Senador @JoseRafaelS depositó proyecto de ley que declara el día 25 de septiembre de cada año como "Día de la Reafirmación Democrática"

7/27/2019 Senador @JoseRafaelS deposit proyecto de ley que declara el da 25 de septiembre de cada ao como "Da de l

1/10

The Bacterial Etiology of DestructivePeriodontal Disease: Current Concepts*3 ~ t ~ ~ l n dSocr-ansky and Ann e D. Haffajee

T H E I N T E R P R E T A T I O N O F DIAGNOS TIC TESTS for the detection of subgingival bacterialspecies is dependent on knowledge of the microbial etiology of destructive periodontaldiseases. Specific etiologic agents of these diseases have been sought for ov er 1 00 years;however, the complexity of the microbiota, an incomplete understanding of the biologyof periodontal diseases, and technical problems have handicapped this search. Nonethe-less, a number of possible pathogens have been suggested on the basis of their associationwith disease, animal pathogenicity, and virulence factors. The immunological responseof the host to a species and the relation of successful therapy to the elimination of thespecies have also been used to support or refute suspected periodontal pathogens. Currentdata suggest that pathogens are necessary but not sufficient for disease activity to occur.Factors which influence activity include susceptibility of the individual host and thepresence of interacting bacterial species which facilitate or impede disease progression.Recent studies have attempted to distinguish virulent and avirulent clonal types of sus-pected pathogenic species and seek transmission of genetic elements needed for patho-genic species to cause disease. Finally, the local environment of the periodontal pocketmay be important in the regulation of expression of virulence factors by pathogenicspecies. Thus, in order that disease result from a pathogen, 1) it must be a virulent clonaltype; 2) it must possess the chromosomal and extra-chromosomal genetic factors toinitiate disease; 3) the host must be susceptible to this pathogen; 4) the pathogen mustbe in numbers sufficient to exceed the threshold for that host; 5 ) it must be located atthe right place; 6) other bacterial species must foster, or at least not inhibit, the process;and 7) the local environment must be one which is conducive to the expression of thespecies' virulence properties. J Periodontol 1992; 63.322-331.Key Words: Periodontal diseasesletiology; periodontal diseasesimicrobiology; periodon-tal diseaseslpathogenesis.

Before beginning a discussion of current concepts of thebacterial etiology of destructive periodontal diseases, it isworth asking whether such knowledge would improve theprevention and treatment of these diseases. Specific an-swers to this question will be provided, at least in part, byDr. Listgarten in the next presentation. However, perspec-tive may be gained by analogy to the success of medicalmicrobiology in determining not only the etiology of manyinfectious diseases, but in designing diagnostic tests fortheir causative agents and, eventually, better methods fortheir control. We note their success but may fail to appre-ciate the time needed for its fruition. Lung infections serveas a useful example. In 1882, Koch' published a paperdescribing the etiologic agent of tuberculosis. Within oneyear, 26 papers were published telling him why he waswrong.2 Criticisms started with concerns about fundamentalbiology, suggesting that there were no such things as bac-Wepartment of Periodontology, Forsyth Dental Center, Boston MA .

teria or that there was only one species of bacteria wimany different forms. Critics went on to criticize his staing and microscopy techniques, animal model systems, dividual specific findings, and his interpretations of thofindings. In 1883, Koch published a specific rebuttal each of his critic^. This rebuttal demo nstrated the raprate of publication in that era and Koch's desire and abilito unmercifully flay his critics. (The tone of his responwould be unthinkable in the current era of peer reviewKoch was clearly correct. However, it is worth noting thspecific treatments based o n this knowledge were not avaable for close to 40 years. Over the next 10 0 years, addtional etiologic agents of lung infections were discoveresome as recently as 1977, when the causative agent of Lgionnai re ' s di sease was dis~overed. '~~as knowledge specific etiology of lung infection been useful? Withosuch knowledge, physicians might have difficulty in indvidual cases, in distinguishing between diseases such tuberculosis, pneumococcal pneumonia, coccidiomycos


2/10

Volume 63Number 4 SOCRANSKY,HAFFAJEE100 YEARS OF PERIODONTAL MICROBIOLOGY

SPECIFIC NON-SPECIFIC- ;* SPECIFIC---- -- J\-____Y--Fusrfoimis Ius~formisv Streptococc~v

SprochetesTAmoebaT

M~xednfectionF~~osplrochetal. ? yrrlgivalisM~xednteclion ? !nt~i!nedjaBlack ptgmented TBacferotdcs C rectusB forsyfhusSpmchete VANUGv

A V ~ S C O S ~ ~ ST- 7- - -- - --

1890 1900 1910 1920 1930 1940 1930 1960 1970 1980 1990

Figure I . D~ ag ram f some of the highlights of the past 100 years of th erearch for the etlolo g~c gents of destructive penodon tal diseases.

and Legionnaire's disease. More importantly, the optimumspecific therapy for each disease might not be employed.The point of this introduction is to illustrate that it takestime to determine the etiologic agents of diseases affectinga given organ system. It takes additional time to developappropriate diagnostic tests for those agents. Finally, it takeseven more time, sometimes decades in the example citedabove, before specific therapies based on knowledge ofspecific etiology can be developed. In the field of perio-dontology, we are fortunate in some ways and unfortunatein others. We are fortunate in that we can draw on theexperiences of physicians who designed and implementeddiagnostic tests useful for other organ systems; and we canemploy newly developed technologies for the rapid iden-tification and enumeration of microbial species. We areunfortunate in that we have a rather complex problem inetiology to unravel. Furthermore, we have a tradition ofnon-use of laboratory diagnostic tests to overcome. Clini-cians have treated periodontal diseases more or less suc-cessfully for over 100 years without benefit of diagn ostictests for the detection and identification of subgingival bac-terial species. They are naturally somewhat skeptical of theneed for such tests. However, when such tests can be dem-onstrated to increase the precision and reliability of peri-odontal disease control, their ultimate acceptance is highlyprobable.100 Years of Periodontal MicrobiologyFigure 1 presents a diagram of some of the highlights ofthe past 100 years of the search for the etiologic agents ofdestructive periodontal diseases. The papers describing eachof the pathogens or pathogen complexes have been re-viewed elsewh ere.' Obv iously not all of the significantfindings have been included, but enough to provide an ap-preciation that the search has been long, that numerousagents have been suggested, and that uncertainty has beena hallmark of this search. The time line is divided into threephases: an initial phase where specificity in the etiology ofdisease was the prominent concept; a period of disillusion-ment in which non-specificity in bacterial etiology (or lackof a role for bacteria) was dominant; and a return to the

concept of specificity in the etiology of periodontal disein the 1970s. Around the turn of the century, investigatwere proposing poss ible e t iologic roles for fus iformamoebae, spirochetes, and streptococci. Even though spcific etiology fell out of fashion at approximately 193mixed infections continued to be studied and specific mcrobial complexes suggested. In the mid-1960s, it was sgested that a specific spirochete m ight be the cau se of acnecrotizing ulcerative gingivitis on the basis of observilarge numbers of this organism in sections of lesions of td i ~ e a s e . ~ , ~n the same decade, a possible role was sugested for Actinomyces viscosus in the etiology of peodontal disease largely on the basis of the pathogenicitythis species in hamster and rat mod el system .^ Specificreturned as a fundamental concept of the etiology of peodontal diseases after studies of localized juvenile peodon titis implicated Actinobacillus actinomycetemcomitaas a possible pathogen in this disease."-" Soo n thereaftPolphyromotzas gingivalis was suggested to be importain adult periodontitis.12-l6 In more recent years, additiospecies have been suggested as discussed below. Th e mapoint of Figure 1 is that, over the years, numerous "patogens" have been proposed, but their role in disease hnot been made entirely clear. This suggests that there habeen problems associated with the search for etiologic ageas outlined below.The Complexity of the ProblemThe difficulties encountered in determining the etiologagents of destructive periodontal diseases have been dcussed at length.'7,'8 In brief, investigators have been hanicapped by the dual problems of technical difficulty ainadequate understanding of the biology of destructive peodontal d iseases. Techn ical difficulties start with the takiof plaque samples. Obtaining precise, "uncontaminated"samples from the correct location in a pocket at a stage active disease has been extremely challenging . Even if susamples were obtained, the investigator would be facwith discriminating the pathogen(s) from among the 300 400 candidate species encountered in subgingival plaquMaking matters worse, many plaque isolates are difficuto grow and maintain and even when this is possible, theare often problems in taxonomy and identification.

While the technical problems are forbidding, lack ofclear understanding of disease pathogenesis has been evmore troublesome. Most difficulties in this area result fromisclassification of disease type and status. Clearly, theare multiple destructive periodontal diseases. Combinisubjects representing two or more disease types into a singgroup diminishes the likelihood of discriminating the pato g e n ( ~ )rom other species. In addition, an individual mighave disease due to different species at different sites. It also conceivable that a site could exhibit consecutive epsodes of disease each due to different species. Failure recognize such occurrences may obscure detection of patogenic species.


3/10

32 4 CURRENT CONCEPTS OF BACTERIAL ETIOLOGYMultiple d iseases with multiple etiologies are not the onlyproblem of misclassification faced by the investigator. De-structive periodontal diseases leave an historical record oftheir progress in terms of attachment and bone loss, pocketformation, and recession. Even the causative organisms mayremain in periodontal pockets, albeit in lower numbers,long past the time in which the disease wa s active. E pisodicdisease will be discussed at greater length by Drs. Jeffcoatand Goodson; however, taking samples from sites (or sub-jects) which are in remission and interpreting them as ifthey were obtained from sites undergoing active destructioncould lead to erroneous conclusions.Another difficulty associated with attempts to defineperiodontal pathogens is the possibility that some of thes ~s pe ct ed athogens may result from, rather than cause, thediseases. Thus, disease caused by a pathogen at a site changesthe environment at that site, quite possibly favoring themultiplication of species which are favored by the new en-vironment. These "opportunistic" species may be re-covered from the site and assumed to be the cause of theobserved pathology. This problem is particularly relevantsince most studies in this field may be considered retro-spective case control studies, wher e sam ples are taken fromcases and controls after disease has taken place. Prospectivestudies should diminish this concern to some extent.It is likely that a subset of the periodontal lesions weobserve result from two or more species acting in concertto cause disease. These mixed infections create another de-gree of complexity for the research worker. The number ofpossib!e co mbina tions of two, three, o r more species thatwould have to be evaluated for their role in disease makes

detection of such mixtures in clinical material more difficultthan defining the role of single species.It is clear that suspected periodontal pathogens can bedetected in low numbers in healthy subjects or in healthysites in diseased subject s. This "carrier state" could rep-resent a long lag phase prior to detection of disease or itcould represent a true carrier state in which that site mightnever show disease. The carrier state has been recognizedfor medically-important bacteria for decades. The presenceof the carrier state complicates analysis, since it providessamples from sites of non-disease which harbor the samepathogens as sites of disease. Detection of pathogens inhealthy sites could also be explained by differences in clonaltype. There are data to indicate that strains of the samesubgingival species exhibit different pathogenicity in ani-mal model Failure to distinguish virulent fromavirulent clonal types would impair our ability to distin-guish pathogenic species. This area will be amplified ingreater detail below.The Approach to Determining Etiologic AgentsThe complexity of the problem outlined above requires theformulation of strategies for distinguishing periodontalpathogens. The classic approach to this problem has beento use the criteria known as Koch's postulates. These are:

J PeriodonApril 1992 (Suppleme1) the agent must be isolated from every case of diseas2) the agent must not be recovered from cases of othforms of disease or non-pathogenically; and 3) after isoltion and repeated growth in pure culture the pathogen muinduce disease in experimental animals. It is worth notithat while these criteria have been repeatedly employed dental research, the third criterion was abandoned by Koin 188426 when he could not induce cholera using Vibrcholera in animal model systems. The recognition of tcarrier-state led to the relaxation of the second criterion 1890.27In recent years, periodontal research workers have etended Koch's postulates somewhat to include criteria association, elimination, animal pathogenicity, host rsponses, and formation of virulence factors. The criteriof association requires that the suspected pathogenic specibe more frequently detected and at higher levels in casthan in controls. For example, the species should be highin actively progressing sites than in healthy sites, non-prgressing sites, or sites showing improvement. Longitudinassessment might also show an increase in the species prito or concomitant with measured disease progression. Tfundamental basis of elimination studies is that treatmeadministered to subjects with a given form of disease shouinfluence both the clinical status of the disease and m embeof the associated microbiota. It is reasonable to expect thsuccessful therapy will diminish the level of a pathogen anhalt disease progression. Failure to eliminate or diminithe level of the pathogen ultimately might lead to furthprogression at that site or in that subject.Testing of pathogenicity in animal model systems cotinues to be used to further support (or refute) possibpathogens. In spite of concerns with animal model syt e m ~ , ' ~hey can provide additional evidence of roles fcertain species in di ~e as e~ $- ~Ond will be particularly usefin defining virulence factors for di ~ e a s e .~ ' he host rsponse has also been employed in attempting to discrimnate periodontal pathogens. Most notable has been thassociation of antibody response to A. actinomycetemcomitans as a marker demonstrating the relationship of thspecies with localized juvenile period on ti ti^.^^.^^ It is fethat a periodontal pathogen which causes destructive perodontal disease often will elicit an elevated immunologicresponse, either locally or systemically. In certain circumstances, it is conceivable that the pathogen may diminisaspects of the host response. Finally, the ability of certaispecies to produce virulence factors has been used to suport possible roles of such species in periodontal diseaseThe production of unique biochemical determinants bpathogens may be important in disease and an indicator othe potent ia l of the species to cont r ibute to d iseasprogression.Currently Suspected PathogensTables 1, 2 and 3 are excerpted from a previously publisheworkI8 and list some of the currently suspected pathogen


4/10

Volume 63Number 4 SOCRANSKY. IIAFFAIEE 325Table 1. Data That Suggest Actinobacillus actinomyceterncomitans asa Possible Etiologic Agent of Destructive Periodontal Diseases(adapted from 18)Criterion FindingsAssociation Elevated in lesions of juvenile periodontitisUnusual in health or gingivitis subjectsElevated in some periodontitis lesions

Elevated in active juvenile periodontitis lesionsDetected in prospective studiesDetected in apical area of pocket or in tissues ofUP lesions

Elimination Elimination resulted in successfu l therapyRecurrent lesions harbored speciesHost response Elevated serum antibody in juvenile periodontitisElevated local antibody in juvenile periodontitisVirulence factors Leukotoxin, collagenase, endotoxin, epitheliotoxin,fibroblast inhibitory factor, bone resorptioninducing factorAnimal studies Induce disease in gnotobiotic rat

Table 2. Data That Suggest Porphyromonas gingivalis as a PossibleEtiologic Agent of Destructive Periodontal Diseases (adapted from18)Criterion FindingsAssociation Elevated in lesio ns of periodontitisUnusual in health or gingivitis subjectsPresent on crevicular epithelial cellsElimination Elimination or suppress ion resulted in successfu ltherapyRecurrent lesions harbored speciesHost response Elevated serum antibody in periodontitisElevated local antibody response in periodontitisVirulence factors Collagenase, trypsin-like activity, fibrinolysin, otherproteases, phospholipase a, phosphatases,endotoxin, H,S, NH,, fatty acids, and factors whichadversely affect polymorphonuclear leukocytesAnimal studies Important role in experimen tal mixed infectionsStudies in monkeys and dogs

Table 3. Data That Suggest Additional Species as Possible EtiologicAgents of Destructive Periodontal Diseases (adapted from 18)Host Virulence AnimalSpecies Association Elimination Response Factors Studies

P. intennediaF. nucleatumB. forsythusC. rectusE. corrodensP. microsSelenomonas sp.Eubacterium sp .(E. brachyE. nodatumE. timidum)Spirochetes*Indicates relative number of publications.

and the nature of some of the data which support their role.Documentation of the findings is provided in that paper.

-- -PATHOGENS SPECIES

IMPA IREDNEUTROPHILS

A acltn~mycefemcomrrans Aciinomyces spB iorsyrhus C ochraceaINADEQUATE OR E cor iodcns S tnriisUNREGULATED F oucleafum S sangurs 1IMMUNOLOGICAL P micros V parvulaRESPONSE P gingrvalisLPS RESPONSIVENESS P irreiined,a

AIDS C recfusDIABETES Selenomonas spSMOKING Eubacferiiini spSoiiochcfcsDRUGSFigure 2. Diagrammatic representation of 3 groups of factors which de-

termine whether active periodontal disease w ill occur in a su bject or at asite.The evidence for these species differs and it is highly prob-able that some of these species will be removed from thelist and others added. Currently, more data are available tosupport the pathogenic roles of A. actinomyceterncomitansand P. gingivalis than are available for other suspectedperiodontal pathogens.Evolving ConceptsThe foregoing discussion indicated the level of complexitythat has been encountered in the search for the etiologicagents of destructive periodontal diseases. In spite of thiscomplexity, a number of species have been suggested whichmay be pathogens of these diseases. However, it is clearthat while presence of pathogens is necessary for diseaseactivity to take place, it is not sufficient. If the presence ofpathogens were sufficient, then detection of the pathogenwould be synonymous with detection of disease activitymaking diagnostic tests and their interpretation quite sim-ple. Clearly, other factors play a role in the initiation, pro-gression, and remission of destructive periodontal diseases.Figure 2 is a diagrammatic representation of three groupsof factors which determine whether active periodontal dis-ease will occur in a subject or at a site. First, the host mustbe susceptible to the pathogens or disease will not occur.Some of the factors which have been suggested to increasesusceptibility of periodontal subjects include impaired neu-trophils, inadequate or unregulated host immunologic re-sponse, differences in LPS responsiveness, AIDS, diabetes,tobacco, and drug use. Some of these factors will be dis-cussed in greater detail by Drs. Genco and Page and willnot be elaborated further here. The second essential factorfor disease initiation and progression is the presence of oneor more pathogens of the right clonal type in sufficientnumbers to cause disease. A number of suspected pathogensare listed in Figure 2 which were derived from Tables 1 to3. The role of "beneficial species" (to the host) in theprogression of disease is less obvious. As will be discussedbelow, high levels of beneficial species, even in the pres-ence of pathogens, may have a marked effect on diseaseinitiation and progression. Finally, the local periodontal en-vironment may have a profound effect on the expression ofvirulence factors by resident pathogens influencing disease.


5/10

326 CURRENT CONCEPTS OF BACTERIAL ETIOLOGYEXTENT OF DISEASE

0 l o 6 L ~ ~ L - U n-r;i_n 1 LL- 7-C) l o 5 - ~- WIDESPREADC ,A

HEALTHY

1 1 3 3 3 1MAXILLA MANDIBLE

Figure 3. Diagram demonstrating the distribution of A. actinomycetem-comitans serorype b in two healthy subjects and two subjects each withlocalized and widespread disea se. In this figure, sites with prior attach -ment loss > 3 mm are indicated by boxes and bacterial levels by thecircles. Empty spaces represent no disease and no infection (there wereno missing samples in 6 subjects). The numbers below each panel indicatequadrant (1,2,3,4) and tooth (1-7, i. e. central incisor to second m olar)from which mesial samples were taken. (Adapted rom 34).

The Role of Disease SusceptibilityIn ongoing studies of periodontal disease at Forsyth DentalCenter, su bjects with destructive periodontal diseases havebeen evaluated longitudinally using clinical, microbiologi-cal, and immunological parameter^.^^,^' Briefly, 67 sub-jects with prior evidence of destructive periodontal diseaseswere monitored every 2 months for pocket depth, attach-ment level, subgingival temperature, bleeding on probing,suppuration, redness, and plaque. A subject was consideredto be active if he or she exhibited one or more sites withattachment loss > 2.5 mm in a Zmonth period. In addition,the levels of 14 subgingival species were determined insubgingival plaque samples taken from the mesial aspect ofeach tooth in each subject at each visit using a "colony-lift" method and DNA probes." Ten suspected pathogens:A. actinomycetemcomitans serotypes a and b, Bacteroidesforsythus, Fusobacterium nucleatum ss vincentii, Pepto-streptococcus micros, P. gingivalis, Prevotella intermed iahomology groups I and 11, Streptococcus intermedius, andCamp ylobacter rectus (formerly Wolinella recta) and foursuspected beneficial species: Capnocytophaga ochracea,Streptococcus sanguis I and 11, and Veillonella pawula wereenumerated. The results of this prospective study confirmedthe role of certain suspected periodontal pathogens.35One of the most interesting findings from this type ofinvestigation was the demonstration that microbial speciesare not evenly distributed from subject to subject or fromsite to site in the same subject. Figures 3 and 4 indicate thedistribution and levels of P. gingivalis and A. actinomy-cetemcomitans serotype b in two healthy, two minimallydiseased, and two widespread disease subjects. A actinomy-cetemcomitans serotype b was found less frequently inwidespread disease subjects than in subjects with localizeddisease (Fig. 3). In the example shown, one widespreaddisease subject (MA) did not harbor the organism at anysampled site, while one of the localized disease subjects(CD ) harbored the species in 12 sites. Interestingly, one of

J PeriodontApril 1992 (SupplemenEXTENT OF DISEA

I I I: I t 6 t i 4 11 12 1 1 21 2 2 2 3 2 % 21 2 5 2 7 3 7 3 C 3131 1 1 3 2 11 4) 6 2 4 3 r < : / 5 r :

MAXILLA MANDIBLE

5sWIDESPREAD

MA

MTLOCALIZED

CD

DOHEALTHY

AM

Figure 4. Diagram demonstrating the distribution of P . gingivalis in healthy subjects and 2 subjects each with localized and widespread disease. F or details regarding the composition of the diagram see F igure 3(Adapted rom 34).

the healthy subjects harbored the species in four sites, albeat low levels. Figure 4 provides examples of different patterns of colonization by P. ging.valis. While this speciewas not detected in any sampled site in one healthy (AMand one localized disease subject (CD), it was found ihigh numbers in 23 of 28 sites in one of the subjects witwidespread disease (MA). Although many subjects harbored this species, the number of sites colonized and thpattern of colonization were different. The lack of homogeneity of distribution of a given species is important noonly in understanding disease initiation and progression, buin designing sampling procedures for diagnostic tests.While monitoring these subjects longitudinally, it became apparent that subjects with widespread disease exhibited more sites showing new attachment loss than subjectwith fewer affected sites at baseline. The relationship between prior extent of disease and new disease progressionhas been repor ted in other ~tudies .~~-~Ohis observatiomight have been due to the presence of higher levels of thesuspected pathogens in the widespread disease subjects. Totest this hypothesis, the mean p ercentages of the total cultivable microbiota for each species in subjects with differenlevels of periodontal destruction at baseline were computed(Fig. 5 ) . Percentages of suspected pathogens were highesin subjects with localized destruction and lowest in thewidespread disease subjects. Such findings could be explained in a number of ways. First, the pathogens of themore widespread forms of disease may not have been amongthe test species. Second, the pathogens m ay have been correct but not of a virulent clonal type. Third, the pathogensmay have been of a virulent clonal type, but the presenceof beneficial species (to the host) or unfavorable environmental conditions precluded disease in the localized diseaseindividuals. Finally, and most relevant to this section, wasthe possibility that the hosts differed in their threshold sus-ceptibility to given pathogenic species. This last hypothesiswas supported by the observation that lower threshold percent of sites colonized by suspected pathogens including,C. rectus, P. intermedia I , P. gingivalis, and A. actinomycetemcomitans serotypes a and b were needed to produceodds ratios > 4 for disease progression in subjects with


6/10

Volume 63Number 4 SOCRANSKY, HAFFAJEE 32716

MEAN %TOTAL

CULTIVABLEFLORA

12

A, actinomi/cetemcomitans aF, nucleatum ss.vlncentl1

16MEAN %TOTAL

CULTIVABLEFLORA

12

A. acti,nomycetemcomi,tans58 A, act~nomycetemcom~tansF, nucleatum ss.vlncentl1

4 P, intermedia IP. intermedia II

0LOCALIZED WIDESPREADINTERMEDIATE

Figure 5. Stacked bar chart of the mean levels of the 10 suspected pathogen s detected using DN A probes insubjects with localized (< 16 % of sites with attachment loss > 3mm at baseline), intermediate (16% to 44%affected sites) and widespread j > 44% affected sites) periodontal destruction. The mean for each species wascomputed by averaging the levels oftha t species for each subject and then averaging the means for all subjectswithin the localized, interm ediate, or widespread disease groups. The shading of the bars is used for o rientationpurposes only.

widespread baseline attachment 10~s.~'ubjects who exhib-ited both widespread disease and high levels of suspectedpathogens exhibited mo re active sites than subjects in otherg r o ~ p s . ~ 'Bacterial InteractionsPeriodontal pathogens do not exist in dental plaque in iso-lation. Indeed this has been one of the most confusing as-pects of research in this area. Undoubtedly, the otherorganisms in the pocket affect the ability of pathogens tocause disease progression. In certain situations, the pres-ence of one species could facilitate the colonization of an-other pathogenic species or act synergistically with thatspecies to cause disease. A lternatively, a bacterial speciescould be beneficial to the ho st acting to prevent or minimizetissue damage.

During the course of enumerating subgingival species inplaque samples using a "colony-lift" method, it was notedthat a number of colonies produced areas of P hemolysison blood agar plates. These colonies included strains ofActinomyces naeslundii, Actinomyces odontolytcus, Acti-nomyces israelii, A. viscosus, Streptococcus pyogenes,Streptococcus constellatus, Prevotella melaninogenica, andPrevotella denticola (unpublished results). The remarkablystrong relationship between the presence of p hemolyticspecies in a subject or at a site and the levels of "black-pigmented Bacteroides" species is of interest to the presentdiscussion. For examp le, mean coun ts of "black-pigmented

Bacteroides" species (P. intermedia I, P. intermedia 11,and P. gingivalis) at sites where P hemolytic species werenot detected were 0.90 -+ 0.20 (95% C1, n = 1758) and2.73 + 0.38 (n = 1035) at sites where P hemolytic specieswere detected (unpublished data). While there are a numberof interpretations of this observation, perhaps the simplestis that the p hemolytic species cause damage to red bloodcells or other host tissue cells and foster the colonizationof the suspected periodontal pathogens. These in turn mightlead to tissue damage. (One alternative explanation is that"black-pigmented Bacteroides" species result from tissuedamage.)Bacterial interactions can also be beneficial to the host.For example, a species could affect disease progression ina number of ways: 1) by "passively" occupyin g a nichewhich might otherwise be colonized by a pathogen; 2) byactively limiting a pathogen's ability to adhere to appro-priate tissue surfaces; 3) by adversely affecting the vitalityor growth of a pathogen; 4) by affecting the ability of apathogen to produce virulence factors; or ' 5) by degradingvirulence factors produced by the pathogen. One well-doc-umented example of a bacterial interaction which is bene-f ic ia l to the hos t is the e ffec t of S. sanguis o n A.actinomycetemcomitans. S. sanguis pro duces H,O,, wh icheither directly or by host enzyme amplification kills A. ac-tinomycetem~omitans.~~*~~nother example of a relation-ship which may be beneficial to the host is provided inFigure 6 where the levels P. gingivalis and C. ochracea in


7/10

328 CURRENT CONCEPTS OF BACTERIAL ETIOLOGYJ PerioA~ r i l 992 (Su~~

INACTIVE ACTIVEMEAN% lo j7-T= 0.08 + 0 . 7 ~ y = 1.78 - 0 . 4 ~P. gingivalis r = 0.72 r = -0.23

0 5 10 0 5 10MEAN % C. ochracea

Figure 6. Scatter plots of the mean % of C . ochracea (x-axis) and mean% P . gingivalis (y-axis) n subjects subset on the basis of showing (active)or not showing (inactive) I or more sites with attachment loss > 2.5 mmin 2 months. Each circle represents the mean values for one subject.(Adapted from 41).

active and inactive subjects are presented. W hen high leve lsof P. gingivalis were found in subjects with high levels ofC. ochracea, the risk of new attachment loss in subjectswas diminished. In contrast, subjects with high levels of P.gingivalis and low levels of C. ochracea tended to exhibitdisease progression.Ano ther exam ple of the role of "beneficia l" spec ies wa sobserved in preliminary results from an ongoing treatmentstudy at Forsyth. Sub jects were treated by Widman flapsurgery and scaling together with one of the four followingsystemic agents: tetracycline, augmentin, ibuprofen, or pla-cebo. Subjects receiving the adjunctive antibiotic therapyshowed a higher percentage of sites with attachment levelgain and higher levels of the suspected beneficial speciesC. ochracea and S. sanguis I1 post-therapy than subjects inthe other two groups. Although this study is far from com-plete, the early results indicated that successful therapy maybe dependent not only on the reduction or elimination ofpathogens, but on subsequent colonization of high levels of"beneficial" species (unpublished data).The Concept of Virulent Clonal Types of PathogensThe concept of bacteria playing important roles in destruc-tive periodontal diseases was reintroduced in the 1950s whenit was suggested that plaque control was essential in thetreatment of periodontal patients. This finding gave rise tothe non-specif ic p laque hypothesi~ .~~ater studies, as in-dicated above, suggested that all organisms in plaque maynot be equally capable of causing destructive periodontaldiseases and thus the concept of specificity in destructiveperiodontal disease re-emerged. In recent years, infectiousdisease workers in medicine have indicated that not all clonaltypes of a pathogenic species are virulent for man. Forexam ple six of 104 clonal types of Haemophilus influenzaeappear to be virulent, w hile two of 15 clonal types of Hae-mophilus influenzae biogroup aegyptius have been shownto be associated with Brazilian purpuric It seemslikely that here might be multiple clonal types of periodon-tal pathogens, some of which are virulent and others ofwhich are not. This is an attractive hypothesis in that it

might explain observations such as the presence of "pogens" in healthy sites or subjects and the return of "pogens" to successfully treated sites. In order to supporhypothesis, it will be necessary to demonstrate: 1) that are multiple clonal types within a pathogenic speciethat these clonal types differ in pathogenicity; and 3)som e clonal types are associated with health and othersdisease. There are a number of ways in which investigcan distinguish clonal types. Among these are the exnation of multi-locus enzyme patterns and of ribotypeindividual strains.48 The latter approach has been useexamine ribotypes of P. [email protected] and C. rectus (unlished data). Preliminary data from these studies havdicated that there are multiple clonal types in each of tspecies. The second requirement, differences in pathnicity for different clonal types, has not yet been elished. However, it has been shown that there are sdifferences in the pathogenicity of P. gingivalis isolatanimal test systems.19-21~23-25hese differences may blated to clonal type, but this has not been determined.final test, the comparison of clonal types isolated from cand controls, has not been initiated as yet; however, testing will be of enormous interest.Transmission of Virulence FactorsThe possibility of the transfer of genetic information cofor virulence properties from o ne strain or species to anodeserves consideration. Imagine, if you will, a pathowith virtually all the genetic factors needed to cause disbut lacking in a single key element. For exam ple, the smight be lacking an essential adhesin, toxin, or hemolyConceivably, this strain might acquire the genetic inmation needed to produce this factor from other cellthat or other species in that environment. There is prdence for this concept in that transfer of virulence facby medically-important bacteria have been shown as as the transfer of antibiotic resistance. Indeed, bactphage-related virulence has been ascribed to strains oactinomycetemcomitans.49~50ntibiotic resistance has transferred from P. denticola to P. i n t e rmed i ~ .~~hesence of an essential extra-chromosomal element mighplain failure of a putative pathogen to cause disease manner analogous to differences in clonal types as cussed above.The Importance of Regulation by the LocalEnvironmentConsider for a m oment the possibility that a virulent cltype of a pathogenic species colonizes a site in seemisufficient numbers to cause disease in a susceptible and yet disease does not occur. The possibility that "beficial" species have ameliorated this process has beencussed above. However, there might be other explanatfor this phenomenon. Until recently, w e and perhaps othad the view that a cell of a virulent pathogenic specontinuously produced its virulence factors. Recent w


8/10

Volume 63Number 4 SOCRANSKY.HAFFAlEEin other fields suggest that this may not be the case. It hasbeen shown that strains of many pathogenic species "turn-on" or "turn -off" production of virulence factors depend-ing on the n atu re of th eir e n v i r ~ n m e n t . ~ ~ ~ ' ~ ~ ~ 'pparentlythe species "senses" its environm ent and a regulating geneaffects a global network of genes which simultaneously turnon or off production of multiple virulence factors." Tem -perature, iron concentration, osmolarity, magnesium, andcalcium all have been shown to be environmental factorswhich influence the production of virulence factor^.^^^'^^^"It has been shown that production of virulence factors bysuspected periodontal pathogens can be influenced by en-vironment. For example, expression of certain cell mem-brane proteins and pathogenicity in animal model systemsby P. gingivalis is influenced by the level of iron or heminin the medium in which the strain is g r ~ w n . ' ~ - ' ~emper-ature has been shown to differ from subgingival site tosubgingival site in the same oral cavity and in differentsubject^.'^ In addition, higher subgingival temperature ap-pears to confer higher risk of new attachment loss and in-fluence (or be influenced by) the levels of certain subgingivals p e ~ i e s . ~ ~ , ~ "t is tempting to speculate that subgingival tem -perature or other environmental factors might influence theexpression of virulence factors by virulent clonal types ofpathogens and turn these quiescent subgingival residentsinto vehicles of destruction. Thus, a change in environmentresulting from a local traumatic incident such as food impac-tion might result in a local inflammation and an alteration inthe environment which induces the production of virulencefactors by a resident pathogen in that site. Although the datarelating environment to pathogenicity are limited, the conceptis so attractive that it warrants further investigation.DISCUSSIONThe material presented above attempts to describe some ofthe changing concepts in the etiology of destructive peri-odontal diseases. These changes are driven, in part, bydiscrepancies between microbial and clinical status and, inpart, by parallel recognitions in infectious disease micro-biology. Infection does not mean instantaneous disease. Evenagents as lethal as the HIV virus can persist in individualsfor months to years before clinical manifestations are ob-served. Infection by a species does not necessarily meanthat disease is imminent. The carrier state of quite patho-genic species such as Salmonella typhi provide examplesof this phenomenon. All strains of a pathogenic species arenot pathogenic. A test of the presence of Escherichia coliin the human intestinal tract would probably reveal the pres-ence of this species in most individuals. However, mostindividuals do not get diarrhea associated with this speciesbecause only a subset of E. coli strains are entero-toxigenic.Thus, lag phase, host susceptibility, and clonal type canaffect interpretation of diagnostic tests.

I t would be simple and convenient if one could take asample of plaque and detect a pathogen and know imme-diately that this was the cause, or about to be the cause, of

disease at that site. The foregoing discussion should wus that this may not be the case. The outcome, diseinitiation or progression, is a resultant of the interplay tween a large number of factors. The presence of a paogenic species is necessary, but not sufficient for diseto take place. In order that disease result from this pathgen, 1) it must be of a virulent clonal type; 2) it mpossess the chromosomal and extra-chromosomal genefactors to initiate disease; 3) the host must be susceptito this pathogen; 4) the pathogen must be in numbers sficient to exceed the threshold for that host; 5) it must located at the right place; 6) other bacterial species mfoster, or at least not inhibit, the process; and 7) the loenvironment must be one which is conducive to the exprsion of the species' virulence properties. With this list mind it is not surprising that disease activity, at least thwhich causes major destruction, appears to be a somewhuncommon event. Assuming that all factors are in order an episode of active disease to take place, the episode itsmarkedly changes the equation leading to a massive loresponse by the host, shifts in the other oral microorgaisms, and establishment of a new equilibrium.The foregoing discussion highlights both the complexof the problem and the fact that investigators have sharened the focus of their search. If one had asked "Whcauses periodontal disease?" 150 years ago, a likely answmight have been "God only knows." Over time, the sponses to that question might have shifted to "accumlations of bacteria" or "specific bacterial spec ies 7' and,the above discussion were correct, to "specific bacteria the right clonal type with the essential genetic elementssufficient numbers for that host with appropriate additionspecies in the right environm ent." Such progression deonstrates both the evolution of the concepts and the iproved focus of investigation. In turn, improvement in focshould lead to improvement in potential diagnostic tesFor example, a diagnostic test might not seek a species, ba specific clonal type of a species or a critical element that species such as the gene for toxin production in thecoli example cited above. In addition, diagnostic tests fbacteria would probably not exist in isolation, but in combination with other diagnostic tests of host susceptibiliand/or status of the local environment. Given sharpenunderstanding of the biologic basis of disease, it may surmised that combinations of diagnostic tests will be usefin determining not only the pathogens of disease, but ttherapies best suited for control.AcknowledgmentsThis work was supported in part by research grants D04881 and DE-02847 from the National Institute of DentResearch.REFERENCES1. Koch R. Die aetiologie der tuberculose. Berliner Klinische W ochschn'ft 1882;19:221-230.


9/10

330 CURRENT CONCEPTS OF BACTERIAL ETIOLOGYJ PeriodontolApri l 1992 (Su ~ole me nt)

2. Koch R. Kritische Besprechung der gegen die Bedeutung der Tub-erkeilbazillen gerichteten Publikationen. Deutsche MedizinischeWochenschrift 1883;9: 137-141.

3. Fraser DW, Tsai T, Orenstein W, et al. Legionnaires' disease I:Description of an epidemic of pneumonia. New England J Med1977;297:1189-1197.

4. McDade JE, Shepard CC, Fraser DW, et al. Legionnaires' disease:Isolation of a bacterium and demonstration of its role in other respi-ratory diseases. New England J Med 1977;297:1197-1203.5. Socransky SS, Haffajee AD, Listgarten MA. Microliiology (plaque).In: Grant DA, Stern IB, Listgarten MA, eds. Periodontics, 6th ed.St. Louis: C.V. Mosby: 1987:147-197.

6. Listgarten MA, Socransky SS . Ultrastructural characteristics of a spi-rochete in lesions of acute necrotizing ulcerative gingivostomatitis(Vincent's infection). Arch Oral Biol 1964;9:95-96.

7. Listgarten MA. Electron microscopic observations of the bacterialflora of acute necrotizing ulcerative gingivitis. J Periodontol1965;36:328-339.

8 . Jordan HV, Keyes PH. Aerobic, Gram-positive, filamentous bacteriaas etiologic agents of experimental periodontal disease in hamsters.Arch Oral Biol 1964;9:401-414.

9. Newman MG, Socransky SS. Predominant cultivable microbiota inperiodontosis. J Periodont Res 1977; 12: 12 c1 28 .

10. Slots J. The predominant cultivable organisms in juvenile periodon-titis. Scand J Dent Res 1976;84:1-10.

11. Newman MG, Socransky SS, Savitt ED, Propas DA, Crawford A.Studies of the microbiology of per iodontosis . J Periodontol1976;47:373-379.

12. Slots J. The predominant cultivable microflora of advanced peri-odontitis. Scand J Dent Res 1977;85:114-121.

13. Spiegel CA, Hayduk SE, Minah GE, Kryolap GN. Black-pigmentedBacteroides from clinically characterized periodontal sites. J Perio-dont Res 1979; 14:376-382.

14. Tanner ACR, Haffer C, Bratthall GT, Visconti RA, Socransky SS.A study of the bacteria associated with advancing periodontal diseasein man. J Clin Periodontol 1979:6:278-307.

15. White D, Mayrand D. Association of oral Bacteroide.~with gingivitisand adult periodontitis. J Periodont Res 1981;16:259-265.

16. Zambon JJ, Reynolds HS, Slots J. Black-pigmented Bacteroides spp.in the human oral cavity. Infect lmmun 1981;32:198-203.

17. Socransky SS, H affajee AD , Sm ith GLF, Dzink JL. Difficu!ties en-countered in the search for the etiologic agents of destructive peri-odontal diseases. J Clin Periodontol 1987;14:588-593.

18. Socransky SS, Haffajee AD. Microbiological risk factors for destruc-tive periodontal diseases, In: Bader JD, ed. Risk Assessment in Den-tistry. Chapel Hill: University of North Carolina Dental Ecology;1990:79-90.

19. Grenier D, Mayrand D. Selected characteristics of pathogenic andnonpathogenic strains of Bacteroides gingivalis. J Clin Microbiol1987;25:738-740.

20. Van Steenbergen TJ, Delemarre FG, Namavar F, De Graaff J. Dif-ferences in virulence within the species Bacteroides gingivalis. An-tonie Van Leeuwenhoek 1987;53:233-244.21. McKee AS, McDermid AS, Wait R, Baskerville A, Marsh PD . Iso-lation of colonial variants of Bacteroides gingivalis W50 with a re-duced virulence. J Med Microbiol 1988;27:59-64.

22. Marsh PD, McKee AS, McDermid AS, Dowsett AB. Ultrastructureand enzyme activities of a virulent and an avirulent variant of Bac-teroides gingivalis W50. Fems Microbiol Let 1989;50:181-185.

23. Neiders ME, Chen PB, Suido H, Reynolds HS, Zambon JJ. Heter-ogeneity of virulence amon g strains of Bacteroides gingiv alis. J Per-iodont Res 1989;24:192-198.

24. Shah HN, Seddon SV, Gharbia SE. Studies on the virulence prop-erties and metabolism and pleiotropic mutants of Porphyromonas gin-givalis (Bacieroides gingivalis) W50. Oral Microbiol lrnmunol1989;4:19-23.

25. Smalley JW, Birss AJ, Kay HM , McKee AS, Marsh PD. The distri-

bution of trypsin-like enzyme activity in cultures of a virulent and anavirulent strain of Bacteroides gingiv alis W50. Oral Microbiol Im-munol 1989;4: 178-181.

26. Koch R. Erste Konferenz zur Eroterung der Cholerafrage. BerlinerKlinische Wochenschrift 1884;30:20-49.

27. Carter KC. Essays of Robert Koch. New York: Greenwood Press;1987; xviii-xix.

28. Kornman KS , Holt SC, Robertson PB. The microbiology of ligature-induced periodontitis in the cynomologus monkey. J Periodont Res1981;16:363-371.

29. Kornman KS, Siegrist B, Soskolne WA, Nuki K. The predominantcultivable subgingival flora of beagle dogs following ligature place-ment and metronidazole therapy. J Periodont Res 1981;16:251-258.

30. Holt SC, Ebersole J, Felton J. Brunsvold M, Kornman KS. Implan-tation of Bacteroides gingivalis in nonhuman primates initiates pro-gression of periodontitis. Science 1988;239:55-57.

31. Socransky SS, Haffajee AD. Microbial mechanisms in the pathogen-esis of destructive periodontal diseases: A critical assessment. J Per-iodont Res; 1991;26: 195-212.

32. Ebersole JL, Taubman MA, Smith DJ, Genco RJ, Frey DE. Humanimmune responses to oral microorganisms. I. Association of localizedjuvenile periodontitis (LJP) with serum antibody responses to Acti-nobacillus actinomycetemcomitans. Clin Exper Immunol 1982;47:43-3L .

33. Ebersole JL, Taubman MA, Smith DJ, Hammond BF, Frey DE.Human immune responses to oral microorganisms. 11. Serum antibodyresponses to antigens from Actinobacillus actinomycetemcomitans andthe correlation with localized juvenile periodontitis. J Clin Immunol1983;3:321-331.

34. Haffajee AD, Socransky SS, Smith C, Dibart S. The use of DNAprobes to examine the distribution of subgingival species in subjectswith different levels of periodontal destruction. J Clin Periodontol1992;19:84-91.

35. Haffajee AD, Socransky SS , Smith C, Dibart S. Relation of baselinemicrobial parameters to future periodontal attachment loss. J ClinPeriodontol 1991 18:744-750.

36. Gunaratnam M, Smith GLF , Socransky SS, Smith CM, H affajee AD.Enumeration of subgingival species on primary isolation plates usingcolony lifts. Oral Microbiol Immunol 1992;7:14-18.

37. Albandar JM , Rise J, Gjermo P, Johansen JR. Radiographic quanti-fication of alveolar bone level changes. A 2-year longitudinal studyin man. J Clin Periodontol 1986;13:195-200.

38. Axelsson P, Lindhe J. Effect of controlled oral hygiene procedureson caries and periodontal disease in adults. J Clin ~eriodon to l1978;5:133-151.

39. Haffajee AD, Socransky SS, Dzink JL, Taubman MA, EbersoleJL. Clinical, microbiological and immunological features of sub-jects with refractory periodontal diseases. J Clin Periodontol1988;15:390-398.

40. Haffajee AD, S ocransky SS, Lindhe J, Kent RL, Okamoto H, Yoney-ama T. Clinical risk indicators for periodontal attachment loss. J ClinPeriodontol 1991; 8: 1 17-1 25.

41. Haffajee AD, Socransky SS, Smith C, Dibart S. Microbial risk in-dicators for periodontal attachment loss. J Periodont Res 1991;26:292-296.

42. Hillman JD, Socransky SS, Shivers M. The relationships betweenstreptococcal species and periodontopathic bacteria in human dentalplaque. Arch Oral Biol 1985;30:791-795.

43. Hillman JD, Socransky SS. The theory and application of bacterialinterference to oral diseases. In: Myers HM, ed. New Biotechnologyin Oral Research. Basel: Karger; 1989: 1-17.

44. Loesche WJ. Chemotherapy of dental plaque infections. Oral Sci Rev1976;9:65-107.

45. Brenner DJ, Mayer LW, Carlone GM, et al. Biochemical, genetic,and epidemiologic characterization of Haemophilus influenzae bio-group aegyptius (Haemophilus aegyptius) strains associated with Bra-zilian purpuric fever. J Clin Microbiol 1988;26:1524-1534.


10/10

Volume 63Number 4 SOCRANSKY, AFFAJEE 3346. Musser JM, K roll JS, Moxon ER, Selander RK. Evolutionary geneticsof the encapsulated strains of Haemo philus influenzae. Proc Natl Acad

Sci (USA) 1988;85:7758-7762.47. Finlay BB, Falkow S. Common themes in microbial pathogenicity.Microbiol Rev 1989;53:210-230.48. Eisenstein BI. New molecular techniques for microbial epidemiologyand the diagno sis of infectious diseases. JInfect Dis 1990;161:595402.49. Preus HR, Olsen I, Namork E. Association between bacteriophage-

infected Actinobacillus actinomycetemcomitans and rapid periodontaldestruction.J Clin Periodontol 1987;14:245-247.50. Preus HR, Olsen I, Namork E. The presence of phage-infected Ac-tinobacillus actinomycetemcomitans in localized juvenile p eriodon titispatients. J Clin Periodontol 1987;14:605409.

51. Guiney DG, Bouic K. Detection of conjug al transfer systems in oral,black-pigmented Bacteroides spp. J Bactenol 1990;172:495497.

52. Maurelli AT. T emp erature regulation of virulence genes in pathog enicbacteria: a general strategy for human pathogens? Microbial Patho-genesis 1989; 7:l-10.53. Miller JF, Mek alanos JJ, Falkow SF . Coordina te regulation and sen-sory transduction in the control of bacterial virulence. Science

1989;243:91&922.

54. McKee AS, McDermid AS, Baskewille A, Dowsett AB, EllwooDC. Effect of hemin on the physiology and virulence of Bacteroidegingivalis W50. Infect Immun 1986;52:349-355.

55. Barua PK, Dyer DW , Neiders ME. Effect of iron limitation on Bacteroides [email protected]. Oral Microb iol Immu nol 1990;5:263-268.56. Bramanti TE, Holt SC. Iron-regulated outer membrane proteins in thperiodontopathic bacterium, Bacteroides gingivalis. Biochem BiophyRes Comm 1990;166: 1146-1 154.

57. Haffajee AD, S ocransky SS, G oodson JM. S ubgingival temperaturI Relation to baseline clinical parameters. J Clin Periodonto1992;19:00-00.

58. Haffajee AD, S ocransky SS, Goodson JM. Su bgingival temperaturI1 Relation to future periodontal attachment loss. J Clin Pt:riodonto1992; in press.59. Haffajee AD, Socransky SS, Smith C, Dibart S, Goodson JM

Subgingival temperature. 111 Relation to micro bial co unt s. J Clin Periodontol 1992; in press.

Send reprint requests to: Dr. S.S. Socransky, Forsyth Dental Cente140 The Fenway, Boston, MA 02115.

Documents

Senador @JoseRafaelS depositó proyecto de ley que declara el día 25 de septiembre de cada año como "Día de la Reafirmación Democrática"