had hy­poth­e­sized that DspB en­ables Aa to dis­tance it­self from H₂O₂-pro­duc­ing Sg. To test this, they gen­er­ated ab­scesses us­ing strains of Aa and Sg ex­press­ing a green and a red flu­o­res­cent pro­tein, re­spec­tively, so that the two strains could be dis­tin­guished. Un­der a con­fo­cal mi­cro­scope, they saw that both strains grow as small, dense ag­gre­gates. When they mea­sured the dis­tance be­tween the ag­gre­gates, they found that Aa WT ag­gre­gates were lo­cated 4 ‒ 13 µm away from Sg ag­gre­gates. Aa dspB- ag­gre­gates, how­ever, were ei­ther right along­side or within 4 µm of Sg ag­gre­gates. These re­sults show that DspB en­ables Aa to put some dis­tance be­tween it­self and its H₂O₂-ex­cret­ing neigh­bor, af­fect­ing the or­ga­ni­za­tion of the biofilm com­mu­nity. The au­thors clev­erly char­ac­ter­ize Aa's han­dling of Sg-pro­duced H₂O₂ as a "fight-and-flight" re­sponse: cata­lase di­rectly detox­i­fies H₂O₂ (fight), and DspB al­lows Aa to move a safe dis­tance from the source (flight).

This study re­veals some­thing of the com­plex dy­nam­ics that can play out be­tween res­i­dents of a mixed-species biofilm. In­ter­est­ingly, Aa's two-pronged fight-and-flight re­sponse to ox­ida­tive stress also ap­pears to con­fer ben­e­fits to Sg, which grows bet­ter when the Aa genes en­cod­ing this re­sponse are in­tact. The re­mark­able syn­ergy be­tween these two neigh­bors serves to en­hance the fit­ness of the biofilm com­mu­nity as a whole. Now every­one is happy... ex­cept the un­for­tu­nate host!

 

Ref­er­ence

Stacy, A., Everett, J., Jorth, P., Trivedi, U., Rum­baugh, K., White­ley, M (2014). Bac­te­r­ial fight-and-flight re­sponses en­hance vir­u­lence in a polymi­cro­bial in­fec­tion Proc Natl Acad Sci USA, 111 (21), 7819–7824 DOI: 10.1073/pnas.1400586111

 

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