75

Molecular Genetic Analysis of Actinobacillus actinomycetemcomitans

Epidemiology* Joseph J. Zambón, Gregory J. Sunday,

and John S. Smutko

Research over the past decade has identified many of the microorganisms involved in the etiology of human Periodontitis such as Actinobacillus actinomycetemcomitans. Efforts are now directed toward defining these species' role in the pathogenic process. Since microbial colonization of host tissues is a key first step in developing a bacterial infection, determining the source of the periodontal pathogens and their route of transmission is likely to be crucial in formulating preventive strategies. Recently, a technique from molecular biology, restriction endonuclease analysis, has been used to track bacterial infections. In the present study, this method was used to investigate the epidemiology of A. actinomycetemcomitans infection. One hundred twenty-four human subgingival plaque isolates of A actinomycetemcomitans were examined including bacterial strains from the United States, Korea, and Norway as well as 15 strains from cynomolgus (Macaca fasicularis) and spider monkeys (Macaca iris) and 4 reference strains. The genomic DNA from each strain was purified, digested with each of 16 restriction endonucleases, and the DNA digests were resolved by electrophoresis. The resulting patterns of DNA fragments were compared and also correlated with the A actinomycetemcomitans serotype determined using serotype-specific antisera in immunofluorescence. Human isolates of A actinomycetemcomitans even from disparate geographic sources showed little diversity by restriction endonuclease analysis. Three major restriction patterns were found. Restriction pattern I was common to all 20 of the serotype a isolates, restriction pattern II was associated with 58% of the 73 serotype b isolates examined, while restriction pattern III was associated with the remaining serotype b strains and with all 15 of the serotype c strains. All A. actinomycetemcomitans isolates from a single site, or from several sites in the same patient, or from a single patient over time showed the same restriction pattern suggesting a clonai infection by this microorganism. By contrast, strains of A actinomycetemcomitans isolated from non-human primates revealed numerous and distinct restriction endonuclease patterns none of which resembled human strains. These results are consistent with the idea that human oral infection by A. actinomycetemcomitans is clonai in any one patient and that human oral infection is associated with a limited number of possible genotypes. Restriction endonuclease analysis of genomic DNA from A actinomycetemcomitans and other periodontal pathogens provides a powerful tool for analyzing the microbial ecology of subgingival plaque as well as determining the mode of transmission and acquisition for individual species. JPeriodontol 1990;61:75-80.

Key Words: Periodontitis/pathogenesis; periodontitis/microbiology; restriction endonuclease analysis; Actinobacillus actinomycetemcomitans/malysis.

Specific microorganisms in dental plaque are thought to be important and, likely, etiologic in the pathogenesis of the different forms of human Periodontitis. Actinobacillus ac*Departmcnts of Pcriodontology and Oral Biology and the Periodontal Disease Clinical Research Center, State University of New York at Buffalo, School of Dental Medicine, Buffalo, NY.

tinomycetemcomitans is one such bacterium which has been

strongly implicated in the etiology of localized juvenile Periodontitis and also in certain cases of rapidly progressing and refractory Periodontitis in adults.1 As the importance of A actinomycetemcomitans in Periodontitis has been elucidated, increased attention has been focused on the means by which A actinomycetemcomitans is transmitted into the

J Periodontol 76

EPIDEMIOLOGY OF A. ACTINOMYCETEMCOMITANS BY DNA FINGERPRINTING

oral

cavity.

In order to facilitate these types of epidemioit is often useful to differentiate bacteria within studies, logie the same species. A. actinomycetemcomitans has, for ex-

been categorized into from three to ten biotypes the variable fermentation of key sugars including dextrin, galactose, mannitol, maltose, and xylose2^, and into three serotypes defined by heat stable carbohydrate

ample,

based

on

antigens.5

In addition to serotyping and biotyping, a number of techniques have been recently developed to examine the epidemiology of infectious diseases. One method, restriction endonuclease analysis, has been used to examine the transmission of microbial pathogens such as Legionella species,6 Neisseria meningitidis,1 and Clostridium difficile* Restriction endonuclease analysis involves the isolation of genomic DNA from a microorganism, digestion with restriction enzymes, and then comparison of the DNA fragments on gel electrophoresis. Since restriction enzymes cut DNA at specific sites based on the sequence of neucleotide base pairs, the resulting electrophoretogram forms a DNA "fingerprint" for that bacterium. If the banding pattern for

different, it suggests that the two strains are genetically different. If two strains show an identical bandtwo strains is

ing pattern

when tested with several different restriction then enzymes, they are likely to be, but not necessarily, identical. Since restriction endonuclease analygenetically sis examines the entire bacterial chromosome, it is often more sensitive than methods such as serotyping or biotyping which examine the variability of only a few genetic loci. In the present study, we utilized restriction endonuclease analysis to examine the epidemiology of A actinomycetemcomitans subgingival infection.

February 1990

caton, Arizona; Korea; and

Norway, isolates from cynomolgus monkeys (Macaca fasicularis) and spider monkeys (Macaca iris) as well as culture collection strains. Isolates were identified as A actinomycetemcomitans if they exhibited small

(0.5

to 1.0 mm in diameter) smooth, cirand translucent colonies with a slightly convex, and an internal irregular edge star-shaped morphology; and, if the bacterial cells were Gram-negative, capnophilic, nonmotile, non-hemolytic, indole negative, fermentative, catalase-positive coccobacilli which do not require X (hemin) or V (NADH) factors nor fermented lactose, sucrose, or trehalose.9 As a control, one strain of the closely related species Haemophilus aphrophilus was also examined by restriction endonuclease analysis. The stability of the A. actinomycetemcomitans restriction profiles was examined using 11 strains taken from the same four patients over a period of between 9 months and 2 years. Since strains of A actinomycetemcomitans may exhibit both rough and smooth colony morphologies, it was also of interest to determine the relationship between restriction enzyme profile and colony morphology. Genomic DNA from the rough and smooth variants of three strains of A actinomycetemcomitans from Norway were examined.

cular,

mm

Immunofluorescence

Serotyping Using serotype-specific rabbit antisera, the A actinomycetemcomitans isolates were categorized into one of three serotypes by means of indirect immunofluorescence.5 Three to 5 day old cultures of each A actinomycetemcomitans isolate on tryptic soy agar containing 5% sheep blood were harvested by means of a platinum loop and suspended in phosphate buffered saline (PBS, pH 7.12) to a concentration O.D.540nm 0.7. Ten µ of the bacterial suspension was placed onto glass slides and gently heat-fixed. The bacterial =

MATERIALS AND METHODS

stained for 10 minutes with 25 µ of serotypeantisera diluted to working titer (the highest two-fold serial dilution still giving brilliant fluorescence of the cell envelope) in phosphate buffered saline (PBS, pH 7.2) containing 0.05% Tween 20 (PBS-T). The slides were then gently rinsed with PBS-T, washed in PBS, and rinsed smears were

Bacterial Strains A total of 124 A. actinomycetemcomitans isolates representing 70 strains were examined by restriction endonuclease analysis (Table 1). These included human isolates from patients in Buffalo, New York; Boston, Massachusetts; SaTable 1: A. Patient

specific rabbit

actinomycetemcomitans

diagnosis

Juvenile Periodontitis Adult Periodontitis Periodontally Normal Juvenile Periodontitis Adult Periodontitis Juvenile Periodontitis Juvenile Periodontitis Periodontitis Periodontitis Reference Strains Total

Isolates Examined

Source Buffalo Buffalo Buffalo Boston Sacaton Korea

Norway

Rochester Buffalo

ATCC

Human Huir, a Human Human Human Human Human

Monkey Monkey

by

Restriction Endonuclease

Number of Strains (Isolates) 27

(83) 5(5) 1 (1) 2(2) 1 (1) 8(8) 7(7) 4(4) 11 (11)

4(4) (126)

70

Analysis

Scrogroup

Distribution of Strains (Isolates) A c 8(17) 16 (63) 3(3) 5 (5) 1 1

(1) (1)

(1) (1) 9(8) 1 1

4(4) 11 (11) 1 (1)

7(7) '

3(3)

Volume 61 Number 2

ZAMBON, SUNDAY, SMUTKO

with distilled water. The slides were incubated with 25 µ of goat anti-rabbit IgG conjugated with fluorescein isothiocyanate (FITC isomer I, fluorescein-to-protein ratio of 25 µg/mg) diluted in PBS-T to working titer. The slides were rinsed and washed as before and then mounted with glycerol in PBS (2:1 v/v), pH 9.0. The bacterial smears were examined within 2 hours of staining using a Zeiss standard 14 microscope equipped for phase contrast illumination and for incident light fluorescence. The light source was a 100 W halogen lamp with a BP 450 to 490 interference filter in its exciting pathway and with an LP 520 barrier filter. Fluorescence was graded from 0 to 4 + with grades 3+ and 4+ considered serologically positive reactions. ,

Preparation of Bacterial

DNA Total cellular DNA was extracted from the bacterial cells by a modification of the method of Silhavy, Berman, and Enquist.10 Each bacterial isolate was inoculated into 100 ml of brain heart infusion broth containing 5 mg/L hemin, 1 gm/L sodium bicarbonate and vitamin Kj and grown overnight at 37°C. The bacterial cells were harvested by centrifugádon at 7000 g for 15 minutes, washed once in STE buffer (0.1 M NaCl, 10 mM Tris-HCl pH 8.0, 1 mM EDTA), and suspended in 4 ml of 50 mM Tris-HCl, pH 8.0, 50 mM EDTA. Lysozyme* was added to a final concentration of 500 µg/ml and the cells were incubated at 0°C for 30 minutes. The cells were lysed by the addition of 0.8 ml STEP buffer (0.5% SDS, 50mM Tris-HCl pH 8.0, 0.4 M EDTA, 1 mg/ml Proteinase ). The samples were gently mixed by inversion and incubated at 65°C for 1 hour. The samples were then extracted three times with equal volumes of buffer-saturated phenol-chloroform-isoamyl alcohol (25:24:1) and three times with chloroform-isoamyl alcohol (24:1). Ammonium acetate was added to a final concentration of 2.5M and the DNA was precipitated by the addition of three volumes of 95% ethanol. The DNA was hooked out with a bent glass rod, rinsed once in 70% ethanol, and redissolved in 3 ml 50 mM Tris-HCl pH 7.5 ImM EDTA. RNAse A* was added to a final concentration of 50 µg/ml and the samples were incubated overnight at 4°C. The following day, SDS and Proteinase were added to final concentrations of 0.5% and 50 µg/ml, respectively. The samples were incubated at 65°C for 1 hour and then extracted twice with phenol-chloroform-isoamyl alcohol (25:24:1) and twice with chloroform-isoamyl alcohol (24:1). The samples were re-precipitated with NH4OAc and ethanol and stored at -20°C in 500 µ of lOmM Tris, pH 8.0, 1

mM EDTA. Restriction Endonuclease

Analysis

of A.

actinomycetemcomitans Purified total cellular DNA was digested with different restriction endonucleases including Apa I, Asp 718, Bam HI,

*Sigma

Chemical

Co., St. Louis,

MO.

77

Cla I, Dra I, Eco RI, Hind III, Pst I, Pvu II,t Bst BI, Bst Ell, Not I, Sfi I, Sty I, Xlw I and Xba Ii. Three to five µg

of DNA was digested to completion in a total volume of 100 µ according to manufacturer's recommendations. Electrophoresis was carried out on a horizontal slab gel apparatus§. DNA fragments were separated on a gel containing 0.7% agarose in TBE (0.1M Tris-Borate, pH 8.3, 2 mM EDTA) containing 0.5 µg/ml ethidium bromide. Optimal conditions were found to be 125 V for 4 hours. The gels were photographed in UV light through a Kodak Wratten no. 23 filter. Aliquots of undigested, purified DNA were electrophoresed on 0.7% agarose gels in order to examine for the presence of plasmids. RESULTS Several restriction endonucleases proved to generate acceptable banding patterns upon electrophoresis in 0.7% agarose gels. In particular, Eco RI and Hind III yielded "fingerprints" that were appropriate for the analysis of restriction enzyme patterns; i.e., resolved approximately 50 bands ranging in size from a few hundred to approximately 20,000 basepairs. The enzymes Not I, Sfi I and Xba I generated fragments that were too large to clearly separate on the gels. The restriction patterns for human A. actinomycetemcomitans from the same periodontal pocket were examined. All isolates from the same periodontal pocket and from the same patient exhibited identical patterns by restriction endonuclease analysis (Fig. 1). Furthermore, the restriction pattern appeared to be stable. Isolates obtained from the same patient over a period of several months to 2 years exhibited the same restriction endonuclease pattern as did both the rough and smooth variants of the same strain. When the distribution of the restriction endonuclease patterns according to serogroup was considered, the profile for the 20 serogroup a isolates was identical (Fig. 2) and compromised a single group-restriction pattern I. The restriction endonuclease pattern for the serogroup b and the serogroup c isolates could be separated into two closely related groups. One group—restriction pattern II—consisted of 58% of the serogroup b strains; while the other group—restriction pattern III—consisted of the remaining serogroup b strains and all 15 of the serogroup c strains. The restriction endonuclease pattern for the monkey A. actinomycetemcomitans isolates, in contrast to the human isolates, revealed considerable genetic heterogeneity. While two isolates from different animals shared the same restriction pattern, each of the remaining isolates exhibited unique patterns (Fig. 3). None of the monkey A. actinomycetemcomitans isolates exhibited a restriction pattern which was similar to that of the human isolates. Furthermore,-the re-

tBoehringcr Mannheim Biochemicals, Indianapolis, IN. rfNew England Biolabs, Beverly, MA. §Max Horizontal Submarine Agarosc Gel, Hoefer Scientific Instruments, San Francisco, CA.

J Periodontol

78

EPIDEMIOLOGY OF A. ACTINOMYCETEMCOMITANS BY DNA FINGERPRINTING

1 23456789 1011 12131415

The restriction endonuclease pattern of A. actinomycetemcomitans isolates from sites within the same patient. Purified bacterial DNA was electrophoresed on 0.7% agarose gels containing 0.5 ßg/ml ethidium bromide at 125 V for 4 hours and then photographed in ultraviolet light. Lanes 1 to 5, isolates from site 7D; lanes 6 to 10, isolates from site 30M; lanes 11 to 15, isolates from site 11D.

Figure 1:

1 23456789 1011 12 13 1415

2.B: Lane 1, FDC Y4; lane 2, SUNYaB 23; lane 3, SUNYa 54; lane 4, SUNYaB 333; lane 5, SUNYaB 373; lane 6, SUNYaB LH30D-3; lane 7, SUNYaB K5; lane 8, SUNYaB 30M-7; lane 9, SUNYaB 57; lane 10, SUNYaB 82; lane 11, Norway 1, all serotype b strains; lane 12, SUNYaB 45; lane 13, SUNYaB 67; lane 14, SUNYaB 113; lane 15, SUNYaB 360, all serotype c strains.

Figure

123456789 10

Figure 2.A: The restriction endonuclease patterns of human A. actinomycetemcomitans isolates correlated with serotype. Lane 1, SUNYaB 75 (serotype a); lane 2, FDC Y4 (serotype b); lane 3, SUNYaB 67 (serotype c); lane 4, SUNYaB 30; lane 5, SUNYaB 32; lane 6, SUNYaB 122; lane 7, SUNYaB 334; lane 8, SUNYaB 347; lane 9, SUNYaB 349; lane 10, SUNYaB 432, all serotype a strains.

February 1990

1 2 3 4 5 6 7 8 9 10 11 12

The restriction endonuclease pattern of A. actinomycetemcomitans isolates from monkeys. Lane 1, SUNYaB 75 (serotype a); lane 2, FDC Y4 (serotype b); lane 3, SUNYaB 67 (serotype c); lane 4, M1M-Aa1; lane 5, 2141-2; lane 6, 10D1T-2; lane 7, 687; lane 8, 750; lane 9, 3D2T-2; lane 10, 4CI-27; lane 11, 14E-29; lane 12, 15E1-25. All strains are serotype b.

Figure 3:

Volume 61 Number 2

striction patterns of both the human and the monkey isolates of A. actinomycetemcomitans were distinct from that seen for the closely related species H. aphrophilus (data not

shown).

Examination of the undigested DNA preparations from the human isolates failed to reveal the presence of any plasmid. Two unique, cryptic plasmids, however, were found in two A. actinomycetemcomitans isolates from the spider

monkeys. DISCUSSION As specific plaque microorganisms have been implicated in the pathogenesis of periodontal disease, there has been increasing interest in the epidemiology of periodontal infections. Specifically, attention has focused on identifying the route by which periodontal pathogens are transmitted into the human oral cavity in order to prevent both subgingival infection with these species and the subsequent development of Periodontitis. Several studies have suggested that one likely source of oral bacteria in general, and A. actinomycetemcomitans in particular, is through contact with infected family members. Microbial transmission between husbands and wives has been suggested in studies by Offenbacher et al.11 For A actinomycetemcomitans, we have

an intra-familial route of transmission12 based on data in which family members of localized juvenile Periodontitis patients were found to harbor the same biotype and serotype of A. actinomycetemcomitans. Other routes of A actinomycetemcomitans infection have also been

previously suggested

A recent study by Preus and Olsen13 suggested the transmission of A actinomycetemcomitans from a pet dog to a child. In order to track bacterial infections, one must be able to differentiate bacteria within the same species. Subspecies classifications of bacteria based on biochemical character-

proposed.

istics—biotyping—or surface antigen variation—serotyping—have been used since the early 1940s. Recently, molecular biology techniques have been used to differentiate bacteria within the same species in the analysis of microbial transmission.14^17 Total genomic DNA is extracted from cultured strains, digested with restriction endonucleases, and the resulting pattern of DNA fragments is compared by gel electrophoresis. A single difference in the restriction profile between strains is considered evidence for genetic dissimilarity; that is, evidence that the strains the same. If two strains show the same restriction when tested with several different restriction enprofile then zymes they are likely, but not necessarily, to be geA potential misinterpretation of these identical. netically results could occur if the microbial population under consideration shows limited genetic diversity, as has been found for some species.18,19 In this case, the finding of the same restriction pattern in isolates from two different sources would not necessarily indicate transmission. Data from the present study indicate that A. actinomycetemcomitans is such a species; that is, the human oral are not

ZAMBON, SUNDAY, SMUTKO

79

isolates exhibit limited genetic diversity by restriction endonuclease analysis. A actinomycetemcomitans serotype a strains exhibit a single restriction endonuclease pattern (I) distinct from the other two serotypes. This distinctive pattern may be useful as a marker for this serotype. A. actinomycetemcomitans serotype c strains exhibit a single restriction endonuclease pattern (III) shared by approximately half of the serotype b isolates while the remaining serotype b strains exhibit an additional distinct restriction endonuclease pattern (II). DNA restriction patterns in other species have also been reported to cross serogroup boundaries. For example, a similar restriction endonuclease profile is shared by a cluster oîLegionella pneumophila strains from all the reported serogroups of that species.20 In addition to the restriction endonuclease analysis patterns, serotype a has been previously differentiated from serotypes b and c by biochemical tests. A. actinomycetemcomitans serotype a strains do not ferment xylose while all serotype b and some serotype c strains ferment this sugar.5 Thus, it appears that serotype a A. actinomycetemcomitans is genotypically, serologically, and phenotypically distinct. There is also recent evidence to suggest that these different groups of A actinomycetemcomitans may be associated with clinically distinct forms of Periodontitis. Serotype a (restriction pattern I) A actinomycetemcomitans has been associated with Actinobacillary Periodontitis in adults (S.S. Socransky, written communication) while serotype b is associated with Actinobacillary Periodontitis in juveniles.5 The limited genetic diversity of A. actinomycetemcomitans seen by restriction endonuclease analysis is in contrast to the significant genetic heterogeneity seen in another periodontal pathogen, Bacteroides gingivalis. For this species, different restriction endonuclease patterns are obtained for isolates from different individuals even within the same family.21 However, B. gingivalis isolated within the same site and within the same individual have the same restriction profile, suggesting that B. gingivalis infection constitutes a clonai infection; that is, an infection derived from a single microorganism. Similarly, the apparent genetic homogeneity of strains isolated from individual subjects and sites within subjects is consistent with a clonai origin of oral A actinomycetemcomitans infections. In all the observed instances, isolates from the same site and isolates from different sites in the same patient exhibited an identical restriction endonuclease pattern. This is consistent with previous studies which find that A actinomycetemcomitans isolated from the same periodontal pocket and from the same patient are of the same biotype4 and serotype.5-12 The restriction endonuclease pattern of A actinomycetemcomitans is an apparently stable feature with regard to both time and colony morphology. Isolates obtained from the same patient over a period of several months to a year and rough and smooth variants of the same strain show the same restriction endonuclease pattern. The stability of restriction endonuclease patterns has been reported in other species. For example, Corynebacterium species isolated over

J Periodontol 80

EPIDEMIOLOGY OF A. ACTINOMYCETEMCOMITANS BY DNA FINGERPRINTING

9 month period demonstrated the same restriction endonuclease pattern22 and a single isolate of N. meningitidis was unchanged over 29 in vitro passages.7 The diversity of restriction endonuclease patterns observed in monkey isolates of A actinomycetemcomitans and the distinction between the strains infecting human and those infecting non-human primates raises some interesting questions. Even from different areas of the world, A actinomycetemcomitans isolates have similar restriction endonuclease profiles suggesting limited genetic diversity among the isolates infecting human subgingival dental plaque. By contrast, monkey A actinomycetemcomitans isolates are apparently derived from a wide variety of A actinomycetemcomitans strains. Thus, it appears that humans are infected with a subset of the many different types of A actinomycetemcomitans which exist in nature. This is similar to Legionella pneumophila in which only a few of the many different genotypes existing in nature actually cause Legionnaire's Disease.23 The limited diversity of the human A actinomycetemcomitans isolates may be the result of selective pressures found in the human oral cavity. Consequently, human A actinomycetemcomitans isolates may express certain virulence factors that are absent from the monkey isolates. For example, human A actinomycetemcomitans isolates may preferentially colonize human epithelial cells or produce leukotoxin effective against human polymorphonuclear leukocytes. Alternatively, the monkey A actinomycetemcomitans isolates may either not produce these same virulence factors or monkey polymorphonuclear leukocytes may not be susceptible. Polymorphonuclear leukocytes from cebus and African monkeys are known to be resistant to the A actinomycetemcomitans leukotoxin.24 In summary, restriction endonuclease analysis of human oral isolates of A actinomycetemcomitans revealed three major genotypes in contrast to analysis of monkey isolates which demonstrated a large number of genotypes. The data suggest that the human oral cavity is clonally infected with a small number of the many different A actinomycetemcomitans genotypes which exist in nature. Determination of the factors involved in the selective colonization of the human oral cavity by this subset of A actinomycetemcomitans genotypes may be useful in the prevention Actinoa

bacillary Periodontitis.

itans in human

periodontal disease: A cross-sectional microbiological investigation. Infect Immun 1980; 29:1013. 5. Zambón JJ, Slots J, Genco RJ. Serology of oral Actinobacillus actinomycetemcomitans and serotype distribution in human periodontal disease. Infect Immun 1983; 41:19. 6. Tompkins LS, Troup NJ, Woods T, et al. Molecular epidemiology of Legionella species by restriction endonuclease and allocnzyme analysis. J Clin Microbiol 1987; 25:1875. 7. Bjorvatn B, Lund V, Kristiansen B, et al. Applications of restriction endonuclease fingerprinting of chromosomal DNA of Neisseria meningitidis. J Clin Microbiol 1984; 19:763. 8. Devlin HR, Au W, Foux L, et al. Restriction endonuclease analysis of nosocomial isolates of Clostridium difficile. J Clin Microbiol 1987; 9. 10.

11.

12.

13.

14.

15.

16.

17.

18. 19.

20.

21.

Acknowledgements USPHS Grant Number DE 07497 and DE Dr. Zambón is the recipient of a Research Career 04898; Award from the National Institute of Dental Development Research.

Supported by

REFERENCES 1. Zambón JJ. Actinobacillus actinomycetemcomitans in human periodontal disease. / Clin Periodontol 1985; 12:1. 2. King EO, Tatum HW. Actinobacillus actinomycetemcomitans and Haemophilus aphrophilus. J Infect Dis 1962; 111:85. 3. Pulverer G, Ko HL. Actinobacillus actinomycetemcomitans: Fermentative capabilities of 140 strains. Appi Microbiol 1970; 20:693. 4. Slots J, Reynolds HS, Genco RL Actinobacillus actinomycetemcom-

February 1990

22.

23. 24.

25:216. Slots, J. Salient biochemical characters of Actinobacillus actinomycetemcomitans. Arch Microbiol 1982; 131:60. Silhavy TJ, Bcrman ML, Enquist LW. Experiments with Gene Fusions. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1984; 137-139. Offenbacher S, Olsvik , Tonder A. The similarity of periodontal microorganisms between husband and wife cohabitants: Association or transmission? / Periodontol 1985; 56:317. Zambón JJ, Christersson LA, Slots J. Actinobacillus actinomycetemcomitans in human periodontal disease: Prevalence in patient groups and distribution of biotypes and serotypes within families. J Periodontol 1983; 54:707. Preus HR, Olsen I. Possible transmittance of A. actinomycetemcomitans from a dog to a child with rapidly destructive Periodontitis. J Clin Periodontol 1988; 23:68. Kaper JB, Bradford HB, Roberts NC, Falkow S. Molecular epidemiology of Vibrio cholerae in the U.S. Gulf Coast. / Clin Microbiol 1982; 16:129. Scherer S, Stevens DA. Application of DNA typing methods to epidemiology and taxonomy of Candida species. / Clin Microbiol 1987; 25:675. Kuijper EJ, Oudbier JH, Stuifbergcn WN, et al. Application of wholecell DNA restriction endonuclease profiles to the epidemiology of Clostridium difficile-induced diarrhea. / Clin Microbiol 25 : 751, 1987; 25:751. Langenberg W, Rauws EAJ, Widjojokusumo A, et al. Identification of Campylobacterpyloridis isolated by restriction endonuclease DNA analysis. J Clin Microbiol 1986; 24:414. Allan I, Loeb MR, Moxon ER. Limited genetic diversity of Haemophilus influenzae (type b). Micro Pathogen 1987; 2:139. Musser JM, Granoff DM, Pattison PE, Seiander RK. A population genetic framework for the study of invasive diseases caused by serotypc b strains of Haemophilus influenzae. Proc Nati Acad Sci USA 1985; 82:5078. van Ketel RJ. Similar DNA restriction endonuclease profiles in strains of Legionella pneumophila from different serogroups. / Clin Microbiol 1988; 26:1838. Loos BG, Herweijer JA, Shlossman M, et al. Epidemiology of blackpigmented Bacteroides using DNA restriction fragment pattern analyses. J Dent Res 1988; 67(Spcc Issue): 369(Abstr. 2049). Khabbaz RF, Kaper JB, Moody MR, et al. Molecular epidemiology of Group JK Corynebacterium on a cancer ward: lack of evidence for patient-to-patient transmission. J Infect Dis 1986; 154:95. Sclandcr RK, McKinncy RM, Whittam TS, et al. Genetic structure of populations of Legionella pneumophila. J Bacteriol 1985; 163:1021. Taichman N, Sakurada S, DiRienzo J, Simpson D. Comparative studies on the biology of Actinobacillus actinomycetemcomitans leukotoxin (Aa-LTX) in primates. / Dent Res 1986; 68 (Spec Issue) 852

(Abstr. 1138). Send reprint requests to Dr. Joseph Zambón, University of New York, Buffalo, NY 14214. Accepted for publication August 1, 1989.

222 Foster

Hall, State