0099-2399/90/1607-0318/$02.00/0 JOURNAL OF ENDODONTICS Copyright 9 1990 by The American Association of Endodontists
Printed in U.S.A.
VOL. 16, No. 7, JULY 1990
Indirect Immunofluorescence Microscopy for the Identification of Actinomyces sp. in Endodontic Disease Robert J. Gohean, DMD, Eugene A. Pantera, Jr., DDS, MS, and George S. Schuster, DDS, PhD
tifying Actinomyces sp. in direct clinical smears or in biopsy sections. Hotchi and Schwarz (3) demonstrated Actinomyces israelii and Actinomyces naesludii in formalin-fixed tissues. They used their technique on paraffin sections with conventional staining methods in the diagnosis of cervico-facial actinomycosis. The fluorescing antibody technique was found to have many advantages compared with conventional histopathology; the most important being identification in mixed infections (2, 4). Happonen et al. (5) showed that immunochemical methods can be used for the diagnosis of Actinomyces sp. in periapical actinomycosis at the species level, again demonstrating the specificity of immunospecific techniques in tissue sections. Although immunofluorescence is successfully used for the identification ofActinomyces sp. in histological tissue sections, there are no studies reporting the use of indirect immunofluorescence (IF) for the identification of the actinomycetes in canal samples of endodontically diseased human teeth. The purpose of this investigation was to use indirect immunofluorescence microscopy to determine the presence of select Actinomyces sp. in a survey of teeth associated with endodontic disease.
Indirect immunofluorescence microscopy was used to determine the presence of select Actinomyces sp. in a clinical survey of teeth with endodontic disease. Thirty canal samples were tested for the presence of Actinomyces sp.: A. israelii, A. odontolyticus, A. viscosus serotype I, and A. viscosus serotype II. Actinomyces sp. were identified in 18 (60%) samples. A. israelii was most frequently identified with immunofluorescence in 16 (53%) cases examined and was the test organism most often associated with endodontic disease within the restrictions of this study. Positive and negative controls were appropriate to support findings. Indirect immunofluorescence microscopy can be used for the clinical identification of Actinomyces sp. in endodontic disease without the necessity of performing a parallel culture study.
Actinomycetes are Gram-positive, non-acid-fast, anaerobic, or microaerophilic filamentous bacteria. These organisms are common saprophytes of the oral cavity and are thought to be of low virulence. Actinomyces sp. are recovered from such sites as tonsillar crypts, dental plaque, carious dentin, gingival crevices, periodontal pockets, and infected dental root canals. The actinomycetes are most often associated with periapical actinomycoses rather than as participants or initiators of pulpal disease. There also appears to be a correlation between endodontic treatment and periapical actinomycosis. It may be possible that endodontic procedures introduce the actinomycetes into the periapical tissues. Secondarily, the organisms may enter through a root canal left open to oral fluids or they may appear as a direct extension of the carious process (t). Alternatively, this may only be an observational anomaly since biopsies of periapical lesions are not routinely performed unless there is nonresolution of a lesion. Since the presence of the actinomycetes is usually based on histopathology and not speciation, more specific means need to be used as many organisms form morphologically similar granules, rendering histologic diagnosis uncertain (2). Specific fluorescent antiserum is the most sensitive method for iden-
MATERIALS AND METHODS
Sample Collection Thirty samples were collected from patients who presented for treatment in the Endodontic Clinic at the Medical College of Georgia, School of Dentistry. Need for endodontic therapy was based on signs and symptoms suggestive of endodontic disease or elective needs. Patients who had received antibiotics within the previous 60 days were excluded from the study. The following data were recorded for each patient: name, an identifying number, tooth involved, pulpal diagnosis (PuDx) after access, periapical diagnosis (PaDx), sensitivity to percussion, tenderness on periapical palpation, presence of periapical radiolucency (PARL), presence of pain, and presence of swelling archetypal of endodontic disease. Samples of canal material were collected in a manner similar to that described by Pantera et al. (6) Following isolation with a rubber dam, a 1-min 5.25% NaOCI scrub was used to disinfect the tooth surface. A sterile
318
Vol. 16, No. 7, July 1990
bur in a high-speed handpiece was used for access preparation. Peripheral caries, if present, were removed before pulpal access, when a new sterile bur was exchanged and pulp chamber penetration achieved. Operator diagnosis of contents within the chamber and existing pulpal condition were recorded. Using sterile surgical gloves and presterilized and packaged endodontic Flex-R files (Union Broach, New York, N-Y), canal contents were sampled. A balanced force technique was used short of the apex with up to three instruments for each canal. The sampling instruments were dropped into screw-topped tubes containing prereduced anaerobic sterilized Ringer's solution. This suspension was transported to the laboratory within 2 h and processed for serodiagnostic examination. The bacteria were dispersed using a vortex mixer for 60 s, distributed to 30 glass slides in two 10-ul aliquots and gently heat fixed. The samples were then stored at -10"C until needed. Use of human material as reported in this study met all criteria as established by the Human Assurance Committee of the Medical College of Georgia.
Production of Antisera Rabbit antisera to Actinomyces odontolyticus (ATCC 17929), A. israelii (ATCC 12102), Actinomyces viscosus serotype I (ATCC 19246), and Actinomyces viscosus serotype II (ATCC 15987) were prepared in a manner similar to that described by Pantera et al.(6). Pure cultures were grown in Todd Hewitt broth and maintained on blood agar supplemented with hemin and Vitamin KI. After incubation for 5 days under an atmosphere of 85% N2, 10% H2, and 5% CO2 at 37~ the cells were pelleted and washed three times in phosphate-buffered saline solution (pH 7.2) and suspended to a concentration of 10 mg/ml (wet weight) in sterile saline. Polyclonal antisera was produced in New Zealand White rabbits by intravenous administration, via the marginal ear vein, of three 1.5-ml aliquots of whole cells the first week, nine 1.0-ml aliquots over the next 3 wk, and three 1.5-ml aliquots through the final week. Following trial bleeding and examination for a satisfactory antibody titer, the rabbits were exsanguinated by cardiac puncture under general anesthesia. Whole blood was collected and allowed to coagulate for 24 h at 4"C. Rabbit sera (supernatant) were collected and stored in 2-ml aliquots at -10~ until needed.
Immune Absorption New ATCC cultures for the target species were raised as above. Stock control slides were prepared in the same manner as the in vivo samples and stored at -10*C. After three phosphate-buffered saline washes (pH 7.2), cells were pelleted and weighed. Aliquots of each antiserum were thawed and incubated at 56"C for I h to inactivat e complement. Whole cells (100 mg wet wt) were added to 1 ml of dissimilar rabbit antiserum individually agitated every 15 min for 60 min at 37"C, then centrifuged at 12,000 x g for 5 min. The antiserum supernatant was removed, and the absorption completed for each antisera-cell combination. The entire cycle was repeated and the absorbed immune antisera were stored at -10*C.
Identification of Actinomyces sp.
319
Indirect Immunofluorescence The rabbit antiserum dilutions for the selected Actinomyces sp. were determined by "checkerboard" dilutions reacted with goat anti-rabbit antisera conjugated to fluorescein isothiocyanate. At room temperature, 15 ul of rabbit antiserum were placed on the positive control bacterial smear (e.g.A. israelii antiserum was placed on A. israelii whole cells), incubated for 20 min at 37~ washed with phosphate-buffered saline containing 0.05% Tween 20 twice for 10 rain each, and then rinsed with distilled water. The slides were incubated with 15 ul of goat anti-rabbit immunoglobulin G conjugated to fluorescein isothiocyanate, washed, and rinsed as before and covered with 0.10% glycerol in phosphate-buffered saline (pH 9.0). To test for cross-reaction, positive control smears of Actinomyces sp. were reacted against the absorbed Actinomyces sp. antisera (e.g.A. israelii antisera reacted with A. odontolyticus cells). Combinations of target organisms and serial 2-fold dilutions of antisera and conjugate (1:32 to 1:2096) were evaluated. The technique was optimum at an antiserum dilution of 1:512, and a conjugate dilution of 1:512 when reacted against reference strains. Cross-reactivity between subspecies, while present at this dilution, was distinguishable from a true positive reaction. Stock cultures of
Fusobacterium nucleatum, Haemophilus actinomycetemycornitans, Bacteroides intermedius, and Capnocytophaga ochracea were reacted with each antiserum and used as negative controls. The specimens were examined at xl000 using a Nikon Optiphot microscope (Nikon Inc., Garden City, NY) equipped for phase contrast illumination and incident light fluorescence. Each specimen was examined for the presence of fluorescing cells. Fluorescence was graded from 1+ to 4+. Only grade 4+ (brilliant fluorescence, brightly defined cell wall with center of cell completely dark) was considered a serologically positive reaction.
Immunofluorescence Microscopy The clinical samples were prepared following the protocol outlined above. A minimum of 60 fields of each smear was examined for each sample. A sample was considered positive for the presence of Actinomyces sp. if it exhibited l0 or more strongly fluorescent cell forms and 4+ fluorescence. Crossreactivity was also noted by less numerous weaker fluorescing cells. Positive and negative control reference strains were included for all IF trials.
Statistical Analysis Statistical measurements were performed with X 2 analysis of cross-tabulated contingency tables using Statgraphics (Ver 2.1; STSC, Inc., RockviUe, MD). The absence, presence, or suspected cross-reactivity for each Actinomyces sp. for all samples was cross-tabulated against each Actinomyces sp., the presence or absence of any Actinomyces sp., and against clinical parameters. The clinical parameters analyzed were: PuDx, PaDx, and presence of PARL. For the purposes of statistical analysis, the clinical parameters were collapsed and
320
Journal of Endodontics
Gohean et al.
recoded so that each parameter was either normal or not normal. For example, five PaDx were recorded for the study. These were chronic alveolar abscess, acute alveolar abscess, Phoenix abscess, acute apical periodontitis, and normal. This clinical parameter was recoded as not normal or normal. For each statistical test, the null hypothesis (1to) stated that groups analyzed were independent. The alternative hypothesis (/-/.4) stated that groups analyzed were not independent. The test for significance was set at a = 0.05.
TABLE 3, Immunofluorescence-positive findings for non-normal clinical diagnosis
AI* PuDx (n = 27) PaDx (n = 26) PARL (n = 18)
AO
AV1
AV2
ALL
0 0 0
4(14.8) 4(15.4) 0
18(66.7) 17(65.4) 13(72.2)
16(59.3)1- 3(11.1) 15(57.7) 3(11.5) 12(61.6) 1(5.5)
* AI, A. israelii; AO, A. odontolyticus; AV1, A. viscosus serotype I; AV2, A. viscosus serotype II. 1" Frequency (%).
RESULTS DISCUSSION Of 30 endodontic cases; PuDx included 17 cases necrotic, 6 partially necrotic, 3 retreatments, and 4 vital or normal teeth; PaDx included 8 cases with chronic alveolar abscess, 3 with acute alveolar abscess, 7 with Phoenix abscesses, 7 with acute apical periodontitis, and 5 normal. Diagnostic findings were collapsed to represent 27 (90%) PuDx that were not normal and 26 (87%) PaDx that were not normal (Table 1). Cases determined to have P A R L radiolucencies were 18 (60%). Data for IF findings in clinical samples are shown in Table 2. Of 30 clinical canal samples, Actinomyces sp. were identified in 18 (60%) samples. Of the total samples, 16 (53%) were positive to A. israelii antiserum, 3 (10%) positive to A. odontolyticus antiserum, 0 positive to A. viscosus serotype I antiserum, and 4 (13%) positive to A. viscosus serotype II antiserum. Cross-reactivity among Actinomyces sp. was observed in 18 (60%) of the samples. Reactions against like stock strains of target Actinomyces sp. were positive for each trial. Immunofluorescence responses against negative control organisms were not cross-reactive. Table 3 lists collapsed clinical findings that were IF positive for Actinomyces sp. The incidence of Actinomyces sp. in the presence of non-normal PuDx was 66.7%. Of teeth with a non-normal PaDx, Actinomyces sp. were identified in 65.4%. In cases with PARL, Actinomyces sp. were identified in 72.2%. The incidence of A. israelii in the presence of non-normal PuDx was 59.3%. Of teeth with a non-normal PaDx, A. israelii were identified in 57.7%. In cases with PARL, A. israelii were identified in 66.6%. The other target species are tabulated in Table 3 as well. Table 4 summarizes statistical correlations among and between target organisms and the selected clinical parameters. TABLE 1. Collapsed clinical findings (non-normal; n = 30)
PuDx
PaDx
PARL
27(90)*
26(87)
18(60)
9 Frequency (%).
TABLE 2. Results of indirect immunofluorescence for Actinomyces sp. (n = 30)
IF+l" IFIF+l-
AI*
AO
AV1
AV2
ALL
16(53):1: 12(40) 2(7)
3(10) 17(57) 10(33)
0(0) 19(63) 11(37)
4(13) 15(50) 11 (37)
18(60) 12(40) --
9 AI, A. israelii; AO, A. odontolyticus; AM1, A. viscosus serotype I; AV2, A. viscosus serotype II; ALL, all organisms. 1" Immunofluorescence negative (-), positive (+), cross-reaction (+/-). ~: Frequency (%).
Our results suggest that the presence of select Actinomyces sp. from endodontic canal samples can be identified in greater incidence through the use of IF than reported in previous culture studies. Culture recovery ofActinomyces sp. from clinical material is difficult. Identification by morphological and biochemical criteria is also difficult and time consuming. Actinomyces israelii is the principal species implicated in periapical actinomycosis lesions; nonetheless, Actinomyces sp. are generally found in mixed infections (7) and are not readily isolated through bacteriologic culturing. Identification of Actinomyces sp. by biochemical tests requires the preparation of special culture media and is further complicated by the broad range of strain variation within the species (4). Successful isolation in all clinical samples requires culturing multiple samples of purulent material in enriched media, under anaerobic conditions (optimally), and allowing adequate time (5 to 10 days) for growth. Actinomyces sp. may be mistaken for other organisms or they may remain unidentified if branching is infrequent or absent. While examining head and neck infections, Richtsmeier and Johns (8) reported only 25% positive cultures in known histological cases of actinomycosis. Of the 67 cases in which culture was attempted by Brown (9), the organism was isolated from only 16 samples. When reviewing culture studies and case reports, there is a high incidence with no reported growth of actinomycetes either in pure culture or with mixed flora (9). Wesley et al. (10) reported a case where Actinomyces sp. could not be recovered either before or after actinomycosis was characterized in biopsy. A review of relevant studies of endodontic pathogens (Table 5) shows that culture recovery of Actinomyces sp. is infrequent. These observations typify the inherent difficulties with culture studies of endodontic pathogens. With appropriate preparation, IF can be used successfully for surveys of canal contents without the need for culture recovery techniques, as is done for other clinical problems. Hemophilus influenzae in meningitis and Legionella species in Legionnaire's disease are routinely detected through immunofluorescence techniques. Similarly, IF has been used successfully in the identification of periodontal pathogens such as H. actinomycetemcomitans, B. intermedius, Bacteroides gingivalis, and Bacteroiodes melaninogenicus. Endodontic pathogens such as B. intermedius, B. gingivalis, and Bacteroides endodontalis have been identified with IF as well (6). These studies all have clearly demonstrated that IF is both more sensitive than culture for identification of target organisms and more rapid.
Identification of Actinomyces sp.
Vol. 16, No. 7, July 1990
The findings of the present study indicate cross-reactivity occurring among the target Actinomyces sp. in clinical samples (Table 2). In a review of their research and that of others, Gerencser and Slack (4) discussed the high magnitude of crossserologic reactivity among species of Actinomyces. They developed species-specific antisera using a direct immunofiuorescence technique. Bowden et al. ( 11 ) evaluated Actinomyces cell wall antigens. They reported that neutral cell wall carbohydrate and polypeptide containing charged antigens are c o r n -
ponents of the Actinomyces antigenic structure. They further concluded that cell wall carbohydrate can be responsible for species-specific reaction and cross-reactions, and strains may possess more than one cell wall carbohydrate determinant. Firtel and Fillery (12) raised monoclonal antibodies against 26 strains of A. viscosus and A. naeslundii. They reported extensive cross-reactions and only had three monoclonal antibodies specific for one strain each. Within the parameters of the present study, the strict criterion of a 4+ reaction was
TABLE 4. Summary of statistical analysis for cross-tabulations AI
AO
AI
x2 df Significance
17.066 2 1.969E-4
ALL*
--
--
AO
x2 df Significance
3.60312 0.165
12.827 4 0.012
--
AV1
x2 df Significance
AV2
x2 df Significance
PuDx
x2 df Significance
2.60811 Yates correction 0.106
5.926 2 0.051
1.76512 0.414
XXr
0.6401" 2 0.726
PaDx
x2 df Significance
0.97411 Yates correction 0.324
2.4161" 2 0.299
0.83212 0.660
0.00011 Yates correction 1.000
3.83812 0.147
PARL
x2 df Signifcance
1.67211 Yates correction 0.196
1.09412 0.579
0.82512 0.662
0.1561" 1 Yates correction 0.692
1.36412 0.506
1.40611 Yates correction 0.236 9.393 2 9.595E-6
2.15612 0.340 11.617 4 0.020
AV1
AV2
5.80212 0.079 12.104 4 0.017
9.845 2 7.279-3
* ALL, all organisms; AI, A. israelii; AO, A. odontolyticus; AV1, A. viscosus serotype I; AV2, A. viscosus serotype II. 1"Reject null hypothesis (Ho). XX, Sample too smalt for valid comparison.
TABLE 5. Comparative results of selected studies Culture +
Population
(Actinomyces sp.)
Biopsy
Wittgow and Sabiston (13)
40 cases 32 positive cultures
1/40 (0.03%)
ND*
Weir and Buck (1) (literature review)
20 cases 10 cultured
2/10 (20%)
Sabiston et al. (14)
65 cases 58 positive cultures
3/65 (0.05%) 3/58 (0.05%)
Happonen and Viander (2)
10 cases 5 cultured
1/5 (20%)
Study
20/20 (100%) ND 5/5 (100%)
Sundqvist (15)
18 cases
5/18 (28%)
ND
Kantz and Henry (16)
24 cases 16 positive cultures
5/24 (21%) 5/16 (31%)
ND
Oguntebi et al. (17)
10 cases
3/10 (30%)
ND
Williams et al. (18)
10 cases
0
ND
Matusow et al. (19)
62 cases
0
ND
Yoshida et al. (20)
36 cases 31 positive cultures
7/36 (19%) 7/31 (23%)
ND
* ND, not determined.
321
322
Journal of Endodontics
Gohean et al.
designated as a positive reaction. It may be that the two cycles of immune absorption were insufficient or that absorption should include another species or strain not included in this study. While it may be attractive to make clinical conclusions based on our statistical findings (Table 4), we are reluctant to do so. The purpose of this study was to use IF to determine the presence ofActinomyces sp. in clinical endodontic canal samples. However, despite the size of the sample population, some observations can be made. As determined by IF, there does seem to be a relationship between Actinornyces sp. and some forms of endodontic disease. This supports both the intuitive clinical observation as well as what has been reported in controlled studies (Table 5). Of the selected target species, A. israelii was most frequently identified with IF and most often associated with endodontic disease within the parameters of this study. Again, this supports culture findings that report A. israelii as the Actinomyces sp. most commonly associated with human disease, including endodontic disease. Conversely, the other target Actinomyces sp. (A. odontolyticus, A. viscosus serotype I, and A. viscosus serotype II) were not frequently associated with our sample population. This finding also supports culture studies of endodontic disease. Our findings indicate that Actinomyces sp. may be participants or major protagonists in microbial climax communities that establish certain endodontic diseases. CONCLUSION Indirect immunofluorescence microscopy can be used for the clinical identification of Actinomyces sp. in endodontic disease without the necessity of performing parallel culture surveys. This research was supported in part by the American Association of Dental Schools/Warner Lambert Clinical Research Fellowship award~,d to E. A. P. and also by a student research support grant from the Endowment and Memorial Foundation of the American Association of Endodontists awarded to R. J. G. The opinions, assertions, materials, and methodologies herein are private ones of the authors and are not to be construed as official or reflecting the views of the American Association of Endodontists. Dr. Gohean is in private practice in Columbia, SC. Dr. Pantera is assistant professor, Department of Endodontics, School of Dentistry, Medical College of
Georgia, Augusta, GA. Dr. Schuster is lone and Arthur Merritt Professor, Department of Oral Biology/Microbiology, School of Dentistry, and associate professor, Cell and Molecular Biology, School of Medicine, Medical College of Georgia.
References 1. Weir JC, Buck WH. Periapical actinomycosis. Oral Surg 1982;54:33640. 2. Happonen R-P, Viander M. Comparison of fluorescent antibody technique and conventional staining methods in diagnosis of cervico-facial actinomycosis. J Oral Patho11982;11:417-25. 3. Hotchi M, Schwarz J. Characterization of actinomycotic granules by architecture and staining methods. Arch Patho11972;93:392-400. 4. Gerencser MA, Slack JM. Serologic identification of Actinomyces using fluorescent antibody techniques. J Dent Res 1976;55(spacial issue A):A18491. 5. Happonen R-P, Soderling E, Viander M, Linko-Kettunen L, Pelliniemi LJ. Immunocytochemical demonstration of Actinomyces species and Arachnia propionica in periapical infections. J Oral Pathol 1985;14:405-13. 6. Pantere EA Jr, Zamben J J, Shih-Levine M. Indirect immunofluorescence for the detection of Bacteroides species in human dental pulp. J Endodon 1988; 14:218-23. 7. Finegold SM. Anaerobic bacteria in human disease. New York: Academic Press, 1977. 8. Richtsmeier WJ, Johns ME. Actinomycosis of the head and neck. CRC Crit Rev Clin Lab Sci 1979;11:175-202. 9. Brown JR. Human actinomycosis, a study of 181 subjects. Hum Pathol 1973;4:319-30. 10. Wesley RK, Osbern TP, Dytewski JJ. Periapical actinomycosis: clinical considerations. J Endodon 1977;3:352-5. 11. Bowden GH, Hardie JM, Fillery ED. Antigens from Actinomyces species and their value in identification. J Dent Res 1976;55(special issue A):A192204. 12. Firtel M, Fillery ED. Distribution of antigenic determinants between Actinomyces viscosus and Actinomyces naeslundii. J Dent Res 1988;67:1520. 13. Wittgow WC, Sabiston CB. Microorganisms from pulpal chambers of intact teeth with necrotic pulps. J Endodon 1975;5:168-71. 14. Sabiston CB Jr, Gringsby WR, Segerstrom N. Bacterial study of pyogenic infections of dental origin. Oral Surg 1976;41:430-5. 15. Sundqvist G. Bacteriological studies of necrotic dental pulps [Dissertation 7]. Umea, Sweden: University of Umea, 1976. 16. Kantz WE, Henry CA. Isolation and classification of anaerobic bacteria from intact pulp chambers of non-vital teeth in man. Arch Oral Bio11974;19:916. 17. Oguntebi B, Slee AM, Tanzer JM, Langeland K. Predominant microflora associated with human dental periapical abscesses. J Clin Microbiol 1982;15:964-6. 18. Williams BL, McCann GF, Schoenknecht FD. Bacteriology of dental abscesses of endodontic origin. J Clin Microbiol 1983;18:770-4. 19. Matusow RJ, Goodall LB. Anaerobic isolates in primary pulpal-alveolar cellulitis cases: endodontic resolutions and drug therapy considerations. J Endodon 1983;9:535-43. 20. Yoshida M, Fukushima H, Yamamoto K, Ogawa K, Toda T, Sagawa H. Correlation between clinical symptoms and microorganisms isolated from root canals of teeth with periapical pathosis. J Endodon 1987;13:24-8.
Printed in U.S.A.
VOL. 16, No. 7, JULY 1990
Indirect Immunofluorescence Microscopy for the Identification of Actinomyces sp. in Endodontic Disease Robert J. Gohean, DMD, Eugene A. Pantera, Jr., DDS, MS, and George S. Schuster, DDS, PhD
tifying Actinomyces sp. in direct clinical smears or in biopsy sections. Hotchi and Schwarz (3) demonstrated Actinomyces israelii and Actinomyces naesludii in formalin-fixed tissues. They used their technique on paraffin sections with conventional staining methods in the diagnosis of cervico-facial actinomycosis. The fluorescing antibody technique was found to have many advantages compared with conventional histopathology; the most important being identification in mixed infections (2, 4). Happonen et al. (5) showed that immunochemical methods can be used for the diagnosis of Actinomyces sp. in periapical actinomycosis at the species level, again demonstrating the specificity of immunospecific techniques in tissue sections. Although immunofluorescence is successfully used for the identification ofActinomyces sp. in histological tissue sections, there are no studies reporting the use of indirect immunofluorescence (IF) for the identification of the actinomycetes in canal samples of endodontically diseased human teeth. The purpose of this investigation was to use indirect immunofluorescence microscopy to determine the presence of select Actinomyces sp. in a survey of teeth associated with endodontic disease.
Indirect immunofluorescence microscopy was used to determine the presence of select Actinomyces sp. in a clinical survey of teeth with endodontic disease. Thirty canal samples were tested for the presence of Actinomyces sp.: A. israelii, A. odontolyticus, A. viscosus serotype I, and A. viscosus serotype II. Actinomyces sp. were identified in 18 (60%) samples. A. israelii was most frequently identified with immunofluorescence in 16 (53%) cases examined and was the test organism most often associated with endodontic disease within the restrictions of this study. Positive and negative controls were appropriate to support findings. Indirect immunofluorescence microscopy can be used for the clinical identification of Actinomyces sp. in endodontic disease without the necessity of performing a parallel culture study.
Actinomycetes are Gram-positive, non-acid-fast, anaerobic, or microaerophilic filamentous bacteria. These organisms are common saprophytes of the oral cavity and are thought to be of low virulence. Actinomyces sp. are recovered from such sites as tonsillar crypts, dental plaque, carious dentin, gingival crevices, periodontal pockets, and infected dental root canals. The actinomycetes are most often associated with periapical actinomycoses rather than as participants or initiators of pulpal disease. There also appears to be a correlation between endodontic treatment and periapical actinomycosis. It may be possible that endodontic procedures introduce the actinomycetes into the periapical tissues. Secondarily, the organisms may enter through a root canal left open to oral fluids or they may appear as a direct extension of the carious process (t). Alternatively, this may only be an observational anomaly since biopsies of periapical lesions are not routinely performed unless there is nonresolution of a lesion. Since the presence of the actinomycetes is usually based on histopathology and not speciation, more specific means need to be used as many organisms form morphologically similar granules, rendering histologic diagnosis uncertain (2). Specific fluorescent antiserum is the most sensitive method for iden-
MATERIALS AND METHODS
Sample Collection Thirty samples were collected from patients who presented for treatment in the Endodontic Clinic at the Medical College of Georgia, School of Dentistry. Need for endodontic therapy was based on signs and symptoms suggestive of endodontic disease or elective needs. Patients who had received antibiotics within the previous 60 days were excluded from the study. The following data were recorded for each patient: name, an identifying number, tooth involved, pulpal diagnosis (PuDx) after access, periapical diagnosis (PaDx), sensitivity to percussion, tenderness on periapical palpation, presence of periapical radiolucency (PARL), presence of pain, and presence of swelling archetypal of endodontic disease. Samples of canal material were collected in a manner similar to that described by Pantera et al. (6) Following isolation with a rubber dam, a 1-min 5.25% NaOCI scrub was used to disinfect the tooth surface. A sterile
318
Vol. 16, No. 7, July 1990
bur in a high-speed handpiece was used for access preparation. Peripheral caries, if present, were removed before pulpal access, when a new sterile bur was exchanged and pulp chamber penetration achieved. Operator diagnosis of contents within the chamber and existing pulpal condition were recorded. Using sterile surgical gloves and presterilized and packaged endodontic Flex-R files (Union Broach, New York, N-Y), canal contents were sampled. A balanced force technique was used short of the apex with up to three instruments for each canal. The sampling instruments were dropped into screw-topped tubes containing prereduced anaerobic sterilized Ringer's solution. This suspension was transported to the laboratory within 2 h and processed for serodiagnostic examination. The bacteria were dispersed using a vortex mixer for 60 s, distributed to 30 glass slides in two 10-ul aliquots and gently heat fixed. The samples were then stored at -10"C until needed. Use of human material as reported in this study met all criteria as established by the Human Assurance Committee of the Medical College of Georgia.
Production of Antisera Rabbit antisera to Actinomyces odontolyticus (ATCC 17929), A. israelii (ATCC 12102), Actinomyces viscosus serotype I (ATCC 19246), and Actinomyces viscosus serotype II (ATCC 15987) were prepared in a manner similar to that described by Pantera et al.(6). Pure cultures were grown in Todd Hewitt broth and maintained on blood agar supplemented with hemin and Vitamin KI. After incubation for 5 days under an atmosphere of 85% N2, 10% H2, and 5% CO2 at 37~ the cells were pelleted and washed three times in phosphate-buffered saline solution (pH 7.2) and suspended to a concentration of 10 mg/ml (wet weight) in sterile saline. Polyclonal antisera was produced in New Zealand White rabbits by intravenous administration, via the marginal ear vein, of three 1.5-ml aliquots of whole cells the first week, nine 1.0-ml aliquots over the next 3 wk, and three 1.5-ml aliquots through the final week. Following trial bleeding and examination for a satisfactory antibody titer, the rabbits were exsanguinated by cardiac puncture under general anesthesia. Whole blood was collected and allowed to coagulate for 24 h at 4"C. Rabbit sera (supernatant) were collected and stored in 2-ml aliquots at -10~ until needed.
Immune Absorption New ATCC cultures for the target species were raised as above. Stock control slides were prepared in the same manner as the in vivo samples and stored at -10*C. After three phosphate-buffered saline washes (pH 7.2), cells were pelleted and weighed. Aliquots of each antiserum were thawed and incubated at 56"C for I h to inactivat e complement. Whole cells (100 mg wet wt) were added to 1 ml of dissimilar rabbit antiserum individually agitated every 15 min for 60 min at 37"C, then centrifuged at 12,000 x g for 5 min. The antiserum supernatant was removed, and the absorption completed for each antisera-cell combination. The entire cycle was repeated and the absorbed immune antisera were stored at -10*C.
Identification of Actinomyces sp.
319
Indirect Immunofluorescence The rabbit antiserum dilutions for the selected Actinomyces sp. were determined by "checkerboard" dilutions reacted with goat anti-rabbit antisera conjugated to fluorescein isothiocyanate. At room temperature, 15 ul of rabbit antiserum were placed on the positive control bacterial smear (e.g.A. israelii antiserum was placed on A. israelii whole cells), incubated for 20 min at 37~ washed with phosphate-buffered saline containing 0.05% Tween 20 twice for 10 rain each, and then rinsed with distilled water. The slides were incubated with 15 ul of goat anti-rabbit immunoglobulin G conjugated to fluorescein isothiocyanate, washed, and rinsed as before and covered with 0.10% glycerol in phosphate-buffered saline (pH 9.0). To test for cross-reaction, positive control smears of Actinomyces sp. were reacted against the absorbed Actinomyces sp. antisera (e.g.A. israelii antisera reacted with A. odontolyticus cells). Combinations of target organisms and serial 2-fold dilutions of antisera and conjugate (1:32 to 1:2096) were evaluated. The technique was optimum at an antiserum dilution of 1:512, and a conjugate dilution of 1:512 when reacted against reference strains. Cross-reactivity between subspecies, while present at this dilution, was distinguishable from a true positive reaction. Stock cultures of
Fusobacterium nucleatum, Haemophilus actinomycetemycornitans, Bacteroides intermedius, and Capnocytophaga ochracea were reacted with each antiserum and used as negative controls. The specimens were examined at xl000 using a Nikon Optiphot microscope (Nikon Inc., Garden City, NY) equipped for phase contrast illumination and incident light fluorescence. Each specimen was examined for the presence of fluorescing cells. Fluorescence was graded from 1+ to 4+. Only grade 4+ (brilliant fluorescence, brightly defined cell wall with center of cell completely dark) was considered a serologically positive reaction.
Immunofluorescence Microscopy The clinical samples were prepared following the protocol outlined above. A minimum of 60 fields of each smear was examined for each sample. A sample was considered positive for the presence of Actinomyces sp. if it exhibited l0 or more strongly fluorescent cell forms and 4+ fluorescence. Crossreactivity was also noted by less numerous weaker fluorescing cells. Positive and negative control reference strains were included for all IF trials.
Statistical Analysis Statistical measurements were performed with X 2 analysis of cross-tabulated contingency tables using Statgraphics (Ver 2.1; STSC, Inc., RockviUe, MD). The absence, presence, or suspected cross-reactivity for each Actinomyces sp. for all samples was cross-tabulated against each Actinomyces sp., the presence or absence of any Actinomyces sp., and against clinical parameters. The clinical parameters analyzed were: PuDx, PaDx, and presence of PARL. For the purposes of statistical analysis, the clinical parameters were collapsed and
320
Journal of Endodontics
Gohean et al.
recoded so that each parameter was either normal or not normal. For example, five PaDx were recorded for the study. These were chronic alveolar abscess, acute alveolar abscess, Phoenix abscess, acute apical periodontitis, and normal. This clinical parameter was recoded as not normal or normal. For each statistical test, the null hypothesis (1to) stated that groups analyzed were independent. The alternative hypothesis (/-/.4) stated that groups analyzed were not independent. The test for significance was set at a = 0.05.
TABLE 3, Immunofluorescence-positive findings for non-normal clinical diagnosis
AI* PuDx (n = 27) PaDx (n = 26) PARL (n = 18)
AO
AV1
AV2
ALL
0 0 0
4(14.8) 4(15.4) 0
18(66.7) 17(65.4) 13(72.2)
16(59.3)1- 3(11.1) 15(57.7) 3(11.5) 12(61.6) 1(5.5)
* AI, A. israelii; AO, A. odontolyticus; AV1, A. viscosus serotype I; AV2, A. viscosus serotype II. 1" Frequency (%).
RESULTS DISCUSSION Of 30 endodontic cases; PuDx included 17 cases necrotic, 6 partially necrotic, 3 retreatments, and 4 vital or normal teeth; PaDx included 8 cases with chronic alveolar abscess, 3 with acute alveolar abscess, 7 with Phoenix abscesses, 7 with acute apical periodontitis, and 5 normal. Diagnostic findings were collapsed to represent 27 (90%) PuDx that were not normal and 26 (87%) PaDx that were not normal (Table 1). Cases determined to have P A R L radiolucencies were 18 (60%). Data for IF findings in clinical samples are shown in Table 2. Of 30 clinical canal samples, Actinomyces sp. were identified in 18 (60%) samples. Of the total samples, 16 (53%) were positive to A. israelii antiserum, 3 (10%) positive to A. odontolyticus antiserum, 0 positive to A. viscosus serotype I antiserum, and 4 (13%) positive to A. viscosus serotype II antiserum. Cross-reactivity among Actinomyces sp. was observed in 18 (60%) of the samples. Reactions against like stock strains of target Actinomyces sp. were positive for each trial. Immunofluorescence responses against negative control organisms were not cross-reactive. Table 3 lists collapsed clinical findings that were IF positive for Actinomyces sp. The incidence of Actinomyces sp. in the presence of non-normal PuDx was 66.7%. Of teeth with a non-normal PaDx, Actinomyces sp. were identified in 65.4%. In cases with PARL, Actinomyces sp. were identified in 72.2%. The incidence of A. israelii in the presence of non-normal PuDx was 59.3%. Of teeth with a non-normal PaDx, A. israelii were identified in 57.7%. In cases with PARL, A. israelii were identified in 66.6%. The other target species are tabulated in Table 3 as well. Table 4 summarizes statistical correlations among and between target organisms and the selected clinical parameters. TABLE 1. Collapsed clinical findings (non-normal; n = 30)
PuDx
PaDx
PARL
27(90)*
26(87)
18(60)
9 Frequency (%).
TABLE 2. Results of indirect immunofluorescence for Actinomyces sp. (n = 30)
IF+l" IFIF+l-
AI*
AO
AV1
AV2
ALL
16(53):1: 12(40) 2(7)
3(10) 17(57) 10(33)
0(0) 19(63) 11(37)
4(13) 15(50) 11 (37)
18(60) 12(40) --
9 AI, A. israelii; AO, A. odontolyticus; AM1, A. viscosus serotype I; AV2, A. viscosus serotype II; ALL, all organisms. 1" Immunofluorescence negative (-), positive (+), cross-reaction (+/-). ~: Frequency (%).
Our results suggest that the presence of select Actinomyces sp. from endodontic canal samples can be identified in greater incidence through the use of IF than reported in previous culture studies. Culture recovery ofActinomyces sp. from clinical material is difficult. Identification by morphological and biochemical criteria is also difficult and time consuming. Actinomyces israelii is the principal species implicated in periapical actinomycosis lesions; nonetheless, Actinomyces sp. are generally found in mixed infections (7) and are not readily isolated through bacteriologic culturing. Identification of Actinomyces sp. by biochemical tests requires the preparation of special culture media and is further complicated by the broad range of strain variation within the species (4). Successful isolation in all clinical samples requires culturing multiple samples of purulent material in enriched media, under anaerobic conditions (optimally), and allowing adequate time (5 to 10 days) for growth. Actinomyces sp. may be mistaken for other organisms or they may remain unidentified if branching is infrequent or absent. While examining head and neck infections, Richtsmeier and Johns (8) reported only 25% positive cultures in known histological cases of actinomycosis. Of the 67 cases in which culture was attempted by Brown (9), the organism was isolated from only 16 samples. When reviewing culture studies and case reports, there is a high incidence with no reported growth of actinomycetes either in pure culture or with mixed flora (9). Wesley et al. (10) reported a case where Actinomyces sp. could not be recovered either before or after actinomycosis was characterized in biopsy. A review of relevant studies of endodontic pathogens (Table 5) shows that culture recovery of Actinomyces sp. is infrequent. These observations typify the inherent difficulties with culture studies of endodontic pathogens. With appropriate preparation, IF can be used successfully for surveys of canal contents without the need for culture recovery techniques, as is done for other clinical problems. Hemophilus influenzae in meningitis and Legionella species in Legionnaire's disease are routinely detected through immunofluorescence techniques. Similarly, IF has been used successfully in the identification of periodontal pathogens such as H. actinomycetemcomitans, B. intermedius, Bacteroides gingivalis, and Bacteroiodes melaninogenicus. Endodontic pathogens such as B. intermedius, B. gingivalis, and Bacteroides endodontalis have been identified with IF as well (6). These studies all have clearly demonstrated that IF is both more sensitive than culture for identification of target organisms and more rapid.
Identification of Actinomyces sp.
Vol. 16, No. 7, July 1990
The findings of the present study indicate cross-reactivity occurring among the target Actinomyces sp. in clinical samples (Table 2). In a review of their research and that of others, Gerencser and Slack (4) discussed the high magnitude of crossserologic reactivity among species of Actinomyces. They developed species-specific antisera using a direct immunofiuorescence technique. Bowden et al. ( 11 ) evaluated Actinomyces cell wall antigens. They reported that neutral cell wall carbohydrate and polypeptide containing charged antigens are c o r n -
ponents of the Actinomyces antigenic structure. They further concluded that cell wall carbohydrate can be responsible for species-specific reaction and cross-reactions, and strains may possess more than one cell wall carbohydrate determinant. Firtel and Fillery (12) raised monoclonal antibodies against 26 strains of A. viscosus and A. naeslundii. They reported extensive cross-reactions and only had three monoclonal antibodies specific for one strain each. Within the parameters of the present study, the strict criterion of a 4+ reaction was
TABLE 4. Summary of statistical analysis for cross-tabulations AI
AO
AI
x2 df Significance
17.066 2 1.969E-4
ALL*
--
--
AO
x2 df Significance
3.60312 0.165
12.827 4 0.012
--
AV1
x2 df Significance
AV2
x2 df Significance
PuDx
x2 df Significance
2.60811 Yates correction 0.106
5.926 2 0.051
1.76512 0.414
XXr
0.6401" 2 0.726
PaDx
x2 df Significance
0.97411 Yates correction 0.324
2.4161" 2 0.299
0.83212 0.660
0.00011 Yates correction 1.000
3.83812 0.147
PARL
x2 df Signifcance
1.67211 Yates correction 0.196
1.09412 0.579
0.82512 0.662
0.1561" 1 Yates correction 0.692
1.36412 0.506
1.40611 Yates correction 0.236 9.393 2 9.595E-6
2.15612 0.340 11.617 4 0.020
AV1
AV2
5.80212 0.079 12.104 4 0.017
9.845 2 7.279-3
* ALL, all organisms; AI, A. israelii; AO, A. odontolyticus; AV1, A. viscosus serotype I; AV2, A. viscosus serotype II. 1"Reject null hypothesis (Ho). XX, Sample too smalt for valid comparison.
TABLE 5. Comparative results of selected studies Culture +
Population
(Actinomyces sp.)
Biopsy
Wittgow and Sabiston (13)
40 cases 32 positive cultures
1/40 (0.03%)
ND*
Weir and Buck (1) (literature review)
20 cases 10 cultured
2/10 (20%)
Sabiston et al. (14)
65 cases 58 positive cultures
3/65 (0.05%) 3/58 (0.05%)
Happonen and Viander (2)
10 cases 5 cultured
1/5 (20%)
Study
20/20 (100%) ND 5/5 (100%)
Sundqvist (15)
18 cases
5/18 (28%)
ND
Kantz and Henry (16)
24 cases 16 positive cultures
5/24 (21%) 5/16 (31%)
ND
Oguntebi et al. (17)
10 cases
3/10 (30%)
ND
Williams et al. (18)
10 cases
0
ND
Matusow et al. (19)
62 cases
0
ND
Yoshida et al. (20)
36 cases 31 positive cultures
7/36 (19%) 7/31 (23%)
ND
* ND, not determined.
321
322
Journal of Endodontics
Gohean et al.
designated as a positive reaction. It may be that the two cycles of immune absorption were insufficient or that absorption should include another species or strain not included in this study. While it may be attractive to make clinical conclusions based on our statistical findings (Table 4), we are reluctant to do so. The purpose of this study was to use IF to determine the presence ofActinomyces sp. in clinical endodontic canal samples. However, despite the size of the sample population, some observations can be made. As determined by IF, there does seem to be a relationship between Actinornyces sp. and some forms of endodontic disease. This supports both the intuitive clinical observation as well as what has been reported in controlled studies (Table 5). Of the selected target species, A. israelii was most frequently identified with IF and most often associated with endodontic disease within the parameters of this study. Again, this supports culture findings that report A. israelii as the Actinomyces sp. most commonly associated with human disease, including endodontic disease. Conversely, the other target Actinomyces sp. (A. odontolyticus, A. viscosus serotype I, and A. viscosus serotype II) were not frequently associated with our sample population. This finding also supports culture studies of endodontic disease. Our findings indicate that Actinomyces sp. may be participants or major protagonists in microbial climax communities that establish certain endodontic diseases. CONCLUSION Indirect immunofluorescence microscopy can be used for the clinical identification of Actinomyces sp. in endodontic disease without the necessity of performing parallel culture surveys. This research was supported in part by the American Association of Dental Schools/Warner Lambert Clinical Research Fellowship award~,d to E. A. P. and also by a student research support grant from the Endowment and Memorial Foundation of the American Association of Endodontists awarded to R. J. G. The opinions, assertions, materials, and methodologies herein are private ones of the authors and are not to be construed as official or reflecting the views of the American Association of Endodontists. Dr. Gohean is in private practice in Columbia, SC. Dr. Pantera is assistant professor, Department of Endodontics, School of Dentistry, Medical College of
Georgia, Augusta, GA. Dr. Schuster is lone and Arthur Merritt Professor, Department of Oral Biology/Microbiology, School of Dentistry, and associate professor, Cell and Molecular Biology, School of Medicine, Medical College of Georgia.
References 1. Weir JC, Buck WH. Periapical actinomycosis. Oral Surg 1982;54:33640. 2. Happonen R-P, Viander M. Comparison of fluorescent antibody technique and conventional staining methods in diagnosis of cervico-facial actinomycosis. J Oral Patho11982;11:417-25. 3. Hotchi M, Schwarz J. Characterization of actinomycotic granules by architecture and staining methods. Arch Patho11972;93:392-400. 4. Gerencser MA, Slack JM. Serologic identification of Actinomyces using fluorescent antibody techniques. J Dent Res 1976;55(spacial issue A):A18491. 5. Happonen R-P, Soderling E, Viander M, Linko-Kettunen L, Pelliniemi LJ. Immunocytochemical demonstration of Actinomyces species and Arachnia propionica in periapical infections. J Oral Pathol 1985;14:405-13. 6. Pantere EA Jr, Zamben J J, Shih-Levine M. Indirect immunofluorescence for the detection of Bacteroides species in human dental pulp. J Endodon 1988; 14:218-23. 7. Finegold SM. Anaerobic bacteria in human disease. New York: Academic Press, 1977. 8. Richtsmeier WJ, Johns ME. Actinomycosis of the head and neck. CRC Crit Rev Clin Lab Sci 1979;11:175-202. 9. Brown JR. Human actinomycosis, a study of 181 subjects. Hum Pathol 1973;4:319-30. 10. Wesley RK, Osbern TP, Dytewski JJ. Periapical actinomycosis: clinical considerations. J Endodon 1977;3:352-5. 11. Bowden GH, Hardie JM, Fillery ED. Antigens from Actinomyces species and their value in identification. J Dent Res 1976;55(special issue A):A192204. 12. Firtel M, Fillery ED. Distribution of antigenic determinants between Actinomyces viscosus and Actinomyces naeslundii. J Dent Res 1988;67:1520. 13. Wittgow WC, Sabiston CB. Microorganisms from pulpal chambers of intact teeth with necrotic pulps. J Endodon 1975;5:168-71. 14. Sabiston CB Jr, Gringsby WR, Segerstrom N. Bacterial study of pyogenic infections of dental origin. Oral Surg 1976;41:430-5. 15. Sundqvist G. Bacteriological studies of necrotic dental pulps [Dissertation 7]. Umea, Sweden: University of Umea, 1976. 16. Kantz WE, Henry CA. Isolation and classification of anaerobic bacteria from intact pulp chambers of non-vital teeth in man. Arch Oral Bio11974;19:916. 17. Oguntebi B, Slee AM, Tanzer JM, Langeland K. Predominant microflora associated with human dental periapical abscesses. J Clin Microbiol 1982;15:964-6. 18. Williams BL, McCann GF, Schoenknecht FD. Bacteriology of dental abscesses of endodontic origin. J Clin Microbiol 1983;18:770-4. 19. Matusow RJ, Goodall LB. Anaerobic isolates in primary pulpal-alveolar cellulitis cases: endodontic resolutions and drug therapy considerations. J Endodon 1983;9:535-43. 20. Yoshida M, Fukushima H, Yamamoto K, Ogawa K, Toda T, Sagawa H. Correlation between clinical symptoms and microorganisms isolated from root canals of teeth with periapical pathosis. J Endodon 1987;13:24-8.
