An­other Cell Bites the Dust

by Janie

Al­most a cen­tury ago, Walther Goebl and Os­wald Av­ery ob­served that Strep­to­coc­cus pneu­mo­niae cells lysed upon en­ter­ing sta­tion­ary phase, but the mech­a­nism be­hind this au­tol­y­sis re­mained elu­sive.

Fig­ure 1. Trans­mis­sion elec­tron micro­scopy im­age of a lysing bac­terium. Source

This chal­lenge is pre­cisely what a study by a team of re­searchers at Har­vard Med­ical School, the Uni­ver­sity of Sheffield, and the Howard Hughes Med­ical In­sti­tute tack­led. The study took a closer look at the role of pep­ti­do­gly­can (PG) hy­dro­lases in bac­te­ri­ol­y­sis. PG hy­dro­lases, also known as au­tolysins, cleave bonds in pep­ti­do­gly­can net­works, pro­vid­ing bac­te­ria with a means of cell wall re­mod­el­ing re­quired for growth and di­vi­sion. Their dou­ble-edged role in pok­ing holes in the cell wall was of in­ter­est: PG hy­dro­lases and their wall-cleav­ing ac­tiv­ity could be es­sen­tial to nor­mal growth un­der one set of con­di­tions, but lethal un­der an­other. Piec­ing to­gether the mech­a­nism of PG-hy­dro­lase-me­di­ated ly­sis is of par­tic­u­lar in­ter­est, as PG hy­dro­lases are also im­pli­cated in beta-lac­tam-in­duced ly­sis. Un­der­stand­ing ex­actly how an­tibi­otics like peni­cillin sub­vert bac­te­r­ial de­fenses, un­der­mine the in­tegrity of the cell wall, and ul­ti­mately cause ly­sis of pathogens would be valu­able in ad­vanc­ing ther­a­peu­tic de­vel­op­ments.

In­tro­duc­ing LytA and TacL: an en­zyme that wears two hats, and its in­hibitor

The re­searchers be­gan to un­ravel the role of PG hy­dro­lases in au­tol­y­sis. S. pneu­mo­niae was the per­fect mi­crobe for the job, as the bac­terium con­ve­niently con­tains only one PG hy­dro­lase, LytA, and is prone to au­tol­y­sis af­ter ex­tended time in sta­tion­ary phase.

Be­cause LytA lev­els stay con­stant through­out growth, from ex­po­nen­tial phase to sta­tion­ary phase to au­tol­y­sis, the team hy­poth­e­sized that there must be an ad­di­tional reg­u­la­tory com­po­nent that in­hibits LytA dur­ing ex­po­nen­tial growth. Such a reg­u­la­tor would be re­quired for growth in cells with func­tional LytA but un­nec­es­sary in cells lack­ing LytA. They per­formed a Tn-seq screen for reg­u­la­tors of the PG hy­dro­lase and found that the gene tacL was re­quired for sur­vival in wild-type S. pneu­mo­niae but dis­pens­able in ΔlytA strains. They con­firmed these re­sults by com­par­ing growth curves for strains with tacL on and off, find­ing that cells lack­ing TacL un­der­went pre­ma­ture au­tol­y­sis dur­ing ex­po­nen­tial phase. Thus, TacL must ex­ert in­hibitory ef­fects on LytA dur­ing ex­po­nen­tial growth.

Fig­ure 2.  Di­a­gram of Strep­to­coc­cus pneu­moniae cell en­ve­lope. The cell wall pepti­do­glycan (PG) (blue) con­tains Wall Te­i­choic Acid (WTA) poly­mers and the lipid bi­layer con­tains Lipote­i­choic Acid (LTA). The con­stituents of the re­peat­ing unit in LTAs and WTAs are in­di­cated; Cho, choline; Glc­Nac, N-acetyl­glu­cosamine; Rbo, ribitol; ATT­Gal, 2‑acetamido-4-amino‑2,4,6‑trideoxygalactose; Glc, glu­cose; Phos, phos­phate. . Source

The next step was to fig­ure out how ex­actly TacL reins in LytA's cell-wall-cleav­ing ac­tiv­ity. One pos­si­bil­ity was that TacL could im­pede se­cre­tion of LytA dur­ing ex­po­nen­tial growth. If this were true, then cells with and with­out TacL should both be equally sen­si­tive to the ex­oge­nous ad­di­tion of pu­ri­fied LytA. In align­ment with the team's pre­dic­tion, the ad­di­tion of LytA to ΔlytA cul­tures and to ΔlytA ΔtacL cul­tures in ex­po­nen­tial growth re­sulted in rapid ly­sis of the lat­ter cells. This showed that TacL pro­tects cells from death-by-LytA at a point af­ter LytA has al­ready been ex­ported to the cell sur­face.

Tip­ping the LTA-WTA scale

With two key en­zymes in the books, the next step was to de­ter­mine their re­la­tion­ship with po­ten­tial cell en­ve­lope sub­strates. Lipote­i­choic acids (LTAs), struc­tural com­po­nents of the lipid bi­layer, and wall te­i­choic acids (WTAs), struc­tural com­po­nents of the cell wall, are sus­pected to be syn­the­sized from the same un­de­caprenyl phos­phate pre­cur­sor in S. pneu­mo­niae. A pre­vi­ous study found ev­i­dence that the mem­brane pro­tein TacL at­taches the te­i­choic acid bit of the pre­cur­sor mol­e­cule to a gly­col­ipid an­chor, pro­duc­ing LTA (hence, TacL stand­ing for techoic acid lig­ase).

The team first con­firmed that TacL is re­quired for LTA for­ma­tion. They mea­sured the lev­els of LTA and WTA in the fol­low­ing strains dur­ing ex­po­nen­tial growth: WT, ΔlytA, ΔlytAΔtacL, and a dou­ble mu­tant with a Zn-in­ducible tacL al­lele, ΔlytAΔtacL+tacL. The re­sults? When LTA lev­els were down, WTA lev­els were up, and vice versa. The ΔlytAΔtacL cells en­tirely lack­ing a source of TacL pro­duced no LTA, but the amount of WTA de­tected in these same cells was sig­nif­i­cantly higher than that of any of the other three strains. This sug­gested that WTA and LTA syn­the­sis are in­versely re­lated, ad­di­tional ev­i­dence that they com­pete for the same un­de­caprenyl phos­phate pre­cur­sor.

So far, the team had elu­ci­dated that the ab­sence of TacL dur­ing ex­po­nen­tial growth causes the switch from LTA to WTA syn­the­sis and trig­gers an au­tol­y­sis path­way in­volv­ing LytA. Be­cause LytA has do­mains for cell wall cleav­ing ami­dase ac­tiv­ity and for bind­ing with te­i­choic acids, they guessed that LytA might lo­cal­ize to the cell wall in the ab­sence of TacL, grant­ing LytA more ready ac­cess to its WTA sub­strate to cause cell wall degra­da­tion and sub­se­quent cell ly­sis.

This hy­poth­e­sis was put to the test. The team tracked LytA lo­cal­iza­tion pat­terns dur­ing three stages of growth—exponential growth, sta­tion­ary growth, and late sta­tion­ary phase when au­tol­y­sis occurs—in both TacL+ and TacL- strains (in an in­ac­ti­vated LytA back­ground, LytA(H26A), to avoid com­pli­ca­tions aris­ing from au­tol­y­sis). As an­tic­i­pated, the TacL+ strain had more LTAs than WTAs dur­ing ex­po­nen­tial growth, and the ma­jor­ity of the LytA(H26A) were still found in the mem­brane. But as the cells reached late sta­tion­ary phase, LTA lev­els dropped while WTA lev­els soared, and LytA(H26A) was now lo­cal­ized to the cell wall, where it would be busy chew­ing away at the ma­trix for im­pend­ing ly­sis.

It seemed likely that peni­cillin-in­duced au­tol­y­sis would fol­low a sim­i­lar path­way. In­deed, re­peat­ing the ex­per­i­ment with LytA(H26A) cells ei­ther ex­posed or not ex­posed to peni­cillin con­firmed this guess, demon­strat­ing the same switch in LTA to WTA syn­the­sis and lo­cal­iza­tion of LytA to the cell wall.

Fig­ure 3. Trans­po­son in­ser­tions in tacL were found much more fre­quently in cells lack­ing lytA. Source

An­other en­zyme be­hind the cur­tain

The rapid dis­ap­pear­ance of LTAs ob­served in the last ex­per­i­ment was an ob­ject of in­trigue: per­haps there was an­other ac­tor in play, re­mov­ing ex­ist­ing LTAs from the cell mem­brane. So, the team mon­i­tored the LTA lev­els of LytA- cells in the mem­brane ver­sus su­per­natant and com­pared with that of LytA+ cells. At the start, LTAs were de­tected in the mem­brane. How­ever, upon en­try into late sta­tion­ary phase or ex­po­sure to peni­cillin, LTAs were hardly de­tectable in the mem­brane. There was also more choline-con­tain­ing ma­te­r­ial re­cov­ered from the su­per­natant, likely sourced from the LTAs. This sug­gested that LTAs are re­leased from lysing cells, al­low­ing LytA to di­rect its en­er­gies to­ward as­so­ci­at­ing with WTAs to cleave the cell wall.

ig­ure 4. Model for LytA reg­u­la­tion. Dur­ing ex­po­nen­tial growth, TacL-de­pen­dent LTA syn­the­sis dom­i­nates over LCP-de­pen­dent pro­duc­tion of WTA. LytA bound to the phos­pho­choline moeities (purple/black balls) on LTA is se­questered away from WTA pre­vent­ing LytA from tar­get­ing the cell wall. Upon en­try into sta­tion­ary phase or ex­po­sure to cell wall tar­get­ing anti­biotics (peni­cillin), TacL is de­graded in an FtsH-de­pen­dent man­ner lead­ing to a red­uction in LTA syn­the­sis and an in­crease in WTA lev­els. As­so­ci­a­tion of LytA with newly syn­the­sized WTA leads to cell wall cleav­age and ly­sis. Source

The next part of the puz­zle was to un­cover how the LTA to WTA syn­the­sis switch is ini­ti­ated. Sus­pect­ing that TacL was the pri­mary reg­u­la­tory tar­get, the team mon­i­tored the ef­fect of tacL over­ex­pres­sion on au­tol­y­sis in­duc­tion. In both late-sta­tion­ary-phase cul­tures and peni­cillin-treated cul­tures, tacL over­ex­pres­sion was found to pre­vent au­tol­y­sis, which fur­ther sug­gested that a re­duc­tion in TacL lev­els does in­deed cause the au­tolytic switch from LTA to WTA syn­the­sis.

Upon FLAG-tag­ging  TacL to track its lev­els in dif­fer­ent stages of growth with and with­out peni­cillin ex­po­sure, the team found that TacL lev­els de­creased with the start of au­tol­y­sis only in the strain with func­tional LytA. The half-life of TacL-FLAG in ΔlytA dropped upon en­ter­ing late sta­tion­ary phase and ex­po­sure to peni­cillin, sug­gest­ing that TacL is de­graded dur­ing au­tol­y­sis. So, what pro­tease is re­spon­si­ble for get­ting rid of TacL? A sec­ond Tn-seq screen pin­pointed the gene ftsH. FtsH is an ATP-de­pen­dent zinc met­al­lopep­ti­dase that has al­ready been shown to de­grade mem­brane pro­teins in a dif­fer­ent study, so it was rea­son­able to sus­pect that FtsH could be the TacL-de­grad­ing cul­prit.

Con­firm­ing their hunch, a com­par­i­son of lev­els of TacL-FLAG, LTA, and WTA in FtsH+ and FtsH- cells demon­strated that FtsH- cells dis­played none of the usual signs of im­pend­ing au­tol­y­sis: no de­crease in TacL-FLAG lev­els, no switch from LTA to WTA syn­the­sis: zilch, un­like the FtsH+ cells.

The path­way re­solved

So, LytA de­grades the cell wall, TacL is the lig­ase that syn­the­sizes LTAs, and FtsH de­grades TacL. At last, the model for S. pneu­mo­niae ly­sis comes to­gether. In au­tolytic con­di­tions, FtsH's degra­da­tion of TacL de­creases the lev­els of LTAs present (which is paired with an in­crease in WTAs). With­out LTAs to bind and se­quester LytA, LytA is now able to lo­cal­ize to the cell wall, where it can then as­so­ciate with WTAs to ig­nite its cell-wall-cleav­ing ac­tiv­ity. In this way, the cell wall is dis­man­tled.

The next step is to zero in on what trig­gers FtsH-me­di­ated degra­da­tion of TacL. This new un­der­stand­ing could prove to be a trump card in the per­pet­ual quest to bring down the bac­te­r­ial cell wall of pathogens.

 

Ref­er­ence

Flo­res-Kim, J., Do­bi­hal, G. S., Fen­ton, A., Rud­ner, D. Z., & Bern­hardt, T. G. (2019). A switch in sur­face poly­mer bio­gen­e­sis trig­gers growth-phase-de­pen­dent and an­tibi­otic-in­duced bac­te­ri­ol­y­sis. eLife, 8, e44912. doi:10.7554/eLife.44912

 

Other Posts