Mechanisms Biofilm Resistance

7/27/2019 Mechanisms Biofilm Resistance

http://slidepdf.com/reader/full/mechanisms-biofilm-resistance 1/6

TRENDS in Microbiology Vol.9 No.1 J anuary 200134 ReviewReviewReviewReviewReview

http://tim.trends.com 0966-842X/01/$ – see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S0966-842X(00)01913-2

Review

Thien-Fah C.Mah

George A.O’Toole*

Dept of Microbiology and

Immunology,

Dartmouth Medical

School,

Hanover, NH 03755, USA.

*e-mail:

[email protected]

B acteria l biofilms a re formed when un icellular

organisms come together to form a community tha t is

at ta ched to a solid surface an d encas ed in an

exopolysaccharide ma trix. B iofilms ca n be m ad e up of

single or mult iple ba cteria l species. For example, it ha s

been estimat ed tha t denta l biofilms contain

>500 different bacterial ta xa1; conversely, in t he lat ter

sta ges of the disease, the prima ry bacterium in the

lungs of cystic fibrosis (CF) pa tient s is Pseudomonas

aeruginosa . It ha s been observed tha t th e resista nce ofbiofilms t o ant ibiotics is increased compared w ith w ha t

is normally seen wit h plankt onic cells. In fact, w hen

cells exist in a biofilm, th ey ca n become 10–1000 times

more resista nt t o the effects of antimicrobial a gents2–5.

It is becoming increasingly clear tha t biofilms ha ve

a n enormous impact on medicine. Biofilms can form

on many medical implant s such as catheters,

ar tificial hips and conta ct lenses a nd, owing to their

increased resista nce to ant imicrobial a gents, these

infections can often only be trea ted by remova l of the

implan t, thus increasing th e trauma to the pat ient

an d the cost of treatm ent. It ha s been estima ted tha t

biofilms a re a ssocia ted w ith 65%of nosocomialinfections 6 an d tha t t reat ment of these biofilm-based

infections costs >$1 billion an nua lly7–9.

The development of biocide resist a nce is not

underst ood, but recent st udies have used a va riety of

model syst ems to determine how a nd wh y biofilms are

so resistant to ant imicrobial a gents. As the

importa nce of biofilms in nosocomia l infections ha s

increased, much energy ha s been directed towa rds th e

study of the effects of an timicrobial a gents on th ese

surfa ce-a tt a ched communities. The key quest ion we

ask in this review is: wha t a re the mechan isms of

biofilm resista nce to a ntimicrobial compounds? Wha t

we w ill empha size is tha t t here are multiple

mechanisms, w hich va ry w ith the ba cteria present in

the biofilm a nd t he drug or biocide being a pplied.

These mechan isms include physical or chemica l

diffusion bar riers to ant imicrobial penetra tion into the

biofilm, slow growt h of the biofilm owing to nut rient

limita tion, activat ion of the genera l stress response

a nd t he emergen ce of a biofilm-specific phenoty pe. In

this review, w e will focus our a tt ention on in vi tro -

derived sing le-species biofilms, a lth ough some dua l-

species biofilm w ork will be high light ed.

Failure of the antimicrobial to penetrate the biofilm

The production of a n exopolysa ccha ride ma tr ix, or

glycocalyx, is one of the distinguishing chara cteristics

of biofilms. It h as been suggested tha t t his ma trix,

a mong other functions, prevents t he a ccess of

a ntibiotics to the ba cterial cells embedded in th e

communit y. We will highlight a few of the m ore recent

studies on t he subject of ant ibiotic diffusion t hrough a

biofilm. For a more comprehen sive review of this

subject, the reader is directed to a review by S tew a rt 10.

Eit her rea ction of the compound w ith, or sorption

to, the components of the biofilm can limit th e

tra nsport of a ntimicrobial a gents to the cells withinthe biofilm. Although ma thema tical models suggest

tha t, for many a ntibiotics, there should be no barrier

to their diffusion into a biofilm, some studies ha ve

shown a n a pparent failure of certa in an timicrobial

a gents t o penetra te th e biofilm. Chlorine, a commonly

used disinfecta nt, did not rea ch >20%of th e bulk

media’s concentra tion with in a mixed Klebsiella

pneumoniae a nd P. aeru ginosa biofilm, as mea sured

by a chlorine-detecting microelectrode11. In fact, the

penetra tion profile wa s suggestive of a subst ra te being

consumed within the ma trix. Suci et al . used infra red

spectroscopy to show t ha t t he rat e of tra nsport of the

a nt ibiotic ciprofloxacin to th e surfa ce of a colonizedsurface was reduced compared w ith tra nsport to a

sterile surface12. These authors suggested tha t t he

ciprofloxacin w as binding t o the biofilm components.

Other groups have ta ken different a pproaches t o

add ress the question of whether t he biofilm a cts as a

barrier to ant imicrobial a gents. On the one hand,

P. aeru ginosa biofilms wer e formed on one side of a

dialysis membrane a nd the a mount of piperacillin tha t

penetra ted the biofilm w as measured. Consistent with

the results discussed a bove, the P. aeru ginosa biofilm

prevented diffusion of this a ntibiotic13. On the other

hand, Staphyl ococcus epiderm id is biofilms formed in a

similar m a nner a llowed for the diffusion of rifampicin

Biofilms are communities of microorganisms attached to a surface.It has

become clear that biofilm-grown cells express properties distinct from

planktonic cells,one of which is an increased resistance to antimicrobial

agents.Recent work has indicated that slow growth and/or induction of an

rpoS -mediated stress response could contribute to biocide resistance.The

physical and/or chemical structure of exopolysaccharides or other aspects of

biofilm architecture could also confer resistance by exclusion of biocides from

the bacterial community.Finally,biofilm-grown bacteria might develop a

biofilm-specific biocide-resistant phenotype.Owing to the heterogeneous

nature of the biofilm,it is likely that there are multiple resistance mechanisms

at work within a single community.Recent research has begun to shed lighton how and why surface-attached microbial communities develop resistance

to antimicrobial agents.

Mechanisms of biofilm resistance toantimicrobial agents

Thien-Fah C.Mah and George A.O’Toole



TRENDS in Microbiology Vol.9 No.1 J anuary 2001

http://tim.trends.com

35ReviewReviewReviewReviewReviewReview

an d van comycin a cross the membrane14, implying tha t

th ese an tibiotics could efficiently penetr a te t his biofilm.

These results sugg est t ha t inh ibition of diffusion can not

a lwa ys explain resista nce to ant imicrobial compounds.

A difference betw een thick an d th in biofilms an d

th eir resista nce to ant ibiotics has been observed.

P enetra tion of a t hin biofilm-covered bead [a vera ge

cell densit y ~ 3.5 log colony-forming un its (cfu) cm−2

]by hyd rogen peroxide wa s observed directly, even

th ough the cells wit hin th e biofilm were more

resistant to the compound compared w ith plankt onic

cells15. B y contrast , th icker biofilms, grown on glass

slides (a vera ge cell density ~ 7.6 log cfu cm−2),

presented a barrier to th e penetra tion of hydrogen

peroxide. Interest ingly, hydr ogen peroxide wa s able

to penetra te a thick biofilm formed by a mut an t st rain

of P. aeru ginosa tha t lacked one of the ma jor cat ala se

genes, katA (Ref. 16). As cata lases a re enzymes tha t

neutr a lize hydr ogen peroxide, th is result suggested

th a t, in t hick biofilms, cells were protected from

hydrogen peroxide penetra tion by th e cata lase-media ted dest ruction of this compound.

Anderl et al . formed K. pneumoniae colony biofilms

on aga r plates with or without a ntibiotic17. By placing

a filter at the t op of the colony, essentia lly

sandw iching the colony, they w ere able to assay

directly for a ntibiotic diffusion from the a ga r plat e

thr ough the colony by performing a sta nda rd zone of

inhibition a ssa y wit h the filter. This breakth rough

study showed tha t a mpicillin wa s unable to penetra te

the biofilm a nd th a t t he production of the a mpicillin-

degrading enzyme β-lacta ma se wa s responsible for

this phenomenon, as t he ampicillin wa s able to

penetra te a biofilm formed by a β-lactama se muta nt.Surprisingly, th e β-lactama se mutant s grown in a

biofilm were still resistan t t o ampicillin, suggesting

tha t other mechanisms contr ibute to the resista nce of

these cells. Furt hermore, ciprofloxacin w as able t o

penetrat e the biofilm, yet, as w as the case with

a mpicillin, it w a s una ble to kill the biofilm bacteria 17.

This simple method a llowed for t he differentiat ion

between tra nsport effects a nd other mecha nisms and

thu s provides a powerful tool for the further a na lysis

of the molecular mechanism of biofilm resista nce to

an timicrobial agents.

From these studies, an d others, it is clear t ha t th e

exopolysa ccha ride ma trix (or other components ofbiofilms) does not form a n impenetra ble ba rrier to th e

diffusion of a ntimicrobial a gents, a nd other

mechan isms mu st be in place to promote biofilm cell

surviva l. However, for certa in compounds, the

exopolysaccha ride ma trix does represent a n initia l

barrier tha t can delay penetra tion of the

a nt imicrobial a gent. The experiment s described

above strongly suggest tha t multiple mechanisms a re

required for overa ll a ntim icrobial resista nce.

Slow growth and the stress response

When a ba cterial cell cultur e becomes sta rved for a

part icular n utrient, it slows its growt h. Tra nsition

from exponentia l to slow or no growt h is genera lly

accompanied by a n increase in resistance to

antibiotics 18,19. Slow growth of the bacteria ha s been

observed in mat ure biofilms20,21. B eca use cells

growing in biofilms a re expected to experience some

form of nutrient limitat ion, it ha s been suggested tha t

th is physiologica l cha nge can a ccount for the

resista nce of biofilms to a nt imicrobial a gents.B y paying close at tention to the growt h phase of

plankt onic cells an d biofilm cells, recent stud ies ha ve

been able t o exa mine specifically t he contribution of a

slow growth rat e to biofilm cell survival a gainst

an tibiotics. Gilbert a nd collea gues examined

growt h-ra te-relat ed effects und er controlled growt h

condit ions for pla nkt onic cultures a nd biofilms of

P. aeru ginosa , Escher ichi a coli a nd

S. epidermi dis 22–24. They ma de the general

observation tha t t he sensitivities of both the

plankt onic a nd biofilm cells to either tobra mycin or

ciprofloxacin increased with increa sing growth ra te,

thus supporting th e suggestion tha t t he slow growt hra te of biofilm cells protects th e cells from

an timicrobial a ction. For P. aeru ginosa at slow

growth r at es, both the plankt onic and int act biofilm

cells were equa lly resista nt t o ciprofloxacin. However,

as t he growth ra te wa s increased, the planktonic cells

beca me more susceptible to ciprofloxacin t ha n t he

biofilm cells. This result support s th e idea t ha t some

other propert y of the biofilm, and not just growin g

slowly, w as importa nt for t he observed recalcitrance

of biofilms to a nt imicrobial tr eat ment 22. The sa me

group reached a simila r conclusion w hile working

wi th Bu rk holderia cepacia . Desai et al . compared the

resista nce of plankt onic a nd biofilm cells at differentsta ges during exponential growth up to the entry into

stat ionary phase25. They found tha t resista nce

increased a s t he planktonic cultures a nd t he biofilm

cells approa ched st at iona ry pha se. The ma ximal

resista nce of both cultures occurred in sta tiona ry

pha se wh ere the biofilm cells were 15-times m ore

resista nt t ha n the plan ktonic cells. These results

suggested that some determinant other tha n growt h

ra te is responsible for a certa in level of resista nce, and

slow growth a dds a dditiona l protection. This

determinant could be relat ed to the fact t ha t cell

density increases dur ing this la te sta ge of exponential

growth (see below). Other studies ha ve suggested tha tmechan isms differ for different a nt ibiotics. For

example, although the slow growth ra te in a

P. aeru ginosa biofilm seemed t o account for biofilm

resista nce to tet ra cycline, it did not seem to a ffect

resistance to t obra mycin26.

Heterogeneity

The experiment a l conditions resulting in t he tight

cont rol of growt h described in the stud ies

summ a rized above a llow ed investiga tors to focus on

th e effect of a specific growt h ra te on bacteria l

susceptibility t o an timicrobial a gents. H owever, w hen

thinking a bout biofilms, a logical a ssumption is t ha t





36 ReviewReviewReviewReviewReviewReview

a ny given cell with in th e biofilm will experience a

slightly different environment compa red with oth er

cells within th e same biofilm, an d thus be growing a t

a different rat e. Gr adients of nutrients, wa ste

products a nd signa ling factors form t o allow for t hisheterogeneity w ithin th e biofilm.

Recent a dvances in t echnology ha ve resulted in the

ability t o visualize the heterogeneity with in a biofilm.

A staining method utilizing a cridine oran ge wa s

employed to identify regions of biofilms th a t conta in

ra pidly or slowly growing cells based on their relat ive

RNA–DNA cont ent 21. The method w a s first used on

bact erial colonies and t he regions of the colonies th a t

tur ned oran ge (high rela tive RNA cont ent) were

correlated to fast growt h ra tes, whereas regions tha t

were sta ined yellow/green (low relat ive RNA cont ent)

represented slowly gr owing cells. When th is method

wa s used on seven-da y-old biofilms, orang e ma rkedth e biofilm–bulk-liqu id inter fa ce a nd yellow /green

ma rked t he center of th e biofilm (Fig. 1). This

heterogeneity w ithin biofilms ha s also been show n for

protein synthesis an d respira tory activity, whereas

DNA content rema ined relatively constant

th roughout th e biofilm27,28.

There is a lso evidence for gra dients of

physiological a ctivity in response to a nt imicrobial

trea tment. F or example, the patt ern of respira tory

a ctivity of a K. pneumoniae biofilm in response t o

monochloram ine (a n oxidat ively act ive biocide)

trea tment sh owed tha t cells closest to the

biofilm–bulk-liquid int erface lost a ctivity first 29

(Fig. 2). Similar ly, when biofilm cells were trea ted

wit h th e ant ibiotic fleroxocin, cell elongat ion wa s

observed a nd w a s most extreme in cells loca ted close

to th e exposed sid e of th e biofilm30. These st udies

reveal t ha t t he response to ant imicrobial a gents can

great ly vary , depending on the locat ion of a part icular

cell with in a biofilm communit y.

General stress response

Recently, it has been suggested tha t t he slow growt h

ra te of some cells wit hin th e biofilm is not owin g to

nutrient limitat ion per se , but to a general str ess

response initiated by growth with in a biofilm31. This

idea is a n a tt ra ctive possibility because t he stress

response results in physiological changes tha t a ct to

protect the cell from va rious environment a l stresses.

Thus, th e cells are protected from the detr imenta l

effects of hea t sh ock, cold shock, cha nges in pH a nd

ma ny chemical a gents32. The centra l regulat or of thisresponse is the alterna te σ factor, RpoS, originally

thought t o be expressed only in sta tionary pha se32.

However, recent st udies suggest t ha t RpoS is induced

by high cell density an d tha t cells growing a t t hese

high densities seem t o have undergone the general

str ess response, as judged by t he production of

treha lose (an osmoprotectant ) an d cata lase33. As cells

in a biofilm experience high cell density, it is logica l to

propose tha t t hese cells would express RpoS.

Accordingly, it ha s been shown by RT-P CR t ha t

rpoS mRNA is present in sputum from CF patients

with chronic P. aeru ginosa biofilm infections34.

Another link betw een RpoS a nd biofilms wa srecently identified: E. coli cells tha t lack rpoS a re

una ble to form n ormal biofilms wh ereas plan ktonic

cells are a pparently una ffected by the a bsence of this

σ factor35. In P. aeru ginosa , it ha s been suggested tha t

an additional σ fact or, AlgT, a cts in concert wit h RpoS

to cont rol th e stress response31,34. Cochra n et al . found

tha t th in biofilms formed by null muta nts of rpoS a nd

algT on algina te gel beads w ere susceptible to

hyd rogen peroxide but not to monochlora mine36.

However, when these mut an ts formed th ick biofilms on

glass slides, they were as resista nt t o both oxidat ive

biocides a s th e wild-ty pe cells. Thus, a lthough th ere is

some evidence to suggest t ha t rpoS a nd algT have a

Fig. 1. Physiological heterogeneity in biofilms. The spatial pattern of growth rate within a Klebsiella pneumoniae biofilm, as judged by

acridine orange staining. In this figure, areas of red–orange staining

correspond to a high relative RNA content and thus rapid growth. Cells

staining yellow/green have low relative RNA content and a slower

growth rate. There are clearly distinct regions of faster and slower

growth throughout the biofilm. The bottom of the image is the portion

of the biofilm attached to the substratum and the top of the image is the

portion of the biofilm exposed to the bulk medium. Reproduced, with

permission, from Ref.21.

Fig. 2. Susceptibility to biocide treatment. A two-species biofilm treatedwith the oxidatively active biocide monochloramine. This figure illustrates

that there is heterogeneity within the biofilm in terms of the response of

individual cells to biocide treatment. Areas of red–orange staining

correspond to respiratory activity. Green cells have no respiratory activity.

Yellow regions represent a mixture of respiring and non-respiring

bacteria. The bottom of the image is the portion of the biofilm attached to

the substratum and the top of the image is the portion of the biofilm

exposed to the bulk medium. Reproduced, with permission, from Ref. 29.





37ReviewReviewReviewReviewReviewReview

role in biofilm resist a nce to oxida tive biocides, it is clear

tha t other factors must contribute to this resista nce.

Quorum sensing

The role of quorum s ensing in biocide resist a nce is not

yet clear. P revious w ork by Da vies and colleagues

showed that a mutant in the l asR–lasI quorum-sensing

system in P. aeru ginosa wa s una ble to form a biofilmwit h norma l ar chitecture37. Moreover, these aut hors

presented data showing tha t lasI muta nt biofilms were

abnormally sensitive to treatment with SDS , although

the question of wh ether these muta nt biofilms had

a ltered antibiotic resistan ce wa s not addressed37.

However, a recent st udy by B rooun a nd co-workers

showed tha t mut an ts defective in quorum sensing were

una ffected in their resista nce to detergents an d

antibiotics26. Furt her complica ting t he interpretat ion

of these studies is a report suggesting a role for RpoS in

regulation of quorum sensing38. Addit ional

experimenta tion is required t o elucidat e the role (direct

or indirect) of quorum s ensing in biocide resista nce.

Induction of a biofilm phenotype

Thus fa r, th e mecha nisms discussed ha ve been ba sed

on general st ra tegies to slow t he effect of

a nt imicrobial a gents on cells in th e biofilm. An

emerging idea in th e field is th a t a biofilm-specific

phenotype is induced in a subpopulation of the

community tha t results in t he expression of a ctive

mecha nisms to combat the detrimenta l effects of

an timicrobial agents 15,39–41.

When cells at ta ch to a surfa ce, th ey will express a

general biofilm phenotype and w ork has begun t o try

to identify genes tha t a re activat ed or repressed inbiofilms compa red w ith plankt onic cells42.

Furth ermore, it is possible tha t a ll or just a subset of

th ese biofilm cells could express increased resista nce

to a ntimicrobial a gents. This resistan t phenotype

might be induced by nut rient limitat ion, certa in types

of stress, high cell density or a combinat ion of these

phenomena. As summa rized below, recent w ork has

focused on th e identificat ion of genes th a t could

cont ribute to this increa sed-resista nce phenotype.

Multidru g efflux pumps can extr ude chemica lly

unrelated a ntimicrobial a gents from the cell. In

E. col i , upregulat ion of the mar operon results in a

mult idrug-resista nt phenotype. The efflux pump

th ought t o be responsible for this resist a nce is AcrAB .

To address t he question of whether t his known

multid rug-resista nce system is involved in biofilm

resista nce to an timicrobial agent s, expression of mar

wa s monitored in batch, chemosta t a nd biofilm

cultures by lacZ fusion 41. Overall, the results did not

support the idea tha t the mar operon is upregula ted

in biofilms, a s t he level of mar w a s lower in biofilmscompa red wit h th e level seen in equiva lent

sta tionary-phase culture grown in bat ch.

Furth ermore, this sa me group made use of mar- a nd

acrAB -deleted stra ins to determine if the resista nce of

E. col i to ciprofloxacin w a s a ffected by loss of these

loci40. Loss of mar a nd acrAB did not ad versely affect

the E. col i biofilms, but const itut ive expression of

acrAB did provide a certa in level of protection aga inst

ciprofloxacin. Although t hese results suggest th a t

upregula tion of the mar operon specifica lly does not

a ccount for E. coli biofilm resista nce to ant imicrobial

a gents , they do not discount th e possibility of oth er

multid rug-resista nce pumps being induced inresponse to life in a biofilm.

There a re th ree known m ultidru g-efflux pumps in

P. aeru ginosa an d there are several other putat ive

pumps tha t ha ve been identified by the P. aeru ginosa

genome project. One study h a s suggested th e

importa nce of one of these pumps in the resista nce to

th e a ntibiotic ofloxa cin26. Using stra ins of

P. aeru ginosa tha t either lacked or overexpressed t he

MexAB –OprM pump, it wa s shown tha t, a t low

concentra tions of ofloxacin, biofilms lacking t he pump

were more susceptible to this drug t ha n biofilms th a t

overexpressed th e pump. However, for a d ifferent

quinolone, ciprofloxacin, th ere w a s no difference.Therefore, as wa s the case w ith t he E. col i studies, the

quest ion of wh ether ind uction of pumps is one of the

key alt era tions conferring resist a nce to biofilm cells

aw aits further experimenta tion.

Another resistance mecha nism tha t can be

induced in biofilm cells is t he a lterat ion of the

membra ne-protein composition in r esponse to

a ntimicrobial a gents. This chan ge could result in

decreas ed permeabilit y of th e cell to th ese

compounds. Muta tions in ompB (a regulat or of the

Fig. 3. Drug resistance in

biofilms. A schematic of

mechanisms that can

contribute to the

resistance of biofi lm-

grown bacteria to

antimicrobial agents. The

extracellular

polysaccharide is

represented in yellow andthe bacteria as blue ovals.

Biofilms are marked by

their heterogeneity and

this heterogeneity can

include gradients of

nutrients, waste products

and oxygen (illustrated by

colored starbursts).

Mechanisms of resistance

in the biofilm include

increased cell density and

physical exclusion of the

antibiotic. The individual

bacteria in a biofilm can

also undergo

physiological changes

that improve resistance to

biocides. Various authors

have speculated that the

following changes can

occur in biofilm-grown

bacteria: (1) induction of

the general stress

response (an rpoS -

dependent process in

Gram-negative bacteria);

(2) increasing expression

of multiple drug

resistance (MDR) pumps;

(3)activating quorum-

sensing systems; and

(4)changing profiles of

outer membrane proteins(OMP).

TRENDS in Microbiology

+ +

+ +

Antibiotic

concentrationNutrient andoxygenconcentration

Quorum

sensing

RpoS

MDR pumps

OMP

Wha t other factors ar e importa nt for

a ntim icrobial resist a nce in multi-species

biofilms?

• Wha t genes ar e induced in biofilm cells tha t

allow for increased resista nce to antimicrobial

agents?

• Wha t signa ls are involved in rpoS regulation

in biofilms?

• Are multidr ug efflux pumps importa nt for

biofilm resistance to an timicrobial a gents?

• Wha t is th e role of quorum sensing in t he

biocide resist a nce developed by biofilms?

Questions for future research





38 ReviewReviewReviewReviewReviewReview

genes encoding t he outer membra ne porin proteins

OmpF and OmpC) and in ompF increa sed the

resista nce of E. col i t o a β-lacta m an tibiotic43.

Mutant s tha t la ck OmpF ha ve been shown to be more

resista nt t o chloramphenicol and t etra cycline44.

Furt hermore, in sta rving cells, the relat ive

proport ions of the ma jor E. col i porins OmpC a nd

OmpF w ere altered, fa voring t he expression of thesma ller porin, OmpC (Ref. 45). The a bove result s

support the suggestion th at a ltering porin expression

a ffects the intrinsic resista nce of bacteria t o

ant imicrobial agents. Recently, it w as shown tha t t he

expression of ompC an d thr ee other osmotica lly

regulat ed genes wa s increased in biofilm bacteria

compared wit h pla nktonic cells46. These da ta

suggested tha t ba cteria in a biofilm a re indeed living

in a n environment of increas ed osmotic st ress. Thus,

the environmenta l conditions wit hin th e biofilm can

lead to a lterat ions w ithin th e cell envelope tha t

protect the ba cteria from the detrimenta l affects of

antimicrobial agents.

Conclusion

There is no one a nsw er to the quest ion of why an d how

ba cteria gr owing in a biofilm develop increased

resistan ce to an timicrobial ag ents. We have seen tha t

there a re man y possible mecha nisms tha t a ccount for

bacterial resistan ce to ant imicrobial compounds

(summa rized in Fig. 3). Depending on th e bacterial

complement of the biofilm, and the a ntimicrobial a gent

used to trea t t he biofilm, different m echa nisms w ill

account for r esistance t o the an timicrobial compound.

Furt hermore, the environmenta l heterogeneity t ha texists w ithin a biofilm might promote the format ion of

a h eterogeneous populat ion of cells, such tha t different

levels of resist a nce ca n be expressed thr oughout th e

commu nity . For example, the cells closest t o the

liquid–biofilm interface might be protected to a sma ll

degree by t he exopolysaccharide ma trix a nd by

enzymes th at inactivat e certa in a ntimicrobial agents.

The cells in an int ermedia te position might be growing

slowly a nd could also be protected by the outerm ost

lay er of cells. Fina lly, a nother s ub-popula tion of cells

might express a biofilm-specific resista nce phenotype

induced by th e particular environmental fa ctors

influencing these cells. It is clear tha t a dditionalstudies must be performed t o further elucidat e how

an d wh y ba cteria growing in complex surface-at ta ched

commu nities can protect t hemselves from the insults of

an timicrobial agent s.

Acknowledgements

We wish to thank Phil

Stewart for permission to

use Figs 1 and 2. This

work was supported by agrant from Microbia, Inc.

and The Pew Charitable

Trusts (to G.A.O.). G.A.O.

is a Pew Scholar in the

Biomedical Sciences.

References

1 Whittaker, C.J . et al . (1996) Mecha nisms of a dhesion

by oral bacteria. Annu . Rev. M icrobiol . 50, 513–552

2 Prosser, B .L. et al . (1987) Method of evalua ting

effects of ant ibiotics on ba cterial biofilm.

Ant im icrob. Agents Chemother. 31, 1502–1506

3 Nickel , J .C. et al . (1985) Tobra mycin r esista nce of

Pseudom onas aeru gin osa cells growing a s a

biofilm on urina ry tra ct catheter. Antimicrob.

Agents Chemoth er.27, 619–6244 Gr i st ina , A.G . et al . (1987) Adhesive coloniza tion

of biomat erials an d a ntibiotic resista nce.

Biomaterials 8, 423–426

5 Evan s, R.C. and Holmes, C.J . (1987) Effect of

van comycin h ydrochloride on Staphylococcus

epidermidis biofilm a ssociat ed with silicone

elastomer. Ant im icrob. Agents Chemother.

31, 889–894

6 Licking, E. (1999) Getting a grip on ba cterial

slime. Bu sin ess Week 13 Septem ber, pp. 98–100

7 Coster ton, J .W. et al . (1995) Microbial bi ofilms.

Annu . Rev. Mi crobiol . 49, 711–745

8 Archibald, L.K. an d Ga ynes, R.P. (1997) Hospita l

acquired infections in th e United St at es: the

importan ce of interhospital comparisons.

Nosocomi al In fect . 11, 245–2559 P otera, C. (1999) Forging a link between biofilms

and disease. Science 283, 1837–1838

10 St ewa rt, P .S. (1996) Theoretical aspects of

an tibiotic diffusion int o microbial biofilms.


11 de Beer , D. et al . (1994) Direct m easurement of

chlorine penetra tion int o biofilms during

disinfection.Appl. Envi ron. M icrobiol .

60, 4339–4344

12 Suci, P.A. et al . (1994) Investigation of

ciprofloxacin penetra tion int o Pseudomonas

aeruginosa biofilms. Anti mi crob. Agents

Chemother . 38, 2125–2133

13 Hoyle , B . et al . (1992) Pseudom onas aeru gin osa

biofilm a s a diffusion barrier to piperacillin.

Ant im icrob. Agents Ch emother . 36, 2054–2056

14 Dunne, W.M. et al . (1993) Diffu sion of rifa mpin

and vancomycin through a Staphylococcus

epidermidis biofilm. Ant im icrob. Agents

Chemother . 37, 2522–2526

15 Cochran, W.L. et al . (2000) Reduced su sceptibilit y

of thin Pseud omonas aerugi nosa biofilms t o

hydrogen peroxide an d monochlora mine. J. Appl.

M icrobiol . 88, 22–30

16 Ste war t , P .S . et al . (2000) Effect of cat a la se on

hydrogen peroxide penetra tion intoPseudomonas

aeruginosa biofilms. Appl. Envir on. Microbiol .

66, 836–838

17 Ander l , J .N. et al . (2000) Role of an tibiot ic

penetra tion limitation in Kl ebsiell a pneumoniae

biofilm resista nce to am picillin an d ciprofloxacin.


18 Tuomanen, E. et al . (1986) The ra te of killin g of

Escheri chia coli by β-lacta m ant ibiotics is strictly

proportional to th e rat e of bacterial growth .

J. Gen. Mi crobiol.132, 1297–1304

19 Tuomanen, E. et al .(1986) Antibiot ic tolera nce

am ong clinical isolat es of bacteria . Antimicrob.

Agents Chemoth er. 30, 521–527

20 Brown, M.R. et al . (1988) Resistan ce of bacteria l

biofilms to antibiotics: a growth-rat e relat ed

effect? J. An tim icrob. Chemother . 22, 777–78021 Wentland, E.J . et al . (1996) Spatia l varia tions in

growth rate within Kl ebsiell a pneumoniae

colonies an d biofilm. Bi otechnol. Pr og .

12, 316–321

22 Evans , D . J . et al . (1991) Su sceptibilit y of

Pseudom onas aeru gin osa a nd Escheri chia coli

biofilms towards ciprofloxacin: effect of specific

growth rate. J. Ant im icrob. Chemother.

27, 177–184

23 Duguid, I .G . et al . (1992) G rowth -ra te-

independent killing by ciprofloxacin of biofilm-

derived Staphylococcus epiderm idi s ; evidence for

cell-cycle depen dency. J. Ant im icrob. Chemother.

30, 791–802

24 Duguid, I .G . et al . (1992) Effect of biofilm cultu re

upon the susceptibility of Staphylococcus

epidermidis to t obramycin. J. Antim icrob.

Chemother. 30, 803–810

25 De sa i , M. et al . (1998) Increasing resista nce of

plankt onic and biofilm cultures ofBurkholderia

cepacia to ciprofloxacin a nd ceftazidim e during

exponential growth. J. Ant im icrob. Chemother.

42, 153–160

26 Brooun, A. et al . (2000) A dose-response s tu dy of

an tibiotic resista nce in Pseud omonas aerugi nosa

biofilms.Ant im icrob. Agents Chemother.

44, 640–646

27 Hua ng, C-T. et al . (1998) Spatia l pat terns of

alkaline phosphata se expression with in ba cterial

colonies and biofilms in response to phosphat e

s tarvation.Appl. Envir on. Microbiol.

64, 1526–1531

2 8 X u, K . D .et al . (2000) B iofilm resista nce to

ant imicrobial agents . M icrobiology

146, 547–549

29 Hua ng, C-T. et al . (1995) Nonuniform spa tia l

patt erns of respiratory a ctivity within biofilms

during disinfection. Appl. Envir on. Microbiol .

61, 2252–2256

30 Korber , D.R. et al . (1994) Evaluation of fleroxacin

activity aga inst established Pseudomonas

fluorescens biofilms. Appl. Envir on. Microbiol.60, 1663–1669

31 Brown, M.R. a nd B arker, J . (1999) Unexplored

reservoirs of pat hogenic bacteria : protozoa a nd

biofilms. Tr ends Mi crobiol . 7, 46–50

32 Hengge-Aronis, R. (1996) Regulation of gene

expression during entry into sta tionary pha se. In

Escheri chia coli an d Salm onell a: Cell ular and

M olecular Bi ology (Neidhart , F.C. et al ., eds),

pp. 1497–1512, ASM P ress

33 Liu, X .et al . (2000) Global a da pta tions resulting

from high population densit ies in Escherichi a coli

cultures. J. B acteri ol . 182, 4158–4164

34 Foley , I . et al . (1999) General stress response

mast er regulator rpoS is expressed in huma n

infection: a possible role in chronicity.

J. Antim icrob. Chemother. 43, 164–165



TRENDS in Microbiology Vol.9 No.1 J anuary 2001 39ReviewReviewReviewReviewReview

http://tim.trends.com 0966-842X/01/$ – see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S0966-842X(00)01910-7

Review

35 Adams, J .L. a nd McLean, R.J . (1999) Impact of

rpoS deletion on Escheri chia coli biofilms. Appl.

Envir on. Microbiol . 65, 4285–4287

36 Cochran , W.L. et al . (2000) Role of RpoS a nd AlgT

in Pseud omonas aerugi nosa biofilm resista nce to

hydrogen peroxide an d monochlora mine. J. Appl.

M icrobiol . 88, 546–553

37 Davies, D.G. et al . (1998) The in volvement of cell-

to-cell signals in t he development of a ba cterial

biofilm. Science 280, 295–298

38 Whiteley, M. et al . (2000) Regula tion of quoru m

sensing by RpoS in Pseud omonas aeru gin osa .

J. Bacteriol . 182, 4356–4360

39 Gilber t , P. et al . (1997) Biofilms susceptibility to

antimicrobials.Ad v. Dent. Res . 11, 160–167

40 Maira-Litran, T. et al . (2000) An evalua tion of the

potential of the multiple ant ibiotic resistan ce

operon (mar ) an d the multidr ug efflux pump

acrAB to moderate resista nce towards

ciprofloxacin in Escherichi a coli biofilms.

J. Antim icrob. Chemother. 45, 789–795

41 Maira-Litran, T. et al . (2000) Expression of th e

multiple an tibiotic resistan ce operon (mar )

during growth of Escheri chia coli as a biofilm.

J. Appl. M icrobiol . 88, 243–247

42 Kuchma , S.L. a nd O’Toole, G.A. (2000) Surfa ce-

induced an d biofilm-induced chan ges in gene

expression. Curr . Opin. Biotechnol . 11, 429–433

43 J a f f e, A. et al . (1982) Role of porin pr oteins

OmpF an d OmpC in the permeation of

β-lactams. Ant im icrob. Agents Chemother.

22, 942–948

44 Pu gsley, A.P . and Schna itma n, C.A. (1978)

Outer membra ne proteins ofEscheri chia coli .

VII. E vidence tha t bacteriophage-directed

protein 2 functions a s a pore. J. Bacteri ol .

133, 1181–1189

45 Liu, X. a nd F erenci, T. (1998) Regulation of porin-

mediat ed outer membra ne permea bility by

nutrient limitat ion in Escheri chia coli.

J. Bacteriol . 180, 3917–3922

46 Prigent-Combaret , C. et al . (1999) Abiotic s urfa ce

sensing a nd biofilm-dependent regula tion of gene

expression in Escheri cia coli . J . Bacteriol .

181, 5993–6002

Since their recognition a s the th ird doma in of life,

Archaea, par ticularly t hose tha t t hrive in extreme

environments , have been the focus of a g rea t dea l of

recent resea rch, including complete genome sequence

determination1. However, th e viruses of these

organ isms a re poorly underst ood. A survey of the

extra -chromosoma l element s of the extremely

th ermophilic and a cidophilic a rcha eon Sulfolobus ha s

revealed t he presence of many novel viruses, a nd both

conjugat ive a nd cryptic plasm ids2,3. The vir uses of

Sulfolobus ha ve been a ssigned t o four novelfamilies – Fuselloviridae (viruses SS V1, SS V2 an d

SSV3)3–5, Rudiviridae (SIR V1 an d SIRV2)6,

Lipothrixviridae (SIFV)7 and G ut tav i ridae

(SNDV)8 – on the ba sis of their unique morphology.

Morphology

The str ucture of typical virus pa rticles from each of

th e four fa milies is shown in F ig. 1. The flexible

filament ous virions of the lipothrixvirus SI FV

(2000 × 24 nm) cont a in a nucleosome-like core of

linear DNA wound as a superhelix around a zipper-

like arr a y of 80-kDa protein subunit s. The core is

covered by a lipid envelope (Fig. 1a). By cont ra st, t he

stiff rod-sha ped virions of rudiviruses

(830–900 × 23 nm) do not possess an envelope. In t he

Rudivirida e, the t ube-like superhelix formed by

linear DNA an d a single 15.8-kDa DNA-binding

protein is closed a t its en ds by ‘plugs’ to w hich ta il

filament s are a tt a ched (Fig. 1b). The spindle-sha ped

fusellovirus v irions (100 × 60 nm) have a core of

positively supercoiled circular D NA associated w ith aDNA-binding protein, pa ckaged in a hyd rophobic

protein envelope, wh ich has a short t a il (Fig. 1c).

Fina lly, virions of the gutt avirus SND V

(100–185 × 70–95 nm) ta ke the form of droplets,

wh ich, on th eir pointed end, ca rry a dense ‘beard ’ of

long t hin filaments (Fig. 1d).

Natural hosts and geographical distribution

Sulfolobus viruses appear to be ubiquit ous in acidic

hot-spring environment s. Fu selloviruses ha ve been

found in Sulfolobus stra ins isolated from solfata r ic

fields in J a pan , Icelan d an d North America. The

na tura l carriers of rudiviruses a nd thelipoth rixvirus SIFV are Sulfolobus isola tes from

diverse locat ions in I celan d, a lthough viruses of

similar morphology ha ve been observed in sam ples

from North America. The gutt a virus SND V has been

found in a Sulfolobus isola te from a field sample

from New Zeala nd.

Virus–host relationships

None of the Sulfolobus viruses is lytic: th e

fuselloviruses a re temperate, a nd th e others a re

present in th eir hosts in a more-or-less-sta ble ca rrier

sta te. This str a tegy could help th e viruses escape

prolonged d irect exposure t o the low pH (1–3) and

Viruses of the extremely thermophilic

archaeonSulfolobus David Prangishvili,Kenneth Stedman and Wolfram Zillig

Viruses of Sulfolobus are highly unusual in their morphology,and genome

structure and sequence.Certain characteristics of the replication strategies of

these viruses and the virus–host interactions suggest relationships with

eukaryal and bacterial viruses,and the primeval existence of common

ancestors.Moreover,studying these viruses led to the discovery of archaeal

promoters and has provided tools for the development of the molecular

genetics of these organisms.TheSulfolobus viruses contain unique

regulatory features and structures that undoubtedly hold surprises forresearchers in the future.

David Prangishvili*

Universität Regensburg,

Lehrstuhl für

Mikrobiologie –

Archaeenzentrum,

Universitätsstraße 31,

93053, Regensburg,

Germany.

*e-mail:

david.prangishvili@

biologie.uni-r.de

Kenneth Stedman

Thermal Biology Institute,

Montana State University,

Bozeman,

MT 59717, USA.

Wolfram Zillig

Max-Planck-Institut für

Biochemie,

82152 Martinsried,

Germany.

Documents

Mechanisms Biofilm Resistance