Vol. 50, No. 4MICROBIOLOGICAL REVIEWS, Dec. 1986, p. 452-4570146-0749/86/040452-06$02.00/0Copyright C) 1986, American Society for Microbiology

Encapsulated Anaerobic Bacteria in Synergistic InfectionsITZHAK BROOK

Departments of Pediatrics and Surgery, Uniformed Services University of the Health Sciences, Bethesda, Maryland 20814

INTRODUCTION ............................................................ 452

ENCAPSULATED ANAEROBIC BACTERIA IN CLINICAL INFECTIONS ....................................452

Capsule and Infectivity of Organisms from Clinical Specimens ....................................................452

Recovery of B. melaninogenicus from Tonsils...................................453

Recovery of Encapsulated Anaerobic Bacteria from Orofacial Abscesses ........................................453

CAPSULE FORMATION IN EXPERIMENTAL MIXED INFECTIONS ..........................................453

ROLE OF A CAPSULE OF BACTEROIDES SP. AND ANAEROBIC COCCI IN BACTEREMIA .........454

SIGNIFICANCE OF ANAEROBIC BACTERIA IN MIXED INFECTION WITH OTHER FLORA .......455

SYNERGY BETWEEN ANAEROBIC AND AEROBIC OR FACULTATIVE ANAEROBIC BACTERIA.455ACKNOWLEDGMENT ............................................................................... 456

LITERATURE CITED ............................................................................... 456

INTRODUCTION

Polymicrobial infections can be more virulent than thoseinvolving single organisms (4, 29). Studies of the successfultransmission of infections in animals that were given definedmixtures of aerobic and anaerobic organisms have beenreported by many investigators (28, 38, 41). The key role ofBacteroides melaninogenicus was established by Roseburyet al. (38). However, none of the earlier studies establishedthe role of encapsulated organisms in the infectious process.

Encapsulation of anaerobic bacteria has been recognizedas an important virulence factor. Several studies demon-strated the pathogenicity of encapsulated anaerobes andtheir ability to induce abscesses in experimental animalseven when inoculated alone. Onderdonk et al. (34) corre-lated the virulence of B. fragilis strains with the presence ofcapsule. Encapsulated B. fragilis strains or purified capsularpolysaccharide alone induced abscesses, whereas non-encapsulated strains seldom caused abscesses unless theywere combined with an aerobic organism. Simon et al. (40)showed that encapsulated Bacteroides strains resistednetrophil-mediated killing, compared with nonencapsulatedstrains.The susceptibility of pathogenic bacteria to phagocytosis

and killing by polymorphonuclear leukocytes and macro-phages is important in determining the outcome of thehost-pathogen interaction. Ingham et al. (19), Tofte et al.(42), and Jones and Gemmell (21) reported that both phago-cytic uptake and killing of facultative bacterial species wereimpaired at high concentrations of B. fragilis and B.melaninogenicus.The presence of a capsule on B. fragilis was shown to

provide the organism with a growth advantage in vivo overnonencapsulated isolates (37). Furthermore, encapsulatedstrains survived better in vitro than nonencapsulated vari-ants when they were grown in an aerobic environment.Another recently described mechanism of protection inde-pendent of encapsulation is the inhibition of polymorphonu-clear migration due to the production of succinic acid byBacteroides sp. (39).The importance of the capsular polysaccharide of B.

fragilis as an immunogen was demonstrated when antibodiesagainst it protected animals from early bacteremia (23).However, prevention of the formation of intra-abdominal

452

abscess by this organism was found to be T-cell mediated(35).

This paper summarizes our recent research on the role ofencapsulation on anaerobic bacteria in mixed infections. Thereview includes (i) clinical evidence that supports the role ofencapsulated anaerobic bacteria in mixed infections, (ii)experiments that investigated the process of encapsulationof anaerobes, and (iii) studies of the influence of encapsula-tion on the relationship of anaerobes with their aerobiccounterparts in infectious sites.

In addition to capsule, anaerobic bacteria possess otherimportant virulence factors. These include the production ofsuperoxide dismutase and catalase, immunoglobulinproteases, coagulation-promoting and -spreading factors(such as hyaluronidase, collagenase, and fibrinolysin), andadherence factors (4). Other factors that enhance the viru-lence of anaerobes include mucosal damage, oxidatior-reduction potential drop, and the presence of hemoglobin orblood in an infected site. However, this review is devotedonly to the role of capsule as a virulence factor.

ENCAPSULATED ANAEROBIC BACTERIA INCLINICAL INFECTIONS

Aspirates obtained from infections next to mucous mem-brane surfaces generally contain a complex bacterial popu-lation consisting of several species (4, 17). Althoughanaerobes are components of mixed infections, their roleand relative importance in the disease process often are notconsidered.

Capsule and Infectivity of Organisms from ClinicalSpecimens

In an attempt to define the important pathogens amongisolates recovered from clinical specimens, we studied thevirulence and importance of encapsulated bacterial isolatesrecovered from 13 clinical abscesses (11). This was done byinjecting each of the 35 isolates (30 anaerobes and 5 aerobes)subcutaneously (s.c.) into mice alone or in all possiblecombinations with the other isolates recovered from thesame abscess. We then observed their ability to induce orsurvive in an s.c. abscess. Sixteen of the isolates wereencapsulated; 15 of them were able to cause abscesses bythemselves and were recovered from the abscesses even

on April 12, 2019 by guest

http://mm

br.asm.org/

Dow

nloaded from

ENCAPSULATED ANAEROBIC BACTERIA 453

when inoculated alone. The other organisms, which were notencapsulated, were not able to induce abscesses wheninoculated alone. However, some were able to survive wheninjected with encapsulated strains. Therefore, the posses-sion of a capsule by an organism was associated withincreased virulence, compared with the same organism'snonencapsulated counterparts, and might have allowedsome of the other accompanying organisms to survive. Wefound this phenomenon to occur in Bacteroides sp.,anaerobic gram-positive cocci, Clostridium sp., and Esche-richia coli. Detection of a capsule in a clinical isolate maytherefore suggest a pathogenic role of the organism in theinfection.

Recovery of B. melaninogenicus from Tonsils

Two recent studies support the importance of encapsu-lated anaerobic organisms in respiratory infections (5b, 9).The presence of encapsulated and abscess-forming orga-

nisms that belong to the B. melaninogenicus group was

investigated in 25 children with acute tonsillitis and in 23children without tonsillar inflammation (control) (9). Encap-sulated organisms of the B. melaninogenicus group were

found in 23 of 25 children with acute tonsillitis versus 5 of 23controls (P < 0.001). Inoculation s.c. into mice of theBacteroides strains that had been isolated from patients withtonsillitis produced abscesses in 17 of 25 instances comparedwith 9 of 23 controls (P < 0.05). These findings suggest a

possible pathogenic role for the B. melaninogenicus group inacute tonsillar infection and also the importance of encap-sulation in the pathogenesis of the infection.

Recovery of Encapsulated Anaerobic Bacteria fromOrofacial Abscesses

Orofacial abscesses contain many species of aerobic andanaerobic bacteria, which are also found in high numbers inthe normal oral flora (4, 17). It is believed, therefore, thatmost of these organisms originate from that source andbecome pathogenic during the inflammatory process, per-

haps by selection. An evaluation was recently done of therole of encapsulation among the most frequently recoveredanaerobes, the anaerobic cocci and Bacteroides sp. A com-

parison was made between the recovery rates of encapsu-lated organisms in chronic inflammatory conditions in andaround the oral cavity and in the pharynx of normal individ-uals (Sb).The presence of encapsulated Bacteroides sp. and

anaerobic gram-positive cocci was investigated in 182 pa-tients who had head and neck or chronic orofacial infectionsand in the pharynx of 26 individuals without inflammation.Forty-nine of the patients had chronic otitis media, 45 hadcervical lymphadenitis, 37 had chronic sinusitis, 24 hadchronic mastoiditis, 10 had peritongillar abscesses, and 12had periodontal abscesses. Of the 216 isolates of B.melaninogenicus and B. fragilis groups, B. oralis, andanaerobic cocci, 170 (79%) were found to be encapsulated inpatients with chronic infections compared with only 34(35%) of 96 controls (P < 0.001).The recovery of a greater number of encapsulated

anaerobic organisms in patients with acute and chronicorofacial infections provides further support for the potentialpathogenic role of encapsulated organisms and the possibleconversion of nonencapsulated strains into encapsulatedones during the inflammatory process. Early and vigorousantimicrobial therapy, directed at both aerobic and

anaerobic bacteria present in these mixed infections, mayabort the infection before the emergence of encapsulatedstrains that contribute to the chronicity of the infection.

CAPSULE FORMATION IN EXPERIMENTAL MIXEDINFECTIONS

Studies were performed to (i) ascertain the effect of theaerobic component in mixed infection on the external cellmembrane of the anaerobic bacteria, mostly through appear-ance of many encapsulated cells in mixed infections (thissection); (ii) explore the role of encapsulated Bacteroidesspp. and anaerobic cocci in bacteremia and translocation(next section); (iii) evaluate the relative importance of eachof the aerobic and anaerobic components of mixed infectionthrough use of selective antimicrobial therapy and the effectof the presence of an encapsulated anaerobe on that rela-tionship (see "Significance of Anaerobic Bacteria in MixedInfection with Other Flora"); and (iv) investigate the syner-gistic or antagonistic capabilities of each of the componentsof mixed infection (last section). The most frequently recov-ered bacteria were used: Bacteroides, Clostridium, andFusobacterium spp. and anaerobic and facultative gram-positive cocci (AFGPC).These studies were conducted in white male albino mice,

weighing 20 to 25 g, obtained from the Naval MedicalResearch Institute mouse colony. The mice were raisedunder conventional conditions. The animal model used in allexperiments was an s.c. abscess model (20) that was suitablefor this evaluation because of its simplicity, the ability tomonitor the infection while the experiment was in progress,and the relatively easy access for obtaining cultures andmeasuring abscess size. Unless differently specified, 0.1 mlof each of the appropriate bacterial suspensions (108 cells perdose) in saline was inoculated s.c. into the right groin in allanimnals.Many of the previous studies of mixed aerobic-anaerobic

infections used an intraperitoneal abscess model. Differ-ences exist between the s.c. and intraperitoneal abscessmodels. While s.c. abscesses appear within 2 to 3 days,intraperitoneal abscesses take at least 7 to 10 days todevelop.The presence of capsules in all organisms was determined

by the Hiss stain (26) and confirmed by electron microscopyafter staining with ruthenium red (22). Ruthenium red stain-ing demonstrated a homogeneous polysaccharide capsulethat was external to the cell wall.The ability of the aerobic component in mixed infections

to enhance the appearance of encapsulated anaerobic bacte-ria in these infections was studied in an s.c. abscess model inmice. The anaerobic bacteria with which they were inocu-lated were those commonly recovered in mixed infections (4,17). B. melaninogenicus group, B. buccalis, B. oris-buccae(8), B. bivius (5a), B. fragilis group (6), and AFGPC (14) didnot induce abscess when isolates that contained only a smallnumber of encapsulated organisms (<1%) were inoculated.However, when these relatively nonencapsulated isolateswere inoculated, mixed with abscess-forming viable ornonviable bacteria ("helpless"), Bacteroides spp. andAFGPC survived in the abscess and became heavily encap-sulated (>50% of organisms had a capsule). Thereafter,these heavily encapsulated Bacteroides isolates were able toinduce abscesses when injected alone (Fig. 1). Of interest isthe observed appearance of pilli along with encapsulation inthe B. fragilis group after coinoculation with Klebsiellapneumoniae (6).

VOL. 50, 1986

on April 12, 2019 by guest

http://mm

br.asm.org/

Dow

nloaded from

MICROBIOL. REV.

Most of the "helper" strains were encapsulated; however,several of the strains were not encapsulated, but were able toinduce abscesses when inoculated alone. The helper orga-nisms used in conjunction with "oral" Bacteroides spp. (B.melaninogenicus group, B. buccalis, and B. oris-buccae) andAFGPC were Staphylococcus aureus, Streptococcuspyogenes, Haemophilus influenzae, Pseudomonas aerugi-nosa, E. coli, K. pneumoniae, and Bacteroides sp. (8, 14).For the B. fragilis group, these organisms were E. coli, K.pneumoniae, Staphylococcus aureus, Streptococcus pyo-genes, and enterococci (6). Neisseria gonorrhoeae waschosen as a helper for B. fragilis and B. melaninogenicusgroups, and also B. bivius (5a). Of interest is the observedinability of N. gonorrhoeae strains to survive in intra-abdominal abscesses and also their disappearance from s.c.abscesses within 5 days of inoculation with Bacteroides sp.(5a).The virulence of Fusobacterium sp. was also associated

with the presence of a capsule. Only encapsulated strains ofFusobacterium nucleatum, F. necrophorum, and F. vaniumwere able to induce abscesses when inoculated alone (15).However, following passage in animals of nonencapsulatedstrains, none of these organisms acquired a capsule.The presence of a thick granular cell wall (30 to 36 nm)

before anitnal passage was associated with virulence ofClostridium sp. (Sc). Such structure was observed beforeinoculation into animals only in Clostridium perfringens andC. butyricum, the only organisms capable of inducing an s.c.abscess when inoculated alone. This structure was observedin other clostridial species only after their coinoculation withencapsulated Bacteriodes sp. or K. pneumoniae.However, other undetermined factors may also contribute

to the induction of an abscess, since most isolates of C.difficile were not able to produce an abscess even thoughthey possessed a thick wall.The selection of encapsulated Bacteroides sp. and AFGPC

with the assistance of other encapsulated or nonencapsul-ated but abscess-forming aerobic or anaerobic organismsmay explain the conversion into pathogens of nonpathogehicorganisms that are part of the normal host flora or areconcomitant pathogens. Although such a phenomenon wasnot observed in Fusobacterium sp., the presence of acapsule in these organisms was a prerequisite for inductionof s.c. abscesses. Some Clostridium spp. also manifestedcell wall changes after animal passage that could be associ-ated with increased virulence. Although the exact nature andchemical composition of the capsule or external cell wall

Bacterodes fragiNs Group

no-abscess abscess

capsule positive

FIG. 1. Summary of the cycle of events leading to capsuleformation by B. fragilis. *"Helper" = viable bacteria or forma-linized bacteria or capsular material.

may be different in each of the anaerobic species studied, thechanges observed tended to follow similar patterns.The mechanism responsible for the observed phenomenon

is as yet unknown and may be due to either genetic trans-formation or a process of selection. It is possible thatcapsular material of aerobes such as K. pneumoniae or ananaerobe such as a Bacteroides sp. enables the selection ofa few encapsulated organisms from a predominantlyunencapsulated population of the tested anaerobic organism.The presence of capsular material in the inoculum wasprobably sufficient to prevent phagocytosis of the organisms(33, 40) and allowed the selection of encapsulated forms. Aninterference can occur in the interaction between polymor-phonuclear cells and the anaerobic bacteria (21). This proc-ess, which required opsonization by both complement andimmunoglobulins, can be inhibited by either Bacteroides sp.or aerobic bacteria by either competing for limited opsoninsor decreasing intracellular killing (25). We postulated that aselection process was the mechanism responsible for ourfinding because we were able to recover heavily encapsu-lated organisms after passage in vivo with a helper organismin all of the slightly encapsulated organisms.The possibility of some genetic transfer of virulence

factors through a bacteriophage from the helper organisms tothe Bacteroides sp. can further facilitate the pathogenesis ofthese organisms. The phage could also have a role inselecting out the encapsulated cells among the population ofB. fragilis which are more resistant to it than the unencapsul-ated cells (3). This phenomenon as well as other virulencefactors could account for the ability of B. fragilis (whichconstitutes only about 0.5% of the normal fecal flora) tobecome a pathogen present in 70 to 80% of intra-abdominalinfections (4, 17). A similar mechanism could account for thevirulence of Bacteroides sp. in pelvic inflammation followinginteraction with encapsulated N. gonorrhoeae. Acquisitionof such virulence-associated factors as pili and capsulesstrengthens the need to apply direct therapy against potentialpathogens such as B. fragilis.The observed change in the cell walls of clostridia may be

a result of exposure to another bacterium in an animalenvironment with possible genetic transfer that occurs onlyin these circumstances. It may also be due to the selectionfrom the introduced population of a small number of cellsthat have such a wall structure and possibly have, as a result,some advantage because of an antiphagocytic effect (33), orassociated effects such as shortened division time, whichallow its progeny to survive and outgrow the others. Anotherpossibility is that the ability to produce such a structure isinherent in these strains and is only expressed in vivo.

ROLE OF A CAPSULE OF BACTEROIDES SP. ANDANAEROBIC COCCI IN BACTEREMIA

Anaerobic bacteremia account for 5 to 15% of cases ofbacteremia (17) and are especially prevalent in polymicrobialbacteremia, associated with abscesses (7).The role of possession of capsular material in the systemic

spread of Bacteroides sp. and AFGPC was investigated inmice following s.c. inoculation of encapsulated strains aloneor in combination with aerobic or anaerobic facultativebacteria (I. Brook, J. Med. Microbiol., in press). Encapsu-lated anaerobes were isolated more frequently from infectedanimal blood, spleen, liver, and kidney than were non-encapsulated organisms.

After inoculation with a single encapsulated anaerobicstrain, encapsulated organisms were recovered in 163 of 420

454 BROOK

on April 12, 2019 by guest

http://mm

br.asm.org/

Dow

nloaded from

ENCAPSULATED ANAEROBIC BACTERIA 455

(39%) animals, whereas nonencapsulated anaerobes were

recovered in only 14 of 420 (3%) animals. Following inocu-lation of B. fragilis mixed with aerobic or facultative flora,encapsulated B. fragilis was isolated more often and forlonger periods of time than was the nonencapsulated strain.Furthermore, encapsulated B. fragilis was recovered moreoften after inoculation with other flora than it was wheninoculated alone.

Therefore, encapsulated strains were more virulent thannonencapsulated strains. These data highlight the impor-tance of encapsulated Bacteroides sp. and AFGPC in in-creasing the mortality associated with bacteremia and thespread to different organs. A similar pathogenic quality wasobserved in other bacterial species, such as Streptococcuspneumoniae (43) and H. influenzae (16), in which the encap-sulated strains showed greater ability for systemic spread.Onderdonk et al. (36), who studied the model of intra-

abdominal sepsis and abscesses in rats, were able to detectbacteremia due to B. fragilis only during the first few hoursafter inoculation of B. fragilis and E. coli. However, theirmodel represented an intra-abdominal infection, whereasours was an s.c. infection. Bennion et al. (2), who inducedcolonic ischemia in dogs, were able to induce prolonged andpresistent bacteremia due to B. fragilis and other anaerobicbacteria.These data also demonstrate the synergy between Bacte-

roides sp. and AFGPC and the other flora present with themin mixed infections. Further studies are warranted to explainthe pathophysiology of increased virulence of encapsulatedBacteroides sp. and AFGPC expressed by persistent bacte-remia and spread to different organs.

SIGNIFICANCE OF ANAEROBIC BACTERIA IN MIXEDINFECTION WITH OTHER FLORA

Although anaerobic bacteria often are recovered mixedwith other aerobic and facultative flora, their exact role inthese infections and their relative contribution to the patho-genic process are unknown. The relative importance of theorganisms present in an abscess caused by two bacteria (anaerobe and an anaerobe) and the effect of encapsulation on

that relationship were determined by comparing the abscesssizes in (i) mice treated with antibiotics directed against oneor both organisms and (ii) nontreated animals (12, 13, 15;Brook and Walker, J. Infect., in press).As judged by selective antimicrobial therapy, the posses-

sion of a capsule in most mixed infections involving Bacte-roides sp. generally made these organisms more importantthan their aerobic counterparts. In almost all instances, theaerobic counterparts in the infection were more importantthan nonencapsulated Bacteroides species (13). Encapsu-lated members of the B. melaninogenicus group were almostalways more important in mixed infections than their aerobiccounterparts (Streptococcus pyogenes, Streptococcus pneu-moniae, K. pneumoniae, H. influenzae, and Staphylococcusaureus). Encapsulated B. fragilis group organisms werefound to be more important than or as important as E. coliand enterococci and less important than Staphylococcusaureus, Streptococcus pyogenes, and K. pneumoniae.

In contrast to Bacteroides sp., encapsulated AFGPC were

found more often to be less important than their aerobiccounterparts (12). Clostridium sp. and Fusobacterium sp.were found to be less or equally important to entericgram-negative rods (15; Brook and Walker, in press). Al-though Fusobacterium sp., AFGPC, and Clostridium sp.were generally equal to or less important than their aerobic

counterparts, variations in the relationships existed. How-ever, as determined by the abscess size, most of theanaerobic organisms enhanced mixed infection.

SYNERGY BETWEEN ANAEROBIC AND AEROBIC ORFACULTATIVE ANAEROBIC BACTERIA

Several studies documented the synergistic effect of mix-tures of aerobic and anaerobic bacteria in experimentalinfections. Altemeier (1) demonstrated the pathogenicity ofbacterial isolates recovered from peritoneal cultures afterappendiceal rupture. Pure cultures of individual isolateswere relatively innocuous when implanted s.c. in animals,but combinations of facultative and anaerobic strainsshowed increased virulence. Similar observations were re-ported by Meleney et al. (29) and Hite et al. (18).We have evaluated (10) the synergistic potentials between

aerobic and anaerobic bacteria commonly recovered inmixed infections. Each bacterium was inoculated s.c. intomice alone or mixed with another organism, and synergisticeffects were determined by observing abscess formation andanimal mortality. The tested bacteria included encapsulatedBacteroides sp., Fusobacterium sp., Clostridium sp., andanaerobic cocci. Facultative and anaerobic bacteria includedStaphylococcus aureus, P. aeruginosa, E. coli, K. pneumo-niae, and Proteus mirabilis. In many combinations, theanaerobes significantly enhanced the virulence of each of thefive aerobes. The most virulent combinations were betweenP. aeruginosa or Staphylococcus aureus and anaerobiccocci or Bacteroides sp.The bacterial combinations used in this study often occur

clinically in infectious sites (4). The anaerobic species usedin this study, B. fragilis, B. asaccharolyticus, AFGPC, C.butyricum, C. perfringens, F. varium, and F. nucleatum, arethe most frequently recovered anaerobes from clinical spec-imens (4, 17). Our data support the finding of other investi-gators who observed synergy between Bacteroides sp. andenteric bacteria (24). We have been able to demonstratesynergy between the Bacteroides sp. and all of the faculta-tive or aerobic organisms that we tested and between mostAFGPC and P. aeruginosa or Staphylococcus aureus. Al-though there was variability of synergy between combina-tions of AFGPC strains, we have also demonstrated synergyin the ability to induce s.c. abscesses between AFGPC andBacteroides sp. or Fusobacterium sp. Such synergy was alsodemonstrated between Fusobacterium sp. and Bacteroidessp. or Clostridium sp. Little to no synergy was observedbetween AFGPC and Clostridium sp.Enhancement of the growth of each bacterial component

in mixed infection was evaluated by studying the relativegrowth of each bacterial component. This was done bycomparing the (i) growth of each organism in an abscesswhen present with other organism to (ii) growth of thatbacteria when inoculated alone (5, 15; Brook and Walker, inpress).

Streptococcus pyogenes, E. coli, Staphylococcus aureus,K. pneumoniae, and P. aeruginosa were enhanced by B.fragilis, B. melaninogenicus (5), peptostreptococci (14),Fusobacterium sp. (15), and Clostridium sp. (Brook andWalker, in press) except C. difficile. Although mutual en-hancement of growth of both aerobic and anaerobic bacteriawas noticed, the number of aerobic and facultative bacteriawas increased many times more than their anaerobic coun-terparts. Encapsulated Bacteroides sp. was able to enhancethe growth of aerobic and facultative anaerobic bacteriamore than nonencapsulated organisms. Exceptions to the

VOL. 50, 1986

on April 12, 2019 by guest

http://mm

br.asm.org/

Dow

nloaded from

456 BROOK

mutual enhancement were noticed in combinations of orga-nisms that generally are not recovered together in mixedinfections, such as enterococci and B. melaninogenicus. Theabove observations suggest that the aerobic and facultativebacteria benefit even more than do the anaerobes from theirsymbiosis.

Several hypotheses have been proposed to explain micro-bial synergy. When this phenomenon occurs in mixtures ofaerobic and anaerobic flora, it may be due to protection fromphagocytosis and intracellular killing (19), production ofessential growth factors (27), and lowering of oxidation-reduction potentials in host tissues (30). Obligate anaerobescan interfere with the phagocytosis and killing of aerobicbacteria (19). The ability of human polymorphonuclear leu-kocytes to phagocytose and kill Proteus mirabilis was im-paired in vitro when the human serum used to opsonize thetarget bacteria was pretreated with live or dead organisms ofvarious Bacteriodes sp. (21). B. gingivalis cells or superna-tant culture fluid was shown to possess the greatest inhibi-tory effect among the Bacteroides sp. (32). Supernatants ofcultures of B. fragilis group, B. melaninogenicus group, andB. gingivalis were capable of inhibiting the chemotaxis ofleukocytes to the chemotactic factors of Proteus mirabilis(31). Another possible mechanism that explains thesynergetic effect of aerobic-anaerobic combinations is thelowering of local oxygen concentrations and the oxidation-reduction potential by the aerobic bacteria. The resultantphysical conditions are appropriate for replication and inva-sion by the anaerobic component of the infection. Suchenvironmental factors are known to be critical for anaerobicgrowth in vitro and may apply with equal relevance to invivo experimental animal studies. Mergenhagen et al. notedthat the infecting dose of anaerobic cocci was significantlylowered when the inoculum was supplemented with chemi-cal reducing agents (30). A similar effect may be produced byfacultative bacteria, which may provide the proper condi-tions for establishing an anaerobic infection at a previouslywell-oxygenated site.The demonstration of the synergistic potentials of

anaerobic bacteria such as B. fragilis, B. assacharolyticus,Fusobacterium sp., Clostridium sp., and anerobic cocci,when mixed with various aerobic and anaerobic bacteria,further indicate their pathogenic role. Further studies areneeded to investigate the exact mechanisms by which suchsynergy occurs and the mode by which capsular materialenhances it.

ACKNOWLEDGMENT

I am grateful to Hayati Atan for secretarial assistance and to T.Elliott for reviewing the manuscript.

LITERATURE CITED

1. Altemeier, W. A. 1942. The pathogenicity of the bacteria ofappendicitis. Surgery 11:374-378.

2. Bennion, R. S., S. E. Wilson, A. I. Serota, and D. A. Williams.1984. The role of gastrointestinal microflora in the pathogenesisof complication of mesenteric ischemia. Rev. Infect. Dis.6:S132-S128.

3. Booth, S. J., R. L. VanTasseli, J. L. Johnson, and T. D. Wilkins.1979. Bacteriophages of Bacteroides. Rev. Infect. Dis. 1:325-334.

4. Brook, I. 1983. Anaerobic infections in childhood. G. K. Hall,Boston.

5. Brook, I. 1985. Enhancement of growth of aerobic and faculta-tive bacteria in mixed infections with Bacteroides sp. Infect.Immun. 50:929-931.

5a.Brook, I. 1986. The effect of encapsulation on the pathogenicityof mixed infection of Neisseria gonorrhoea and Bacteroidesspp. Am. J. Obstet. Gynecol. 155:424-428.

5b.Brook, I. 1986. Recovery of encapsulated anaerobit bacteriafrom orofacial abscesses. J. Med. Microbiol. 22:171-174.

5c.Brook, I., R. Cole, and R. I. Walker. 1986. Changes in the cellwall of Clostridium species following passage in animals.Antonie Leeuwenhoek J. Microbiol. Serol. 55:273-280.

6. Brook, I., J. C. Coolbaugh, and R. I. Walker. 1984. Pathogenic-ity of piliated and encapsulated Bacteroides fragilis. Eur. J.Clin. Microbiol. 3:207-209.

7. Brook, I., G. Controni, W. Rodriguez, and W. J. Martin. 1980.Anaerobic bacteremia in children. Am. J. Dis. Child.134:1052-1056.

8. Brook, I., J. D. Gillmore, J. C. Coolbaugh, and R. I. Walker.1983. Pathogenicity of encapsulated Bacteroides melanino-genicus group, Bacteroides oralis, and Bacteroides ruminicolain abscesses in mice. J. Infect. 7:281-286.

9. Brook, I., and A. E. Gober. 1983. Bacteroides melaninogenicus,its recovery from tonsils of children with acute tonsillitis. Arch.Otolaryngol. 109:818-820.

10. Brook, I., V. Hunter, and R. I. Walker. 1984. Synergistic effectsof anaerobic cocci, Bacteroides, Clostridia, Fusobacteria, andaerobic bacteria on mouse mortality and induction of subcuta-neous abscess. J. Infect. Dis. 149:924-928.

11. Brook, I., and R. I. Walker. 1983. Infectivity of organismsrecovered from polymicrobial abscesses. Infect. Immun.42:986-989.

12. Brook, I., and R. I. Walker. 1984. Pathogenicity of anaerobicgram-positive cocci. Infect. Immun. 45:320-324.

13. Brook, I., and R. I. Walker. 1984. Significance of encapsulatedBacteroides melaninogenicus and Bacteroides fragilis groups inmixed infections. Infect. Immun. 44:12-15.

14. Brook, I., and R. I. Walker. 1985. The role of encapsulation inthe pathogenicity of anaerobic Gram-positive cocci. Can. J.Microbiol. 31:176-180.

15. Brook, I., and R. I. Walker. 1986. Pathogenicity of Fusobacter-ium species. J. Med. Microbiol. 21:93-100.

16. Chandler, C. A., L. D. Fothergill, and J. H. Dingle. 1937. Studiesof Haemophilus influenzae. II. A comparative study of viru-lence of smooth, rought, and respiratory strains of Haemophilusinfluenzae as determined by infection of mice with mucinsuspension of the organism. J. Exp. Med. 66:789-799.

17. Finegold, S. M. 1977. Anaerobic bacteria in human disease.Academic Press, Inc., New York.

18. Hite, K. E., M. Locke, and H. C. Hesseltine. 1949. Synergism inexperimental infections with nonsporulating anaerobic bacteria.J. Infect. Dis. 84:1-9.

19. Ingham, H. R., P. R. Sisson, D. Tharagonnet, J. B. Selkon, andA. A. Codd. 1977. Inhibition of phagocytosis in vitro by obligateanaerobes. Lancet ii:1251-1254.

20. Joiner, K., B. Lowe, J. Dzink, and J. G. Bartlett. 1982. Com-parative efficacy of 10 antimicrobial agents in experimentalinfections with Bacteroides fragilis. J. Infect. Dis. 145:561-568.

21. Jones, G. R., and C. G. Gemmell. 1982. Impairment by Bacte-roides species of opsonization and phagocytosis of entero-bacteria. J. Med. Microbiol. 15:351-361.

22. Kasper, D. L. 1976. Chemical and biological characterization ofthe lipopolysaccharide of Bacteroides fragilis subspecies fra-gilis. J. Infect. Dis. 134:59-66.

23. Kasper, D. L., and A. B. Onderdonk. 1982. Infection withBacteroides fragilis pathogenesis and immunoprophylaxis in ananimal model. Scand. J. Infect. Dis. Suppl. 31:28-33.

24. Kelly, M. J. 1977. Mixtures of aerobic and anaerobic humanpathogens: a 4-plate counting technique. Br. J. Exp. Pathol.58:4478-4483.

25. Klemper, M. S. 1984. Interactions of polymorphonuclear leuko-cytes with anaerobic bacteria. Rev. Infect. Dis. 6:540-544.

26. Lennette, E. H., A. Balows, W. J. Hausler, Jr., and J. P. Truant(ed.). 1980. Manual of clinical microbiology, 3rd ed. AmericanSociety for Microbiology, Washington, D.C.

27. Lev, M., K. C. Krudell, and A. F. Milford. 1971. Succinate as agrowth factor of Bacteroides melaninogenicus. J. Bacteriol.

MICROBIOL. REV.

on April 12, 2019 by guest

http://mm

br.asm.org/

Dow

nloaded from

ENCAPSULATED ANAEROBIC BACTERIA 457

108:175-178.28. MacDonald, J. B., S. S. Socransky, and R. J. Gibbons. 1963.

Aspects of the pathogenesis of mixed anaerobic infections ofmucous membranes. J. Dent. Res. 42:529-544.

29. Meleney, F., S. Olpp, H. D. Harvey, and H. Jern. 1932.Peritonitis. II. Synergism of bacteria commonly found in peri-toneal exudates. Arch. Surg. 25:709-721.

30. Mergenhagen, S. E., J. C. Thonard, and H. W. Scherp. 1958.Studies on synergistic infections. I. Experimental infectionswith anaerobic streptococci. J. Infect. Dis. 103:33-44.

31. Namavar, F., A. M. J. J. Verweji, M. Bal, T. J. Martijn van

Steenbergen, J. deGraaf, and D. M. MacLaren. 1983. Effect ofanaerobic bacteria on killing of Proteus mirabilis by humanpolymorphonuclear leukocytes. Infect. Immun. 40:930-935.

32. Namavar, F., A. M. J. J. Verweji-Van Vught, W. A. C. Velj M.Bal, and D. M. MacLaren. 1984. Polymorphonuclear leukocytechemotaxis by mixed anaerobic and aerobic bacteria. J. Med.Microbiol. 18:167-172.

33. Okuda, K.* and I. Takozoe. 1983. Antiphagocytic effects of thecapsular structure of a pathogenic strain of Bacteroldesmelaninogenicus. Bull. Tokyo Dent. Coll. 14:99-104.

34. Onderdonk, A. B., f. L. Kasper, D. L. Cisieros, and J. G.Bartlett. 1977. The capsular polysaccharide of Bacteroidesfragilis as a virulence factor: comparison of the pathogenicpotential of encapsulated strains. J. Infect. Dis. 136:82-89.

35. Onderdojk, A. B., R. R. Markham, D. F. Zalenznik, R. L.Cineros, and D. L. Kasper. 1982. Evidence for T cell-dependentimmunity to Bacteroides fragilis in an intra-abdominal abscessinodel. J. Clin. Invest. 69:9-16.

36. Onderdonk, A. B., W. M. Weinstein, N. M. Sullivan, J. G.

Bartlett, and S. L. Gorbach. 1974. Experimental intraabdominalabscesses in rats: quantitative bacteriology of infected animals.Infect. Immun. 10:1256-1259.

37. Patrick, S., J. H. Reid, and M. H. Larkin. 1984. The growth andsurvival of capsulate and non-capsulate Bacteroides fragilis invivo and in vitro. J. Med. Microbiol. 17:237-246.

38. Rosebury, T., A. R. Clank, S. G. Engel, and F. Tergis. 1950.Studies of fusospirochetal infection. I. Pathogenicity for guineapigs of individual and combined cultures of spirochetes andother anaerobic bacteria derived from human mouth. J. Infect.Dis. 87:217-225.

39. Rutstein, D. D., T. L. Pruett, V. D. Fiegel, R. D. Nelson, andR. L. Simmons. 1985. Succinic acid in a metabolitic by-productof Bacteroides sp. inhibits polymorphonuclear leukocyte func-tion. Infect. Immun. 48:402-408.

40. Simon, G. L., M. S. Kelmpner, D. L. Kasper, and S. L.Gorbach. 1982. Alteration in opsonophagocytic killing by neu-trophils of Bacteroides fragilis associated with animal andlaboratory passage: effect of capsular polysaccharide. J. Infect.Dis. 145:72-79.

41. Socransky, S. S., and R. J. Gibbons. 1965. Required role ofBacteroides melaninogenicus in mixed anaerobic infections. J.Infect. Dis. 115:247-253.

42. Tofte, R, W., P. K. Peterson, D. Schmeling, J. Bracke, Y. Kim,and P. G. Quie. 1980. Opsonization of four Bacteroides species:role of the classical complement pathway and immunoglobulin.Infect. Immun. 27:784-792.

43. Wood, W. B., and M. R. Smith. 1949. Inhibition of surfacephagocytosis by the capsular slime layer of Pneumococcus typeIII. J. Exp. Med. 90:86-97.

VOL. 50, 1986

on April 12, 2019 by guest

http://mm

br.asm.org/

Dow

nloaded from