Immunoregulation in severe generalized periodontitis

CLINICAL

IMMUNOLOGY

AND

IMMUNOPATHOLOGY

lmmunoregulation KATHLEEN

34, 84-93 (1985)

in Severe

Generalized

Periodontitisl

MCANULTY, ROSLYN STONE, GEOFFREY JAMES CLAGETT, AND DAVID ENGEL?

Departments of Periodontics, for Research in Oral

Pathology, Microbiology-Immunology, Biology, University of Washington,

HASTINGS,

Biostutistics, und the Center Seattle. Washington 98195

Severe generalized periodontitis (SGP) is an inflammatory disease which leads to extensive alveolar bone loss in young adults. Peripheral blood lymphocytes from SGP patients have been previously reported to exhibit an by vitro hyperproliferative response when exposed to B cell mitogens derived from Staphy/ococctrs ailrem and Actinomyces viscosus. Therefore hyperresponsiveness to B-cell mitogens could be an important pathogenic factor in the susceptibility to and progression of SGP. We have tested whether the hyperproliferative response of lymphocytes from SGP patients was due to (i) a functional deficiency of suppressor T cells, or (ii) to numerical alterations of lymphocytes. Supernatant fluids from concanavalin A-stimulated T cells from 14 SGP patients and I4 normal subjects were compared for their ability to suppress the IgM synthesis of B-cell mitogen-stimulated mouse splenocytes. No significant differences were noted in suppressor T-cell function between control subjects and SGP patients. However. SGP patients had significantly higher lymphocyte counts than control subjects, and there was a positive correlation between high lymphocyte counts and high mitogen-stimulated proliferation. SGP patients also had higher lymphocyte:monocyte ratios than control subjects, suggesting that a defect in macrophage-mediated suppression might be involved in the hyperproliferation phenomenon. Our data do not support the hypothesis that a suppressor T-cell defect is the cause of mitogen-induced hyperproliferative responsiveness of peripheral blood lymphocytes from SGP patients. Rather. hyperproliferation may be due to an expansion of the lymphocyte pool which responds to mitogens. or/and a regulatory disturbance which arises because of altered 1ymphocyte:macrophage ratios. i’ 1985 Academic PI-~\\. Inc

INTRODUCTION

In its usual form, periodontitis is a chronic, slowly progressive disease which may affect only a portion of the dentition and which is most often seen in the fifth to seventh decades of life (1). Another form of periodontitis which is rapidly progressive, and which affects relatively young adults, has recently been described (2, 3). In this form of the disease, which we call severe generalized periodontitis (SGP),3 most or all of the teeth are affected by advanced alveolar bone loss and connective tissue breakdown. It is generally agreed that periodontitis is caused by bacteria (4), and that the ’ This work was supported by Public Health Service Grant-DE-02600. * Correspondence should be sent to Dr. David Engel, Department of Periodontics, SM-44. University of Washington, Seattle, Wash. 98195. 3 Abbreviations used: AVIS, a mitogenic preparations from Actinomyces viscosas; Con A. concanavalin A; ELISA, enzyme-linked immunosorbent assay; FBS. fetal bovine serum; Ig, immunoglobulin; PBMNC, peripheral blood mononuclear cell; PBS-T, phosphate-buffered saline with Tween 20; SGP, severe generalized periodontitis; WBC. white blood cell. 0090- 1229185 $1.50 Copyrghf r’ 198? by Academw Pren. Inc. All rights of reproductmn m any form reerved.

84

IMMUNOREGULATION

IN

PERIODONTITIS

85

host inflammatory response to bacteria may lead to collagen loss and bone resorption. The host response to oral bacteria is complex and involves numerous facets of the inflammatory-immunologic network, including polymorphonuclear leukocytes, immunoglobulins, complement, macrophages, and lymphocytes. These features have been reviewed by Seymour and his colleagues (5) and the reader is referred there for additional details. It appears likely that the lymphocyte response is of major importance in the immunopathogenesis of severe periodontitis. Although T cells are the major lymphocyte subpopulation in the early gingivitis lesion, the predominant inflammatory cell types identified in advanced periodontal lesions are B lymphocytes and plasma cells (6). This suggests that the severe periodontal lesion in man is primarily a B-cell lesion, and that B-cell responses play a central role in the pathogenesis of the disease (6). Severe periodontitis may occur as a consequence of B-cell hyperreactivity to components of plaque bacteria which have gained entrance into the periodontal soft tissues (7). Mitogenic substances derived from the cell walls of plaque organisms may in part initiate this disease by nonspecifically activating B cells in the periodontal connective tissues (7, 8). Peripheral blood lymphocytes from patients with severe periodontitis have been reported to be hyperresponsive to B-cell mitogens derived from Sfuphylococcus aureus (7) and Actinomyces viscosus (9). Mitogen-activated B cells are capable of secreting large amounts of osteolytic mediators, as well as other potentially tissue-destructive lymphokines (IO- 13) which may participate in the pathogenesis of periodontitis. We have attempted to elucidate the biological basis for mitogen-induced hyperresponsiveness. We have begun by testing for functional abnormalities of regulatory T lymphocytes, since T-cell regulation has been shown to be important in modulating the response to Actinomyces viscosus (14- 16). We have also tested the total numbers and differential counts of white blood cells from SGP patients and normal subjects. We found no evidence of functional abnormality in the suppressor T cells of SGP patients. However, we did find subtle but highly significant differences in the overall number of lymphocytes in the peripheral blood of SGP patients, as well as a significant difference in the mean ratio of lymphocytes to monocytes. These findings may indicate an increase in the mitogenreactive subset(s) of lymphocytes in SGP patients, or a heretofore unsuspected alteration in the immunoregulatory network of these individuals. MATERIALS

AND METHODS

Human subjects. Fourteen patients with severe generalized periodontitis between the ages of 22 and 42 years were selected for the study. These individuals had nearly complete dentitions, but exhibited advanced loss of gingival connective tissue attachment to the teeth and generalized, severe alveolar bone loss. As a measurement of gingival attachment loss, interproximal pocket depths of all teeth were determined with a periodontal probe in standard manner (17), and the mean interproximal pocket depth was calculated. Dental radiographs were used to calculate degree of bone loss by the method of Schei (18). Control subjects were individuals with healthy periodontal tissues who were either seeking routine dental care in other clinics, or who were working in the

MC ANULTY

86

ET AL.

Health Science Center of the University of Washington. All SGP patients and control subjects received complete periodontal examinations and radiographs. None of the patients or controls had a history of systemic illness, and none was taking anti-inflammatory drugs, or other prescribed medications. The patient and control groups were balanced with respect to sex. and the age distributions of the two groups were similar. Blood samples from all patients and control subjects were submitted to the University of Washington hematology laboratory for complete blood counts and differential white cell counts. None of the subjects included in the study had remarkable hematological abnormalities. A summary of the clinical data obtained from the SGP patients and control subjects is shown in Table 1. Human lymphocyte preparation. Approximately 100 ml of heparinized whole venous blood was obtained from each patient and control subject. The buffy coat was isolated and the mononuclear cells collected by density-gradient centrifugation on Ficoll-Hypaque solution (LSM; Litton Bionetics, Kensington, Md.). These peripheral blood mononuclear cells (PBMNC) were washed twice in RPM1 1640, and resuspended in RPM1 1640 medium supplemented with glutamine (2 mM), Garamycin (gentamicin sulfate, USP, 10 mgiml), 4 mM HEPES buffer, and 10% (v/v) heat-inactivated fetal bovine serum (FBS) (Hyclone; Sterile Systems, Logan, Ut.). This supplemented medium will be referred to as complete medium. A T-cell-rich suspension was prepared by rosetting peripheral lymphocyte suspensions with 2-aminoethylisothiouronium bromide-treated sheep erthyrocytes (Triple J Farms, Redmond, Wash.) as previously described (19, 20). The sheep TABLE 1 CLINICAL STATUS OF STLIO~ PAK I ICIPANTS Bone Control subject JC VD PE SE JF RF EG HH AJ BJ MS MT BV EW Mean (SEW

Sex M F F M M31 F F M M F M M F F

Age 39 29 39 32

No. teeth

32 33 40 39 37 34 21 34 34

27 27 24 30 28 28 28 28 30 28 27 32 27 31

34.4 (1.1)

28.2 (0.6)

Pocket depth” (mm)

SGP patient

Sex

0.9 1.2 1.9 0.0 0.0 0. I 0.6 0.5 0.9 1.9 0.0 0.9 0. 1 I.5

2.8 3.3 2.5 3.0 2.7 3.0 3.1 2.8 2.9 3.5 3.2 2.6 2.7 3.2

JB MB GE LK AL. CL AM IR cs GS SS SS ST ‘l-l-

M F M F M F F F M F F F M M

0.75 (0.19)

2.95 (0.08)

Mean (SEM)

IOSS"

CT;)

Age 21 36 36 41 39 42 28 33 39 32 a: 38 42

No. teeth

Bone loss” (%I

Pocket depthb (mm)

30 28 '7 25 30 26 28 31 ‘9 26 23 29 26 76

19.5 22.5 42.0 32.6 28.6 37.2 74.9 15.7 29.0 74.6 20.0 41.5 39.0 26.0

4.4 5.4 6.3 1.2 5.2 4.9 -5.7 4.6 1.x 6.8 4.5 5.x 6.4 4.0

35.4 27.6 29.6 5.46 (1.8) (0.6) (2.4) (0.25) -_____-..-__ ” Bone loss is expressed as the mean for all teeth as determined on radiographs using the method of Schei (18). b Data are expressed as the mean interproximal periodontal pocket depths in mm for each patient.

IMMUNOREGULATION

IN

PERIODONTITIS

a7

erythrocyte rosette-forming cells were collected by centrifugation through FicollHypaque, the erythrocytes lysed with 0.87% Tris-buffered NH&l, and the lymphocytes washed three times and suspended in complete medium at a concentration of 2 X lo6 cells per ml. Generation of Con A-activated T-cell supernatants. Human peripheral blood T cells were cultured in the presence of 10 pg/ml of concanavalin A (Con A; Calbiochem, San Diego, Calif.) for 16 hr to generate suppressor cells (21, 22). Cells were cultured in flat-bottomed wells of 24-well plates (Linbro No. 76-033-05, McLean, Va.) at a concentration of 2 x lo6 cells per ml and plated out at 2 ml per well. After 16 hr the cells were washed three times with RPM1 1640 medium to remove residual Con A. The cells were resuspended in complete medium at a density of 2 x lo6 cells per ml, plated out at 2 ml per well, and cultured for an additional 28 hr. The supernatants of these cultures were collected, filtered through a 0.22~km Millipore tilter to remove the cells, and stored at -20°C. Mouse lymphocyte preparation. Mouse splenic lymphocytes were used as responder cells to test the suppressor activities of the T-cell supernates (see ELISA assay protocol below). Twelve-week-old C57BL/6J mice (Jackson Laboratories; Bar Harbor, Me.) were sacrified by cervical dislocation and their spleens removed. Spleen cells were prepared as previously described (8) and were suspended in RPM1 1640 medium containing glutamine (2 mM), Garamycin (gentamicin sulfate, USP, 10 mg/ml), and 5% (v/v) human serum. The lymphocytes from three mice were pooled for use as responder cells in each experiment. Use of ELZSA assay to measure suppressor activity. Suppressor activity of the T-cell supernatants was measured by the ability of the supernatants to suppress IgM synthesis in mouse lymphocyte cultures stimulated with a B-cell mitogen from Actinomyces viscosus bacteria (AVIS) (8). We chose to assay for IgM production because we previously determined that AVIS-stimulated murine B lymphocytes preferentially synthesize IgM (23). An enzyme-linked immunosorbent assay (ELISA) was used to make these determinations of IgM production. Mouse splenocytes were cultured for 7 days in flat-bottom wells of 96-well plates (Linbro No. 76-003-05; McLean, Va.) at a cell density of 4 x lo5 cells per well, in the presence or absence of AVIS mitogen and T-cell supernatants. T-cell supernatants were used at dilutions of 1:2, 1:4, 1:8, 1:20, 1:40. After 7 days, the fluid phases of these cultures were collected and stored at - 2o”C, prior to assaying for IgM content. Flat-bottomed wells of 96-well plates (Linbro No. 76-003-05; McLean, Va.) were coated with affinity-purified goat anti-mouse IgM (TAGO, Inc; Burlingame, Calif.) by incubation at 37°C for 3 hr with 2000 ng antibody per well. The plates were washed 3 times with phosphate-buffered saline with Tween 20 (PBS-T). Mouse splenocyte culture supernatants (diluted 1: lo), or standards of IgM (Litton Bionetics; Charleston, S.C.) diluted in PBS-T plus 5% FBS, were added to the anti-immunoglobulin treated plates. After 2 hr incubation at room temperature, the plates were washed 3 times with PBS-T. Alkaline phosphatase-conjugated goat anti-mouse IgM in PBS-T FBS were added to the plates, incubated for 2 hr at room temperature, and washed 3 times with PBS-T. The phosphatase substrate (Sigma: St. Louis, MO.) diluted in 10% diethanolamine buffer was added. After

88

MCANULTY

ETAL.

incubation at room temperature for 30 min in the dark, the reaction was terminated by addition of 5 N NaOH solution to the wells. Absorbence was read at 410 nm on an autoreader spectrophotometer (Dynatech, MR 580; Alexandria, Va.). Transformation of ELISA data for analysis. The ELISA assay was run on 6 different occasions. For each experiment, observed absorbence values for the dilutions of the IgM standard were used to construct a curve relating optical density to nanograms of IgM. Transforming the optical density values to their square roots, and calculating the log (nanogram + 1) linearized this relationship. A least-squares line was then fit to the transformed standard data for each of the six experiments. Although the six separate calibration curves fit the data quite well (R2 = 0.98), the curves were nonparallel and therefore could not be combined to provide one standard curve for all the data. Conversion of the mean absorbence values of the IgM supernatant samples to nanogram values was based on experiment-specific standard curves. Statistical analyses. Analysis of suppression was done in the nanogram scale, and all suppression comparisons between SGP patients and controls were done within experiment. This necessitated excluding a single subject who was assayed in the sixth experiment. A univariate two-way mixed model analysis of variance was computed for each dilution level of the suppression variable to test the hypothesis of no difference between patients and controls, controlling for the random effect of day. Concentration of IgM produced by cultures treated with Con A-stimulated Tcell supernatants, relative to IgM produced by cultures treated with supernatants from unstimulated T cells, was compared at each dilution level using a paired t test. The SGP patients and controls were compared with respect to differential IgM production under the two conditions using 2-sample t tests on the difference scores (Con A-stimulated minus medium control). The peripheral blood values of the SGP patients and controls were compared using 2-sample t tests for equality of group means. Spearman rank correlations were computed to assess the relationship of lymphocyte counts to mitogen-stimulated proliferation for a subset of 10 subjects for whom proliferation data was available from a previous study (9). RESULTS

Effects of Con A-activated T-cell supernatants on IgM synthesis. Suppressor T-cell activity can be quantified by determination of the amount of IgM produced by mitogen-stimulated B cells in the presence or absence of putative suppressor T-cell substances. High IgM production may indicate low suppressor T-ceil activity; conversely, low IgM production may reflect high suppressor T-cell activity. In our experiments, an ELISA assay was used to quantify IgM production by AVIS-stimulated B cells in the presence or absence of Con A-activated T-cell supernatants. The ELISA assay was run and analyzed on five different occasions, and groups of three to six subjects were examined in each experiment. The results in Fig. 1 are from a representative experiment. No significant differences were found in

IMMUNOREGULATION

0 CCmcen+rot,on

215 of T-cell

IN PERIODONTITIS

HP ,s 50 supernatanf

(pi/well

89

100 1

FIG. 1. The amount of IgM produced by AVIS-stimulated mouse splenocytes treated with varying concentrations of Con A-activated or medium control T-cell supematants. The number of subjects tested in this representative experiment were 3 control subjects with healthy periodontiums (HP) and 3 SGP patients. Standard errors of the means are indicated. The differences between the group means of the SGP patients and control subjects were not statistically significant at any of the T-cell supernatant concentrations tested.

the amount of IgM produced by cultures treated with putative suppressor supernatants from Con A-activated control subject T cells or analogous supernatants from Con A-activated T cells of SGP patients (P > 0.39, controlling for the random effect of day. No difference in the concentration of IgM produced was found between medium control and Con A-stimulated T cells at supernatant concentrations of 5 or 10 t.~l per well. At T-cell supernatant concentrations of 25 and 50 t.~l per well, the supernatants from Con A-stimulated T cells elicited higher IgM production than the supernatants from unstimulated T cells (P < O.OOl), suggesting that helper factors were generated. However, Con A-stimulated T-cell supernatants from SGP patients did not elicit more help than Con A-stimulated T-cell supernatants from control subjects (P > 0.60). At the highest tested concentration of Con A-stimulated T cell supernatant (100 t.~l per well), the AVIS-stimulated mouse splenocyte cultures showed significantly less IgM production (P < O.OOl), indicating the presence of suppressor factors in the Con A-stimulated T cell supernatants. However, no differences were observed between control subjects and SGP patients with respect to the ability of their Con A supernatants to suppress IgM synthesis (P > 0.60). White blood cell counts. Complete blood counts and differential white cell counts were done on peripheral blood from all SGP patients and control subjects. The results are presented in Fig. 2. SGP patients had significantly higher total WBC counts (P = 0.002), lymphocyte counts (P = O.OOl), and neutrophils (P = 0.015). The 1ymphocyte:monocyte ratio values (Fig. 3) were also higher in SGP patients (P = 0.05). This apparently was due to the significantly higher lymphocyte counts found in the SGP patients, since no difference between SGP patients and control subjects was found for monocytes (P = 0.42).

MC ANULTY

ET AL.

“1 T

FIG. 2. Comparison of mean cell numbers per mm’ of peripheral blood in normal control (89) and SGP (69) subjects. Vertical bars indicate standard errors of the means. Total white blood cells (WBC), lymphocyte counts, and neutrophil counts were significantly higher in SGP patients. There was no difference between the SGP and control subjects’ mean monocyte counts.

DISCUSSION

Patients with severe, generalized periodontitis appear to be a phenotypically distinct group of individuals who, at a relatively young age, develop profound periodontal inflammation with extensive alveolar bone resorption in response to

FIG. 3. Lymphocyte (Ly) to macrophage (MI+) ratios of control subjects (0) and SGP patients (e). Each dot represents the ratio of an individual subject. Group means and standard errors of the means are indicated. The mean LY:M+ ratio of SGP patient was significantly higher than that of the control subjects.

IMMUNOREGULATION

IN PERIODONTITIS

91

bacterial challenge. Little is known about the pathogenesis of this syndrome. We and others have previously shown that peripheral blood lymphocytes from many of these patients are hyperresponsive to bacterial-derived B-cell mitogens (7, 9). Several studies have found that regulatory T cells exert modulating influence over the bacterial-mitogen induced response (14-16) and thus disturbances in these subsets of T cells could play a role in the hyperresponse phenomenon. However, there are no apparent differences in the numbers or ratios of helper and suppressor T cells in these patients (9). Functional abnormalities of suppressor T cells have, however, been reported in several other diseases with a prominent inflammatory component, including chronic active hepatitis (24>, idiopathic systemic lupus erythematosus (25), and active inflammatory bowel disease (26). In this study we have tested for possible functional defects of suppressor T cells in patients with SGP. We were not able to demonstrate any differences between our control subjects and SGP patients using the suppressive activity of Con A activated T cell supernatants as a measure of suppressor T cell function. It is conceivable, however, that SGP patients may have a type of suppressor T cell defect which could not be detected by the method we used. We have also tested total and differential WBC counts of the subjects in this study. The SGP patients had significantly higher total WBC, lymphocyte, and neutrophil counts than did the control subjects. The total WBC counts of all control subjects were within the normal range, while two SGP patients were slightly above this range. However, the SGP patients’ mean values for total WBC, lymphocyte, and neutrophil counts were within normal laboratory ranges. The lymphocyte counts of the control subjects were at the low end, or slightly below, the reference range; values for SGP patients were all in the higher end of the range. This is in contrast to the findings of Ranney and co-workers (2), who reported no difference for total WBC and differential leukocyte counts between healthy subjects and severe periodontitis patients. They reported that all mean values were within normal laboratory ranges, and there was no difference in the number of individuals from either group with values slightly above or below published normal laboratory values. We have no explanation for why our results differ from those of Ranney et al., although they may reflect differences in disease activity of the patient populations tested. Whether or not elevated numbers of lymphocytes has any relationship to hypet-responsiveness to B-cell mitogens in SGP patients is an important question. We therefore tested for a correlation of the high lymphocyte counts obtained in this study and the high proliferation responses observed in our previous study (9). Due to the large number of patients who ceased participation after the first study was completed, we had test results on only five control subjects and five SGP patients who participated in both studies. Nevertheless, we found high correlations between lymphocyte number and the magnitude of proliferative responsiveness. When lymphocyte number vs proliferation values at days 3, 5, and 7 of culture for all 10 subjects (control and SGP) were tested, we obtained P values of 0.001, 0.001, and 0.05 respectively. Therefore, it appears that subtle elevations in lymphocyte numbers in SGP patients may be an important factor in the mitogen-induced hyperproliferative response of PBMNC from these patients. Why

92

MC ANULTY

ET AI..

this is the case remains unresolved. One possibility is that the increase in lymphocytes leads to alteration of the normal homeostatic ratio of lymphocytes to another cell type with regulatory functions, the macrophage. In support of this hypothesis, we did find that the mean 1ymphocyte:macrophage ratio of SGP patients was significantly different than that of the control subjects (P = 0.05). This suggests that an aberation in the relative proportions of lymphocytes to macrophages may also be an important factor in hyperresponsiveness. Although there was only a weak positive correlation (P = 0.064) between lymphocyte: macrophage ratio and high proliferative responsiveness in the 10 subjects who participated in both the previous study (9) and this study, the small sample size does not allow a very powerful test. There is evidence that macrophages act as a modulating influence in the response of lymphocytes to polyclonal B-cell activators from periodontopathic organisms (27), most probably through mechanisms involving production and release of prostaglandin E2 (28). Page ef ul. have previously shown that prostaglandin E2 is a potent inhibitor of the proliferative response of human PBMNC to AVIS mitogen (29). It therefore seems worthwhile to further test the relationship of 1ymphocyte:macrophage ratios and proliferative responsiveness in future studies on these patients. It is also possible that mitogen-induced hyperprohferative responsiveness is due to an expansion of the lymphocyte subset(s) which respond to these mitogens. An expansion of the responding subset(s) could be due to (i) genetic factors. or (ii) immunomodulating activity of periodontopathic bacteria. If genetic factors are primarily responsible, lymphocyte subset expansion and resultant hyperresponsiveness would be stable characteristics of these patients. As a model for this hypothesis, hyperproliferative responsiveness of B cells to AVIS mitogen has been described as a stable phenotype of the SMiJ strain of inbred mice. This hyperresponse trait is determined by two or more autosomal, non-H-2 linked genes (30), which appear to induce an expansion in the B cell subset responsive to AVIS mitogen (Engel et nl., unpublished). On the other hand, if bacterialinduced immunomodulation is playing a key role, hyperresponsiveness would be expected to be transient and would correlate with infection of the SGP patient with one or more of the organisms associated with severe periodontitis. While there have been reports on the transient immunomodulating activities of certain bacteria associated with severe periodontal disease, these effects have been primarily to suppress proliferative response to mitogens, rather than to enhance them (31, 32). Finally, the elevated neutrophil counts observed in SGP patients probably bears no relationship to the regulation of lymphocyte responses to mitogens, which was the main theme of this study. However, our observation may be of relevance to the report of Page er al. (3) that five of the seven SGP patients they studied had abnormal neutrophil chemotactic responses. Although it is likely that differences in the numbers and functions of lymphocytes and neutrophils are separate, unrelated alterations in the immunologic network of SGP patients, there could be a common etiologic factor, genetic or environmental, which elicits these alterations. The importance of these alterations in the susceptibility to and/or progression of SGP remains unclear.

IMMUNOREGULATION

IN PERIODONTITIS

93

ACKNOWLEDGMENTS We thank Dr. Robert H. Johnson and Dr. Thomas Stanton for their helpful advice in planning these experiments, and Dr. John Tew for his careful review of, and suggestions for improving, the manuscript.

REFERENCES 1. Henry. J. L., and Sinkford, J. C., Pub. Health Rep. 94, 172, 1979. 2. Ranney, R. R., Ruddy, S., Tew, J. G., Weishimer, H. J., Palcanis, K. G.. and Segreti, A., J. Periodont. Res. 16, 390, 1981. 3. Page, R. C., Altman, L. C., Ebersole, J. L., Vandesteen, G. E., Dahlberg, W. H., Williams. B. L.. and Osterberg, S. K., J. Periodontol. 54, 197, 1983. 4. Socransky. S. S., J. Periodontol. 48, 497, 1977. 5. Seymour, G. J., Powell, R. N., and Davis, W. I. R.. J. Oral P&ho/. 8, 249, 1979. 6. Seymour, G. J., and Greenspan, J. S., J. Periodont. Res. 14, 39, 1979. 7. Smith, S., Bick, P. H., Miller, G. A., Ranney, R. R., Rice, P. L., Lalor, J. H., and Tew, J. G., C/in. Immunol. Immunopathol. 16, 354, 1980. 8. Engel, D., Clagett, J., Page, R., and Williams, B., J. Immunol. 118, 1466, 1977. 9. Engel, D., Monzingo, S., Rabinovitch, P, Clagett. J., and Stone, R.. C/in. Immunol. Immunopathol. 30, 374, 1984. 10. Mackler, B. E, Altman, L. C., Rosenstreich. D. L., and Oppenheim, J. J., Nature (London) 249, 834, 1974. 11. Bloom, B. R., Stoner, G., Gaffney, J., Shevach, E., and Green, I., Eur. J. Immunol. 5, 218, 1975. 12. Chess, L.. Rock& R. E.. MacDermott. R. P., David, J. R., and Schlossman, S. F.. J. Immunol. 115, 315, 1975. 13. Chen, P., Trummel, C., Horton, J., Baker, J. J., and Oppenheim, J. J.. J. Zmmunol. 6, 732, 1976. 14. Engel, I)., and Felts, C., J. Dent. Res. 61, 330, 1982. 15. Lopatin, D. E., Mangan, D. F., and Horner, I. S., Clin. Immunol. Immunopathol. 19, 394, 1981. 16. Carpenter, A. B., Sully, E. C., Ranney, R. R., and Bick, P. H., Infect. Immun. 43, 326, 1984. 17. Schluger, S.. Yuodelis, R. A., and Page, R. C., “Periodontal Disease.” Lea & Febiger. Philadelphia, 1977. 18. Schei, O., Waerhaug, J., Loudal, A., and Arno, A., J. Periodontol, 30, 7. 1959. 19. Kaplan, M. E., and Clark, C., J. Immunol. Methods 5, 131, 1974. 20. Pellegrino, M. A., Ferrone. S., Dierich. M. P., and Reisfeld. R. A., C/in. Immunol. Immunopathol. 3, 324, 1975. 21. Greene, W. C., Fleisher, T. A., and Waldmann, T. A., J. Immunol. 126, 1185, 1981. 22. Fleisher, T. A., Greene, W. C., Blaese, R. M., and Waldmann, T. A.. J. Immunol. 126, 1192. 1981. 23. Clagett, J., Engel, D., and Chi, E., Infect. Immun. 29, 234, 1980. 24. DeGalocsy, C., Jenkins, P. J., Mielivergani, G., Eddleston, A. L. W. E, and Williams, R., C/in. Exp. Immunol. 43, 486, 1981. 25. Krakauer, R. S., Clough, J. D., Frank, S., and Sundeen, J. T., C/in. Immunol. Immunopathol. 14, 327, 1979. 26. Hodgson, H. J. F., Wands, J. R., and Isselbacher, K. J., C/in. Exp. Immunol. 32, 451, 1978. 27. Carpenter, A. B., Sully, E. C., Palcanis, K. C., and Bick, P. H., Infect. Zmmun. 42, 853. 1983. 28. Goodwin, J. S.. and Ceuppens, J., J. Clin. Immunol. 3, 295, 1983. 29. Page, R. C.. Clagett, J. A., Engel, L. D., Wilde, G., and Sims, T.. C/in. Immunol. Immunopathol. 11, 77, 1978. 30. Engel, D., Clark, E. A., Held, L., Kimball, H., and Clagett, J.. J. Exp. Med. 154, 726, 1981. 31. Shenker, B. J., McArthur, W. P., and Tsai, C. C., J. Immunol. 128, 148, 1982. 32. Bolton. R. W., and Dyer, J. K., Infect. Immun. 39, 476, 1983. Received May 7, 1984; accepted with revisions August 16. 1984.