Evo­lu­tion-Proof­ing An­tibi­otics

by Vil­helmi­ina Haav­isto

Since the dis­co­ve­ry of pe­ni­cil­lin and its de­vel­op­ment in­to a the­ra­peu­tic agent, an­ti­bio­tics have re­vo­lu­tion­ized me­di­cine and saved count­less lives. How­ever, the way we use these life-sav­ing me­di­cat­ions is fun­da­men­tal­ly un­sus­tain­able as pa­tho­gens ra­pid­ly evolve re­sist­ance, ren­der­ing ma­ny se­ri­ous in­fect­ions ef­fect­ive­ly un­treat­able. The prob­lem is se­ri­ous and glo­bal; an­ti­bio­tic re­sist­ance has been cal­led an "in­vi­sible pan­de­mic," and could claim as ma­ny as 10 mil­lion lives an­nu­al­ly by 2050.

Part of the prob­lem lies pre­cisely in the very ef­fec­tive way in which an­tibi­otics work. Most in­hibit es­sen­tial com­po­nents of the cel­lu­lar ma­chin­ery, ef­fi­ciently killing sus­cep­ti­ble pathogens but also mem­bers of our com­men­sal mi­cro­biota. This ex­erts a mas­sive se­lec­tion pres­sure, where the only route to sur­vival is the evo­lu­tion of re­sis­tance. As such, we are ef­fec­tively locked in an evo­lu­tion­ary arms race with path­o­genic mi­croor­gan­isms whereby we dis­cover new an­tibi­otics, they evolve re­sistance to them, and the cy­cle re­peats over and over. This echoes the Red Queen hy­poth­e­sis, in­spired by Lewis Carroll's Through The Look­ing-Glass: "It takes all the run­ning you can do, to keep in the same place." While there are many promis­ing av­enues for an­tibi­otic dis­cov­ery, a Red Queen dy­namic is hardly ideal.

A new hope

Fig­ure 1. An­tibi­otic re­sis­tance poses a com­plex puz­zle for hu­man­ity to solve. (Ori­ginal im­age)

The need for so­lu­tions is greater than ever, and ide­ally, these should aim to break us out of the evo­lu­tion­ary arms race. Clearly, there is a need to de­velop ap­proaches that re­duce the risk of re­sis­tance evolv­ing in the first place. Such 'evo­lu­tion-proof' com­pounds, to which pathogens are un­able or un­likely to evolve re­sis­tance, would be more sus­tainable fu­ture al­ter­na­tives in the face of our cur­rent an­ti­biotic re­sis­tance cri­sis.

One 'evo­lu­tion-proof' op­tion is to tar­get and limit vir­u­lence, the dam­age that pathogens cause dur­ing an in­fec­tion, via an 'an­tivir­u­lence' ap­proach. Vir­u­lence is idio­syn­cratic to each in­fec­tion, but al­ways in­volves com­plex in­ter­ac­tions be­tween pathogen and host. The patho­gen's con­tri­bu­tion is of­ten through the pro­duc­tion of vir­u­lence fac­tors, which help them to suc­cess­fully col­o­nize and in­fect their host, and are the tar­gets of the an­tivir­u­lence ap­proach.

The key idea be­hind an­tivir­u­lence is that by dis­arm­ing pathogens rather than di­rectly killing them, the se­lec­tion pres­sure for re­sis­tance is re­duced or even elim­i­nated. By lim­it­ing vir­u­lence and thus re­duc­ing the neg­a­tive ef­fects of in­fec­tion on the host, with­out nec­es­sar­ily af­fect­ing pathogen ab­undance, the host im­mune sys­tem is given more time to clear the in­fec­tion. An­other ben­e­fit of an­tivir­u­lence ap­proaches is that they are un­likely to harm the host's com­men­sal mi­cro­biota, in con­trast to tra­di­tional an­tibi­otics whose spec­trum of ac­tiv­ity is of­ten very broad.

Vir­u­lence fac­tors make the in­fec­tion go 'round

Vir­u­lence fac­tors are es­sen­tial to the lifestyle of many pathogens. For ex­am­ple, Pseudomonas ae­ruginosa, an op­por­tunis­tic pathogen that com­monly in­fects the lungs of pa­tients with cys­tic fi­bro­sis, pro­duces the vir­u­lence fac­tor py­over­dine. Py­over­dine is an iron-bind­ing mol­e­cule (a sidero­phore) which helps P. aerug­i­nosa ob­tain that es­sen­tial nu­tri­ent, and is also in­volved in biofilm for­mation and con­trol­ling the ex­pres­sion of other vir­u­lence fac­tors.

This is where an­tivir­u­lence comes in. Py­over­dine pro­duc­tion can be in­hib­ited by flucy­to­sine, a drug usu­ally used to treat fun­gal in­fec­tions; such treat­ment sig­nif­i­cantly de­creased P. aeruginosa's vi­rulence in a mouse model of in­fec­tion. In the ex­per­i­ment, 80% of the an­i­mals treated with flucyto­sine sur­vived, com­pared to just 20% with­out any in­ter­ven­tion. The sur­vival rates fol­low­ing flucy­tosine treat­ment were sta­tis­ti­cally in­dis­tin­guish­able from sur­vival of an in­fec­tion with ge­net­i­cally en­gi­neered P. aerug­i­nosa that can­not pro­duce py­over­dine at all, pro­vid­ing more ev­i­dence that flucy­to­sine is an ef­fec­tive an­tivir­u­lence com­pound for this in­fec­tion.

Fig­ure 2. Pseudomonas aerug­i­nosa (yel­low) in­fect­ing lung cells (blue). CC BY-NC-ND 2.0 Laven­tie Benoit-Joseph

Sal­mo­nella en­ter­ica serovar Ty­phimurium is an­other patho­gen whose vir­u­lence fac­tors are es­sen­tial for its abil­ity to cause food poi­son­ing. Its suite of vir­u­lence fac­tors help Sal­monella es­tab­lish a foothold in the hu­man gut, and in­cludes se­cre­tion sys­tems for di­rect in­jec­tion of pro­teins into host cells and iron-bind­ing mol­e­cules to swin­dle the gut micro­bio­ta out this es­sen­tial nu­tri­ent. In­ter­est­ingly, by in­hibit­ing the pro­tein fold­ing ma­chin­ery that is es­sen­tial for the func­tioning of its vir­u­lence fac­tors, both the growth and fit­ness of S. Ty­phimurium can be sig­nif­i­cantly re­duced in vitro. How­ever, un­like the flucy­to­sine ex­per­i­ments with P. aeru­ginosa, the ef­fects of this pro­tein fold­ing in­hibitor have not been tested in an­i­mal mod­els of in­fec­tion.

A slightly dif­fer­ent an­tivir­u­lence ap­proach is to mop up the tox­ins that pathogens pro­duce. For ex­am­ple, bac­te­r­ial pathogens in the genus Strep­to­coc­cus, of­ten as­so­ci­ated with se­vere pneumo­nia and sep­sis, se­crete tox­ins that kill host cells by punch­ing holes in their mem­branes. These tox­ins can be se­questered in en­gi­neered li­po­somes, tiny vesi­cles that trap the tox­ins and stop them from harm­ing host cells. Mouse ex­per­i­ments have shown that al­though pathogen abun­dance in the lungs dur­ing Strep­to­coc­cus pneu­mo­niae in­fec­tion is un­af­fected, us­ing li­po­somes sig­nificantly en­hances host sur­vival, seem­ingly be­cause the host's im­mune sys­tem is bet­ter able to clear the in­fec­tion in the ab­sence of tox­ins.

None of this is to say that tra­di­tional an­tibi­otics will be­come ob­so­lete. In fact, they can be used in com­bi­na­tion with an­tivir­u­lence com­pounds, of­ten with pos­i­tive syn­er­gis­tic ef­fects. For in­stance, the sus­cep­ti­bil­ity of an­tibi­otic-re­sis­tant P. aerug­i­nosa in­creases when tra­di­tional an­tibi­otics are com­bined with an­tivir­u­lence com­pounds. This ap­proach can even re­verse re­sis­tance to some com­mon an­tibi­otics in vitro, high­light­ing its po­ten­tial for re­in­stat­ing drugs with high re­sis­tance rates. En­gi­neered li­po­somes can also be com­bined with an­tibi­otics, and such treat­ments achieve bet­ter host sur­vival com­pared to an an­tibi­otics-only ap­proach.

Nev­er­the­less, they re­sisted

Yet mi­crobes are ex­tremely re­silient. It may not sur­prise you that re­sis­tance to an­tivir­u­lence com­pounds, whereby pathogens re­cover pro­duc­tion or func­tion of their pre­cious vir­u­lence fac­tors, oc­curs. For ex­am­ple, P. aerug­i­nosa evolves re­sis­tance to flucy­to­sine through mu­ta­tions in the non-es­sen­tial en­zyme Upp, which is re­quired to ac­ti­vate the com­pound in­side the cell. Since Upp is non-es­sen­tial, mu­tat­ing it is no big deal for P. aerug­i­nosa, mak­ing it an eas­ily evolv­able re­sis­tance mech­a­nism. Mean­while, al­though re­sis­tance to the pro­tein fold­ing in­hibitors used to fight S. Typh­imurium does not ap­pear to evolve in vitro, the com­plex en­vi­ron­ment in­side hosts could very well yield a dif­fer­ent out­come.

Fig­ure 3. Sal­mo­nella Ty­phimurium cells (yel­low) in­vade a hu­man gut ep­ithe­lial cell (blue). CC BY 2.0 NIAID

Though some an­tivir­u­lence ap­proaches ap­pear promis­ing, whether any treat­ment can be com­pletely 'evo­lu­tion-proof' is de­bat­able. While some com­pounds may be more evolu­tionarily ro­bust than oth­ers, an­tivir­u­lence ap­proaches over­all may al­ways carry some risk of re­sis­tance evo­lu­tion re­gardless of the spe­cific mech­a­nisms in­volved. This is sim­ply be­cause they too ex­ert a se­lec­tion pres­sure on the popula­tion, forc­ing pathogens to evolve if they are to sur­vive and thrive. Be­cause vir­u­lence is so cen­tral to the path­o­genic lifestyle, it may be that the dif­fer­ence in se­lec­tive pres­sure com­pared to an­tibi­otics that kill di­rectly is ul­ti­mately quite small.

How­ever, there is nu­ance to this de­bate; some have called for a dis­tinc­tion be­tween the abil­ity of pathogens to evolve re­sis­tance to an­tivir­u­lence com­pounds, which may be in­evitable, and whether this re­sis­tance will spread rapidly due to strong se­lec­tion. The po­ten­tial of an­tivir­u­lence re­sis­tance to spread on a large scale, akin to that of tra­di­tional an­tibi­otic re­sis­tance, is still un­clear. This is cri­tical to de­ter­mine if we want to know whether we will soon find our­selves in an­other un­winnable arms race. In the mean­time, pro­mot­ing an­tibi­otic re­sis­tance aware­ness and re­spon­si­ble steward­ship is also ex­tremely im­por­tant; this can be ac­com­plished by ini­tia­tives like the World Health Or­ganization's an­nual An­timi­cro­bial Aware­ness Week, which runs from 18 to 24 No­vem­ber this year.

An­tivir­u­lence treat­ments: fever dream or dream-come-true?

De­vel­op­ing an­tivir­u­lence drugs is not a straight­for­ward process, and it is not yet clear how promis­ing re­sults from lab­o­ra­tory stud­ies will trans­late into treat­ments of real in­fec­tions. This is obvious­ly com­pli­cated by cost and dif­fi­cul­ties in car­ry­ing out clin­i­cal tri­als and get­ting drugs to the mar­ket, as well as eth­i­cal lim­i­ta­tions of study­ing re­sis­tant in­fec­tions in hu­mans. Nar­row-spec­trum antivi­ru­lence com­pounds also re­quire a more pre­cise re­search ap­proach in con­trast to tra­di­tional broad-spec­trum an­tibi­otics, as the goal is to de­velop treat­ments that are of­ten in­fec­tion-spe­cific. All in all, de­vel­op­ing drugs based on the an­tivir­u­lence con­cept is a Her­culean task, con­sid­er­ing the sheer num­ber of clin­i­cally rel­e­vant pathogens and vir­u­lence fac­tors.

This task, how­ever, is not im­pos­si­ble. There are cur­rently five an­tivir­u­lence drugs ap­proved by the US Food and Drug Ad­min­is­tra­tion (FDA), all tar­get­ing tox­ins made by pathogens caus­ing bot­u­lism, an­thrax, and se­ri­ous gas­troin­testi­nal in­fec­tions. How­ever, only 17 new an­tibi­otics have been ap­proved by the FDA in the last decade. This de­vel­op­ing dearth of new treat­ments, an­tivir­u­lence-based or not, em­pha­sizes our press­ing need for new and im­proved drugs to com­bat in­creas­ingly re­sis­tant in­fec­tions.

An­tivir­u­lence re­search is an ex­cit­ing field con­stantly yield­ing new dis­cov­er­ies and in­sights into pa­thogen ecol­ogy and evo­lu­tion. Over­all, un­der­stand­ing how dif­fer­ent pathogens re­cover vir­u­lence fac­tor ac­tiv­ity, and the rates at which this hap­pens and spreads, looks to be es­sen­tial for the road ahead. How­ever, the po­ten­tial of an­tivir­u­lence ap­proaches as an 'evo­lu­tion-proof' sil­ver bul­let for the an­tibi­otic re­sis­tance cri­sis is yet to be de­ter­mined. It might be wiser to think of them as 'evo­­lu­tion-aware' in­stead; ul­ti­mately, evo­lu­tion is an ex­tremely pow­er­ful force that is dif­fi­cult to over­come, even with "all the run­ning [we] can do." In the im­mor­tal words of Dr. Ian Mal­colm, Jeff Gold­blum's char­ac­ter in Juras­sic Park: "life, uh, finds a way."

 

Vil­helmi­ina Haav­isto is a Mi­cro­bi­ol­ogy and Im­munol­ogy MSc stu­dent at ETH Zürich in Switzer­land, where she works with ma­rine mi­crobes and mi­cro­bial com­mu­ni­ties.