217
Stimulation of Lymphocytes in Vitro by Bacteroides intermedius and Bacteroides (Porphyromonas) gingivalis Sonicates J.E. Raber-Durlacher* W.P.
Zeijlemaker,t A.A.P. Meinesz,* L. Abraham-Inpijn*
The present study was designed to assess whether the in vitro stimulation of lymphocytes by sonicates of Bacteroides intermedius and Bacteroides (Porphyromonas) gingivalis is antigen specific or non-specific. In addition, the role of and lymphocytes in these responses was assessed. Peripheral blood lymphocytes obtained from healthy volunteers were cultured in the presence of these bacterial preparations and the proliferative response was measured. In similar experiments the response of umbilical cord blood lymphocytes did not exceed background values. In limiting dilution experiments only 1:4000, 1:6800, and 1:8200 of the lymphocytes initially reacted to B. intermedius, which strongly argues for the antigen-specificity of the response. Purified cells, in the presence of monocytes, proliferated when stimulated with B. intermedius and B. gingivalis. As for cell stimulation, the bacterial extracts were capable of inducing IgM production, which appeared to be cell dependent. These findings support the notion that B. intermedius and B. gingivalis induce specific cell activation; secondarily, a cell activation may occur. J Periodontol 1990; 61:217cell dependent, polyclonal 223.
Key Words: IgM; T-cell activation; B-cell activation; Bacteroides intermedius; Bacteroides
(Porphyromonas) gingivalis; lymphocytes
performed on the antigenlymphonon-specific specific cyte response to oral bacteria, preparations of Bacteroides gingivalis and/or Bacteroides intermedius have been used as stimulants. The latter microorganisms are considered to play an important role in the etiology of periodontal diseases. Consequently, defining host responses to these particular bacteria may contribute to knowledge of the pathogenesis of these diseases. Studies of the in vitro blastogenic response of whole peripheral blood lymphocytes (PBL) to isolates of B. asaccharolyticus (now classified as B. gingivalis) have suggested antigen specificity.1 On the other hand, bacterial preparations from B. gingivalis and B. intermedius were reported to possess the capacity to induce mitogenesis in germ-free mouse spleen cells, human umbilical cord blood lymphocytes (UCL), and PBL from periodontally healthy adults.2 With regard to the capacities of extracts of B. intermedius and B. gingivalis to induce a polyclonal antibody In
a
few studies that have been versus
nature of the in vitro
'Department of General Pathology and Internal Medicine, Academic Center for Dentistry Amsterdam, (ACTA), Amsterdam, The Netherlands. tDepartment of Clinical Immunobiology and Hybridoma Laboratory, Central Laboratory of the Netherlands Red Cross Bloodtransfusion Service Amsterdam, The Netherlands.
response in cultures of PBL, these particular bacteria appeared to be weak activators.3 A number of studies have been performed on separated lymphocyte populations in order to assess the role of cells in the antibody response to oral bacterial antigens. The generation of IgG and IgM responses induced by a preparation of B. intermedius was shown to be strictly dependent upon the presence of cells.4 It was shown that B. intermedius could act as a polyclonal activator for human
lymphocytes, inducing Immunoglobulin production higher than that induced by pokeweed mitogen (PWM). A preparation of B. gingivalis was not included in the panel of stimulants. In view of these inconclusive and apparently conflicting data, the objectives of the present study were to: analyze in vitro PBL reactivity to preparations of B. intermedius and B. gingivalis; determine whether the response is antigen-specific or non-specific; and assess the role of T- and B-lymphocytes in these responses. First, optimal culture conditions and in vitro PBL reactivity to preparations of B. intermedius and B. gingivalis were assessed. Similar experiments were performed on umbilical cord blood lymphocytes to determine whether these stimulants are capable of inducing polyclonal activation in a presumably non-sensitized lymphocyte population. The is-
218
of antigen-specificity was further investigated by means of limiting dilution (LD) experiments. Finally, and nonT cells (B cells) were separated in order to assess their role in the in vitro blastogenic responses to B. intermedius and B, gingivalis. Proliferative responses were measured by 3Hthymidine incorporation and cell activation by means of sue
IgM production.
MATERIALS AND METHODS Human Subjects Twelve volunteers
(5 males and 7 females, aged 24 to 40 years) without clinical or historical features òf Periodontitis were selected (pocket depths ranging from 1.7 mm to 2.0 mm; Periodontal Pocket Bleeding Index, PPBI5 ranging from 0.2-0.3). All subjects were in good health and no medication that could influence immunological responses had been taken for the previous 3 months. After informed consent was secured, 60 ml of venous blood
was obtained from these individuals for the initial experiments, in which optimal culture conditions were assessed and PBL responses to B. intermedius and B. gingivalti were analyzed. After these initial experiments were completed, 3 of the subjects twice donated an additional 200 ml of venous blood for limiting dilution experiments and for the experiments in which the cell populations involved in blastogenic responses and IgM
production were analyzed. Lymphocyte
(ATCC 25611) and B. gingivalis strain HG762 were kindly provided by Dr. TJ.M. van Steenbergen (Department of Oral Microbiology, Academic Center for Dentistry Amsterdam (ACTA)). Bacteria were cultured in liquid BM medium,6 centrifuged, and handled in phosphate buffered saline. Extracts were prepared by freeze-thawing, sonication, and centrifugation of cell wall fragments as described by Baker
and Tondreau.7 Protein concentrations of the bacterial sonicates were measured by means of the Bicinchoninic Acid Protein Assay.8 The protein concentration of the B. intermedius antigen preparation was 0.85 mg/ml; the sonicate of B. gingivalis had a protein concentration of 3.2 mg/ml.
Mitogens. Pokeweed mitogen (PMW)* final concentration
(purified Phytohaemagglutinin HA16)§ fiµg/ml. ALS, horse anti-human lymphocyte serum (CLB) optimal final dilution 1:32 (Con A Concanavalin no C-2010 type IV)11 final concentration 60 µg/ml. µg/ml.
Standard Test Antigens. Purified protein derivate (PPD, RIVM), final concentration 12 µg/ml; tetanus toxoid (Tet, RIVM),' final concentration 18 LF/ml; and diphtheria toxoid (Dipth, RIVM),1 final concentration 18 LF/ml. Isolation of Mononuclear Cells. Peripheral blood lymphocytes (PBL) were isolated from heparinized blood by
Ficoll-Isopaque density layer centrifugation. Mononuclear cell suspensions were also prepared from heparinized umbilical cord blood lymphocytes (UCL), sampled after normal deliveries from uncomplicated pregnancies. Culture Procedures All cultures were performed in round-bottom microtitre plates. Lymphocytes were cultured in Iscove's modified Dulbecco's medium (IMDM) supplemented with 20% heat-inactivated pooled human serum and standard antibiotics. The culture plates were incubated at 37°C in a humidified 95% air, 5% C02 atmosphere. The cultures with B. intermedius and B. gingivalis were incubated for 5 to 9 days. Cultures with PHA, Con A, and ALS were incubated for 4 days and those with PPD, Tet, and Dipth for 6 days. Proliferative responses were determined by measuring the incorporation of 3H-thymidine (0.4 µ / ^ : activity 200 Ci/mol) added 24 hours before harvest of the cultures. The cells were harvested onto glass fiber filters and radioactivity was measured with a liquid scintillation counter. All cultures were performed in triplicate. Variability among triplicate values was always less than 10%.
Stimulants
Preparations of Oral Bacteria. B. intermedius strain HG110
50
J Periodontol April 1990
RABER-DURLACHER, ZEIJLEMAKER, MEINESZ, ABRAHAM-INPJJN
PHA
nal concentration 12
*Gibco Laboratories, Grand Island, NY. 5Wellcorae Diagnostics, Dartford, UK. "Sigma, St. Louis, Mo.
Limiting Dilution Analysis To determine the actual number of lymphocytes initially responding to oral bacterial antigens, a limiting dilution analysis was performed. To this end, graded numbers of lymphoid cells (from 105 cells/well to 780 cells/well in a series of two-fold dilutions) were cultured in round-bottom microtiter plates. The cell number in all cultures was made up to 105 per well by the addition of feeder cells. The feeder cells were autologous PBLs, which had been irradiated with a dose of 7000 rad. At each cell number, 48 replicate cultures were performed. All cultures were stimulated with an optimal dose of bacterial extract as antigen. Cultures were
incubated for 8 days and the 3H-thymidine incorporation assessed. A culture was scored positive when the thymidine incorporation exceeded 150 cpm. The data were analyzed by means of Poisson statistics. The fraction of negative wells (Fo) was plotted on a semilog plot against the number of lymphocytes at each cell number. Straight lines were obtained, as assessed by the least squares method. This indicates single-hit kinetics: one cell type is limiting. From these plots, the number of cells containing one limiting (antigen-reactive) cell is obtained by determining the cell number at which Fo 0.37.9 was
=
'RIVM, Bilthoven, The Netherlands.
Volume 61 Number 4
Separation of
IN VITRO LYMPHOCYTE STIMULATION BY BACTEROIDES
and
219
cpmxIO3
Cells
35t
and non-T lymphocytes were separated by -rosette sedimentation, using neuraminidase-treated sheep erythrocytes. Interphase cells, depleted of cells, usually contained 30% lymphocytes, 30% non-T, non-B lymphocytes, and 40% monocytes. Contamination with -rosette forming cells was always less than 5%.10 By means of differential counts, it was determined that the cell fraction always contained less than 1% monocytes. Activation of
Cells to
IgM
Production
Lymphocytes were cultured in round-bottom microtiter plates in
a
was
final volume of 170
µ per well. The culture medium IMDM, supplemented with 10% FCS and standard
antibiotics. The unseparated PBLs were cultured at cell numbers of 103 lymphocytes per well. Non-T cells 103 to 160 20 cultured were together with different numbers (50 103) 103 to 80 of irradiated (dose: 4000 rad) cells (10 IO3 per well). The cultures were stimulated with Pokeweed mitogen and with sonicates of B. intermedius and B. gingivalis. Culture plates were incubated for 7 days at 37°C, in a humidified 95% air, 5% C02 atmosphere. Cultures were set up in quadruplicate wells for Immunoglobulin production. These cultures were terminated by collecting the supernatants of the four wells. The amount of IgM produced was determined by ELISA. In each assay a standard curve was obtained by using five different dilutions of a normal human serum with a known concentration of IgM. The amounts of IgM in supernatants of cultured lymphocytes were expressed as nanograms per culture well. The proliferative responses of the PBL, cells and non-T cells, stimulation with sonicates of B. intermedius and B. gingivalis were determined by the incorporation of 3H thymidine. Harvesting of cells and counting of incorporated radioactivity were performed as described.
I: Estimation of optimal antigen concentration o/B. intermedius (B!) and B. gingivalis (BG) bacterial extracts. Both donors respond to B. gingivalis, but one does not respond to B. intermedius. As to the antigen doses, the optimum is rather broad. A concentration of about 1:256 appears to be optimal. However, at the concentration routinely used (1:64), a positive response (if present ) is not missed.
Figure
BACTEROIDES INTERMEDIUS
200*10 ly
RESULTS 100*10
Assessment of Conditions and PBL Responses Lymphocytes were isolated from heparinized venous blood from two individuals to assess optimal culture conditions. PBL were cultured in the presence of different antigen concentrations (Fig. 1), during various culture periods and at different numbers of lymphocytes per well (Figs. 2 and 3). As a positive control the lymphocytes were cultured in the
presence of PHA. This led to 30xl03 cpm and 24xl03 cpm for the two donors. Optimal antigen concentrations were estimated using serial dilutions of the bacterial extracts. From data shown in Figure 1 it can be concluded that both B. intermedius and B. gingivalis sonicates can be diluted to 1:256 to obtain optimal lymphocyte responses. The dose-response curve reveals a rather broad range. From experiments using different numbers of lymphocytes (Figs. 2 and 3), it can be concluded that relatively high numbers of lymphocytes (200
ly
80*10
ly
40*10
ly
days
Figure 2: Time course of the lymphocyte proliferative response to B. intermedius. Different numbers of lymphocytes from one donor were measured for 5 to 9 days as shown.
220
RABER-DURLACHER, ZEIJLEMAKER, MEINESZ, ABRAHAM-INPLJN
cpm*10
BACTEROIDES GINGIVALIS
25-
20-
200-10
ly
15-
10H
100-103ly
.80-1031 y
-40-10 ly
days
Figure 3: Time course of the lymphocyte proliferative response to B. gingivalis. Different numbers of lymphocytes from one donor were cultured from 5 to 9 days as shown.
103/well) are required to obtain optimal responses when these cells are cultured in the presence of B. intermedius and B. gingivalis. It can also be concluded from data shown in Figures 2 and 3 that, at 200 x 103 cells/well, the responses were optimal on day 8. PBL of 12 individuals were cultured under these conditions in the presence of B. intermedius and B. gingivalis, to determine whether proliferation to these periodontal pathogens did occur in this periodontally healthy population. PBL of all subjects did show proliferation to these antigens; responses to B. intermedius were higher than to B. gingivalis (data not shown). Also in Figures 2 and 3, the response to B. intermedius appears to be higher than the response to B. gingivalis. Cultures of Umbilical Cord Blood Lymphocytes Umbilical cord blood lymphocytes (UCL) are thought to be relatively immunologically naive cells.11 Therefore, we decided to determine whether UCL would display a proliferative response to B. intermedius and B. gingivalis preparations. If these bacterial antigens would act as polyclonal activators, they should be able to induce proliferation of UCL exceeding background stimulation values.
J Periodontol
April 1990
The results shown in Table 1 indicate that B. intermedius and B. gingivalis antigen preparations apparently do not act as polyclonal activators. From these data it is clear that the UCL are strongly stimulated by the non-specific mitogens PHA, ALS, and Con A, (response measured at day 3 of culture). When cells were cultured in the presence of specific antigens (PPD, tetanus toxoid, diphtheria toxoid) the 3H-thymidine incorporation was assessed at day 6 of culture. It can be seen that at day 6 the UCL show a considerable background 3Hthymidine incorporation in the absence of any added stimulant. When cultured in the presence of the antigens, the 3H-thymidine incorporation did not exceed this background value. When UCL were cultured in the presence of B. intermedius and B. gingivalis antigens, 3H-thymidine incorporation was assessed at day 8 of culture (Table 1). At that time, background 3H thymidine incorporation was still higher than at day 6. In cultures in the presence of B. intermedius and B. gingivalis antigens 3H thymidine incorporation never exceeded background values.
Estimation of Antigen-Reactive Lymphocytes To investigate whether the lymphocyte proliferative response to a preparation of B. intermedius is antigen-specific (as opposed to polyclonal), limiting dilution experiments were performed as outlined in the Materials and Methods section above. The results of one experiment are given in Figure 4. From these data it can be seen that individual culture wells differ greatly in their 3H-thymidine incorporation. At each cell number the proportion of negative (nonresponding) cultures (i.e., with a 3H-thymidine incorporation of less than 150 cpm) was assessed. From this, the proportion of negative cultures was obtained for each set of 48 replicate cultures. These data were plotted in a semilogarithmic plot against the number of non-irradiated lymphocytes per cultures (Fig. 5). According to Poisson statistics the finding of a straight line in this plot indicates singlehit kinetics; i.e., only one cell type is limiting under the conditions used. From this graph, fitted by the least squares method, the number of cells was determined at which a fraction 0.37 of the cultures were non-responding (Fig. 5). The data from three experiments with lymphocytes from three different donors are shown in Table 2. Role of and Non-T Cells in the Response to B. intermedius and B. gingivalis antigens In order to assess the role of and lymphocytes, and non-T cells were separated and their proliferative responses analyzed. PBLs were obtained from three periodontally and otherwise healthy donors and cells were separated as described above. The interphase cells, depleted of cells, contained lymphocytes, non-T/non-B cells, and monocytes. Responses to PWM and B. intermedius and B. gingivalis appeared to be monocyte dependent. The cell populations cultured in the absence of antigenpresenting cells did not show significant proliferation. When
Volume 61 Number 4
IN VITRO LYMPHOCYTE STIMULATION BY BACTEROIDES
Table 1: Proliferative
Response of UCL to
Several
Mitogens
and
Antigens
Day 3 cord blood lymphocytes 2 10s cells/well
Experiment
A
C D E
221
Dipht
neg
8 1:64 BI
1798 8723 3641 1773
3093 3302 5545 9318 5291
3936 5724 6090 8952 6309
Day
Day 6
neg
PHA
ALS
Con A
neg
PPD
Tet
357 501 1318 972 734
8213 44885 32507 10696 29629
5583 55880 46528 13099 18692
21506 15736 42138 38626 27268
1640 2194 8255 3304 2236
1100 1822 5748 3974 1804
1361 1933
Experiments A through E were performed with UCL obtained from different deliveries.
NT
=
8746 5595 3040
1:64 BG
2936 4105 3638 2320 NT
not tested.
cpm*103
25x103ly
20
12-
1110· 9-
876-
Figure 5: Statistical evaluation of the limiting dilution experiment of Figure 4. Experimental data plotted according to Poisson analysis.
5-
Frequency of Reactive Units in Limiting Dilution Experiments Performed on Peripheral Blood Lymphocytes Obtained Table 2: The
From 3 Donors Donor A C
0^ 0
781
1562
.
3125
""" 6250
12500
25000
lymphocytes
Figure 4: Lymphocyte proliferative response to B. intermedius under limiting dilution culture conditions. Experimental set-up as described in Materials and Methods. Each point represents an individual culture well. irradiated non-T cell fractions containing substantial numbers of monocytes were added to the cultures, proliferation did occur, as measured by the 3H thymidine incorporation. Proliferation of the non-T cells when cultured in the presence of B. intermedius and B. gingivalis is shown in Table 3. Since the responses never exceeded background values, there appeared to be little or no contribution of cells in the proliferative response to B. intermedius and B. gingivalis. In other experiments, cell differentiation was measured by means of IgM production. At termination of the cultures, supernatants were assayed for IgM by means of an ELISA.
Frequency of Reactive
Units
1:4000 1:6800 1:8200
Preliminary
studies showed that optimal IgM production occurred on day 7. Therefore, all culture supernatants were tested after 7 days of incubation. Data obtained at the maximum stimulatory dose of activator are shown in Table 4. A response of cells to antigens requires the presence of antigen presenting cells (APC). Indeed, in our experiments cells by themselves did not proliferate in the presence of B. intermedius and B. gingivalis, except for one instance stimulated by B. intermedius. DISCUSSION The present studies indicate that the in vitro stimulation of human peripheral blood lymphocytes by preparations of B. intermedius and B. gingivalis reflects an antigen-specific cell response.The question whether preparations of bacteria induce an antigen-specific response, an aspecific polyclonal response, or both specific and aspecific lymphoblastic in
Table 3: Role of - and Non-T
Lymphocytes
in the Proliferative
3H thymidine incorporation (cpm)
Lymphocytes PBL non-T
BG
4698 215 2979 4994
non-T +
Table 4: Introduction of
IgM production (mg/well) Number of Lymphocytes PBL
Non
20xl03 40 80 lOxlO3 20 40 80
100
TR"
=
BI
BG
1778 261 444 1043
5933 845 1790 6472
2024 306 1095 2263
IO3
neg 44 120 881 988
Donor A PMW BI 186 6 16 996 1627 1963 1245 1757
977 3693 4272
75 153 75 195
55 94 187 563
3430 4290 4803 5035
67 72 196 285
6 761 1430 2129
75
2642
75
120
BG 183
(BI)
and B.
gingivalis (BG).
_
Cell Differentiation by B. intermedius (BI) and B. gingivalis in the presence of B. intermedius (BI) and B. gingivalis (BG)
160
NonT50xl034-TR"
to B. intermedius
neg
_
903 85 503 995
Response
_Donor
_Donor A BI
_Donor C
neg
BI
BG
neg
1762 353 1086 2500
4941 608 2341 6050
1150 944 1099 3136
950 1273 1706 417
(BG).
Donor PWM 75 342 631 915
BI 75 183 191 191
BG 75 75 155 208
neg 75 81 291 1362
619 463 1166 1071
1513 1185 1139 1672
75 75 132 245
75 163 88 126
264 183 1147 2149
75
1750
75
121
535
Donor C PWM BI 600 75 75 806 254 2174 4174 1138
BG 600 600 600 656
60000
75 155 227 560
122 168 223 328
38923
412
544
14682 17696 43400
cells irradiated with 4000 rad.
vitro lymphocyte responses may be addressed in different ways: 1. Following the papers published by Ivanyi and colleagues12'13 in the early 70s, there have been attempts to relate the periodontal status and in vitro proliferative responses to oral bacteria. It was postulated that a direct correlation between these parameters would indicate antigen-specificity, whereas the lack of such a relationship would indicate non-specificity. A preparation of B. gingivalis elicited statistically more frequent positive responses in patients with destructive Periodontitis as compared to subjects with
gingivitis, suggesting an antigen-specific blastogenic response to this periodontally pathogenic microorganism.1 However, in vitro blastogenic PBL responses to mitogens and to preparations of oral bacteria show great inter- and
intra-individual variations. Therefore, it is difficult to draw conclusions from PBL blastogenic responses and clinical data with respect to the specific versus aspecific nature of the response. Nevertheless, the individual variability of the responses to B. intermedius and B. gingivalis provides an argument in favor of the antigen specific nature of response: some donors respond strongly to these two antigens, whereas others respond only to one of these two antigens (Fig. 1). Such data were obtained in a group of 10 other donors (data not
J Periodontol April 1990
RABER-DURLACHER, ZEIJLEMAKER, MEINESZ, ABRAHAM-INPJJN
222
shown).
2. An alternative approach to study the issue of antigen specificity is the use of human umbilical cord blood lymphocytes (UCL) which can be regarded immunologically as a relatively naive population. In our study UCL showed an appreciable background proliferation in non-stimulated cultures; we were unable to detect any proliferative response
to B. intermedius and B.
gingivalis, using UCL from five different donors. It has been suggested that the non-responsiveness of UCL might be due to the presence of a high proportion of suppressor cells,11,14 and that the cells might respond at lower numbers.2 However, the UCL we tested responded well to polyclonal stimulants such as PHA, Con A, and ALS, but not to B. intermedius and B. gingivalis. Although we did not characterize the UCL, these data argue against B. intermedius and B. gingivalis as polyclonal activators. 3. The question whether the lymphocyte proliferative response to antigenic extracts of oral bacteria is specific or polyclonal was also addressed by means of limiting dilution experiments. Our results show that only a small proportion of lymphocytes determines the reactivity to the bacterial antigens of the sonicate of B. intermedius; in different individuals we measured frequencies of reactive cells of 1:4000, 1:6800, and 1:8200. This finding strongly argues in favor of the antigen specific nature of the response. If a bacterial extract would merely act as a polyclonal stimulator, a much higher frequency of reactive cells was to be expected.15 Unfortunately, we have been unable to perform limiting dilution experiments with B. gingivalis, due to a lack of donor cells. Together with the fact the bacterial extracts were unable to increase the proliferation of UCL, our findings indicate antigen-specificity of the responses. 4. The antigen-specific responding cells in the. limiting dilution experiments are very likely to be cells. To ascertain this, we analyzed the reactivity of separated T- and non-T cells to preparations of B. intermedius and B. gingivalis. Proliferation of lymphocytes cultured in the près-
Volume 61 Number 4
of these antigens appeared to be strictly dependent on monocytes and cells. This is in agreement with the results of Stashenko et al.,16 who have reported that separated cells in the presence of a well-defined number of monocytes responded to a preparation of B. gingivalis. In addition to the assessment of the nature and the cell types involved in the proliferative PBL responses to preparations of B. intermedius and B. gingivalis, it is relevant cell to analyze the nature of the cell types involved in activation to proliferation and differentiation. Antigen-specific cells are capable of fulfilling a helper function for primed cells, resulting in the production of specific antibodies. However, antigen-specific stimulation of cells may also be followed by non-specific polyclonal activation of lymphocytes to the production of Immunoglobulins. cell independent cell Another possibility would be stimulation. It should be realized that bacterial cell walls cell activation are complex antigens, which may induce to different With the requirement of respect by pathways. cells in the culture system, our results are in agreement with those of Carpenter et al.4 These investigators reported that immunoglobin synthesis when lymphocytes were cultured in the presence of B. intermedius (formerly B. melaninogenicus subspecies intermedius) did not occur in the absence of cells. In their experiments the polyclonal nature of the Immunoglobulin response to Actinobacillus actinomycetemcomitans was clearly demonstrated by means of adsorption experiments. However, such experiments were not performed to investigate the nature of the antibody response to preparations of other oral bacteria. In studies using the plaque-forming cell assay to assess polyclonal antibody production, B. intermedius and B. gingivalis appeared to be rather weak polyclonal activators as compared to other oral microorganisms.3 In our studies we did not determine the specificity and the subclass distribution of the antibodies produced in vitro as a result of peripheral lymphocyte stimulation with preparations of B. intermedius and B. gingivalis. However, there is convincing evidence from recent literature that systemic antibodies to B. intermedius and B. gingivalis in patients infected with these microorganisms are the result of specific stimulation of the immune system.17 Taken together, our data provide support for the notion that antigen preparations of B. intermedius and B. gingivalis are capable of inducing specific lymphocyte activation, followed by cell dependent IgM production by lymphocytes. enee
Acknowledgements
The authors thank Dr. C. de Groot for his critical comments, Dr. J. C. Dijkstra for his help in obtaining human umbilical cord blood, and Ms. Marijke Roos for skillful
IN VITRO LYMPHOCYTE STIMULATION BY BACTEROIDES
223
technical assistance. The editorial assistance of Ms. Stephanie Weinreich is gratefully acknowledged. This study was supported by the Dutch Praeventiefonds grant 28-1176. REFERENCES 1. Patters MR, Chen P, McKenna J, Genco RJ. Lymphoproliferative 2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
responses to oral bacteria in humans with varying severities of periodontal disease. Infect Immun 1980; 28:777. Donaldson SL, Ranney RR, Tew JG. Evidence of mitogenic activity in periodontitis-associated bacteria, infect Immun 1983; 42:487. Bick PH, Carpenter AB, Holdeman LV, et al. Polyclonal B-cell activation induced by extracts of Gram-negative bacteria isolated from periodontally diseased sites. Infect Immun 1981; 34:43. Carpenter AB, Scully EC, Ranney RR, PH Bick. T-cell regulation of polyclonal B-cell activation induced by extracts of oral bacteria associated with periodontal diseases. Infect Immun 1984; 43:326. Van der Veiden U. Probing force and the relationship of the probe tip to the periodontal tissues. J Clin Periodontol 1979; 6:106. Van Winkelhoff AJ, Carlee AW, de Graaff J. Bacteroides endodontalis and other black-pigmented Bacteroides species in odontogenic abscesses. Infect Immun 1985; 49:494. Baker JJ, Tondreau SP. The stimulation of human peripheral blood lymphocytes by oral bacteria: Macrophage and T-cell dependence. / Dent Res 1985; 64:909. Redinbaugh MJ, Turley RB. Adaptation of the bicinchoninic acid protein assay for use with microtiter plates and sucrose gradient fractions. Anal Biochem 1986; 153:267. Waldmann H, Cobbold S, Lefkovits I. Limiting dilution analysis. In: Klaus GGB ed. Lymphocytes, a Practical Approach, Oxford: IRL Press; 1987:163-188. Van Oers MHJ, Zeijlemaker WP, Schellekens PTA. Separation properties of EA-rosette forming lymphocytes in humans. Eur J Immunol 1977; 7:143. Chen P, Doroszczak N. Differences in lymphoproliferative responses to the bacterium Actinomyces viscosus in various mammalian species. Arch OralBiol 1982; 27:319. Ivanyi L, Lehner T. Stimulation of lymphocyte transformation by bacterial antigens in patients with periodontal disease. Arch Oral Biol
1970; 15:1089. 13.
14.
15.
16.
17.
Wilton JMA, Lehner T. Cell-mediated immunity in periodontal disease: cytotoxicity, migration inhibition and lymphocyte transformation studies. Immunology 1972; 22:141. Baker JJ, Tondreau SP. Solubilized dental plaque is mitogenic for nylon wool-purified human cord blood lymphocytes. / Periodont Res 1987; 22:94-122. Moretta A, Pantaleo G, Moretta L, Mingari MC, Cerottini JC. Quantitative assessment of the pool size and subset distribution of cytolytic lymphocytes within human resting or alloactivated peripheral blood cell populations. / Exp Med 1983; 158:571. cell reStashenko P, Resmini LM, Haffajee AD, Socransky SS. sponses of periodontal disease patients and healthy subjects to oral microorganisms. / Periodont Res 1983; 18:587. Ebersole JL, Holt SC. Serum antibodies to periodontopathic microorganisms: Specific induction. In: Periodontology Today. International Congress. Guggenheim, B, ed. Zürich. 169, Basel, Karger, 1988.
Ivanyi L,
Send reprint request to: Dr. J.E. Raber-Durlacher, Department of General Pathology and Internal Medicine, Academic Center for Dentistry Amsterdam, (ACTA), Louwesweg 1, 1066 EA Amsterdam, The Netherlands. Accepted for publication October 14, 1989.
Stimulation of Lymphocytes in Vitro by Bacteroides intermedius and Bacteroides (Porphyromonas) gingivalis Sonicates J.E. Raber-Durlacher* W.P.
Zeijlemaker,t A.A.P. Meinesz,* L. Abraham-Inpijn*
The present study was designed to assess whether the in vitro stimulation of lymphocytes by sonicates of Bacteroides intermedius and Bacteroides (Porphyromonas) gingivalis is antigen specific or non-specific. In addition, the role of and lymphocytes in these responses was assessed. Peripheral blood lymphocytes obtained from healthy volunteers were cultured in the presence of these bacterial preparations and the proliferative response was measured. In similar experiments the response of umbilical cord blood lymphocytes did not exceed background values. In limiting dilution experiments only 1:4000, 1:6800, and 1:8200 of the lymphocytes initially reacted to B. intermedius, which strongly argues for the antigen-specificity of the response. Purified cells, in the presence of monocytes, proliferated when stimulated with B. intermedius and B. gingivalis. As for cell stimulation, the bacterial extracts were capable of inducing IgM production, which appeared to be cell dependent. These findings support the notion that B. intermedius and B. gingivalis induce specific cell activation; secondarily, a cell activation may occur. J Periodontol 1990; 61:217cell dependent, polyclonal 223.
Key Words: IgM; T-cell activation; B-cell activation; Bacteroides intermedius; Bacteroides
(Porphyromonas) gingivalis; lymphocytes
performed on the antigenlymphonon-specific specific cyte response to oral bacteria, preparations of Bacteroides gingivalis and/or Bacteroides intermedius have been used as stimulants. The latter microorganisms are considered to play an important role in the etiology of periodontal diseases. Consequently, defining host responses to these particular bacteria may contribute to knowledge of the pathogenesis of these diseases. Studies of the in vitro blastogenic response of whole peripheral blood lymphocytes (PBL) to isolates of B. asaccharolyticus (now classified as B. gingivalis) have suggested antigen specificity.1 On the other hand, bacterial preparations from B. gingivalis and B. intermedius were reported to possess the capacity to induce mitogenesis in germ-free mouse spleen cells, human umbilical cord blood lymphocytes (UCL), and PBL from periodontally healthy adults.2 With regard to the capacities of extracts of B. intermedius and B. gingivalis to induce a polyclonal antibody In
a
few studies that have been versus
nature of the in vitro
'Department of General Pathology and Internal Medicine, Academic Center for Dentistry Amsterdam, (ACTA), Amsterdam, The Netherlands. tDepartment of Clinical Immunobiology and Hybridoma Laboratory, Central Laboratory of the Netherlands Red Cross Bloodtransfusion Service Amsterdam, The Netherlands.
response in cultures of PBL, these particular bacteria appeared to be weak activators.3 A number of studies have been performed on separated lymphocyte populations in order to assess the role of cells in the antibody response to oral bacterial antigens. The generation of IgG and IgM responses induced by a preparation of B. intermedius was shown to be strictly dependent upon the presence of cells.4 It was shown that B. intermedius could act as a polyclonal activator for human
lymphocytes, inducing Immunoglobulin production higher than that induced by pokeweed mitogen (PWM). A preparation of B. gingivalis was not included in the panel of stimulants. In view of these inconclusive and apparently conflicting data, the objectives of the present study were to: analyze in vitro PBL reactivity to preparations of B. intermedius and B. gingivalis; determine whether the response is antigen-specific or non-specific; and assess the role of T- and B-lymphocytes in these responses. First, optimal culture conditions and in vitro PBL reactivity to preparations of B. intermedius and B. gingivalis were assessed. Similar experiments were performed on umbilical cord blood lymphocytes to determine whether these stimulants are capable of inducing polyclonal activation in a presumably non-sensitized lymphocyte population. The is-
218
of antigen-specificity was further investigated by means of limiting dilution (LD) experiments. Finally, and nonT cells (B cells) were separated in order to assess their role in the in vitro blastogenic responses to B. intermedius and B, gingivalis. Proliferative responses were measured by 3Hthymidine incorporation and cell activation by means of sue
IgM production.
MATERIALS AND METHODS Human Subjects Twelve volunteers
(5 males and 7 females, aged 24 to 40 years) without clinical or historical features òf Periodontitis were selected (pocket depths ranging from 1.7 mm to 2.0 mm; Periodontal Pocket Bleeding Index, PPBI5 ranging from 0.2-0.3). All subjects were in good health and no medication that could influence immunological responses had been taken for the previous 3 months. After informed consent was secured, 60 ml of venous blood
was obtained from these individuals for the initial experiments, in which optimal culture conditions were assessed and PBL responses to B. intermedius and B. gingivalti were analyzed. After these initial experiments were completed, 3 of the subjects twice donated an additional 200 ml of venous blood for limiting dilution experiments and for the experiments in which the cell populations involved in blastogenic responses and IgM
production were analyzed. Lymphocyte
(ATCC 25611) and B. gingivalis strain HG762 were kindly provided by Dr. TJ.M. van Steenbergen (Department of Oral Microbiology, Academic Center for Dentistry Amsterdam (ACTA)). Bacteria were cultured in liquid BM medium,6 centrifuged, and handled in phosphate buffered saline. Extracts were prepared by freeze-thawing, sonication, and centrifugation of cell wall fragments as described by Baker
and Tondreau.7 Protein concentrations of the bacterial sonicates were measured by means of the Bicinchoninic Acid Protein Assay.8 The protein concentration of the B. intermedius antigen preparation was 0.85 mg/ml; the sonicate of B. gingivalis had a protein concentration of 3.2 mg/ml.
Mitogens. Pokeweed mitogen (PMW)* final concentration
(purified Phytohaemagglutinin HA16)§ fiµg/ml. ALS, horse anti-human lymphocyte serum (CLB) optimal final dilution 1:32 (Con A Concanavalin no C-2010 type IV)11 final concentration 60 µg/ml. µg/ml.
Standard Test Antigens. Purified protein derivate (PPD, RIVM), final concentration 12 µg/ml; tetanus toxoid (Tet, RIVM),' final concentration 18 LF/ml; and diphtheria toxoid (Dipth, RIVM),1 final concentration 18 LF/ml. Isolation of Mononuclear Cells. Peripheral blood lymphocytes (PBL) were isolated from heparinized blood by
Ficoll-Isopaque density layer centrifugation. Mononuclear cell suspensions were also prepared from heparinized umbilical cord blood lymphocytes (UCL), sampled after normal deliveries from uncomplicated pregnancies. Culture Procedures All cultures were performed in round-bottom microtitre plates. Lymphocytes were cultured in Iscove's modified Dulbecco's medium (IMDM) supplemented with 20% heat-inactivated pooled human serum and standard antibiotics. The culture plates were incubated at 37°C in a humidified 95% air, 5% C02 atmosphere. The cultures with B. intermedius and B. gingivalis were incubated for 5 to 9 days. Cultures with PHA, Con A, and ALS were incubated for 4 days and those with PPD, Tet, and Dipth for 6 days. Proliferative responses were determined by measuring the incorporation of 3H-thymidine (0.4 µ / ^ : activity 200 Ci/mol) added 24 hours before harvest of the cultures. The cells were harvested onto glass fiber filters and radioactivity was measured with a liquid scintillation counter. All cultures were performed in triplicate. Variability among triplicate values was always less than 10%.
Stimulants
Preparations of Oral Bacteria. B. intermedius strain HG110
50
J Periodontol April 1990
RABER-DURLACHER, ZEIJLEMAKER, MEINESZ, ABRAHAM-INPJJN
PHA
nal concentration 12
*Gibco Laboratories, Grand Island, NY. 5Wellcorae Diagnostics, Dartford, UK. "Sigma, St. Louis, Mo.
Limiting Dilution Analysis To determine the actual number of lymphocytes initially responding to oral bacterial antigens, a limiting dilution analysis was performed. To this end, graded numbers of lymphoid cells (from 105 cells/well to 780 cells/well in a series of two-fold dilutions) were cultured in round-bottom microtiter plates. The cell number in all cultures was made up to 105 per well by the addition of feeder cells. The feeder cells were autologous PBLs, which had been irradiated with a dose of 7000 rad. At each cell number, 48 replicate cultures were performed. All cultures were stimulated with an optimal dose of bacterial extract as antigen. Cultures were
incubated for 8 days and the 3H-thymidine incorporation assessed. A culture was scored positive when the thymidine incorporation exceeded 150 cpm. The data were analyzed by means of Poisson statistics. The fraction of negative wells (Fo) was plotted on a semilog plot against the number of lymphocytes at each cell number. Straight lines were obtained, as assessed by the least squares method. This indicates single-hit kinetics: one cell type is limiting. From these plots, the number of cells containing one limiting (antigen-reactive) cell is obtained by determining the cell number at which Fo 0.37.9 was
=
'RIVM, Bilthoven, The Netherlands.
Volume 61 Number 4
Separation of
IN VITRO LYMPHOCYTE STIMULATION BY BACTEROIDES
and
219
cpmxIO3
Cells
35t
and non-T lymphocytes were separated by -rosette sedimentation, using neuraminidase-treated sheep erythrocytes. Interphase cells, depleted of cells, usually contained 30% lymphocytes, 30% non-T, non-B lymphocytes, and 40% monocytes. Contamination with -rosette forming cells was always less than 5%.10 By means of differential counts, it was determined that the cell fraction always contained less than 1% monocytes. Activation of
Cells to
IgM
Production
Lymphocytes were cultured in round-bottom microtiter plates in
a
was
final volume of 170
µ per well. The culture medium IMDM, supplemented with 10% FCS and standard
antibiotics. The unseparated PBLs were cultured at cell numbers of 103 lymphocytes per well. Non-T cells 103 to 160 20 cultured were together with different numbers (50 103) 103 to 80 of irradiated (dose: 4000 rad) cells (10 IO3 per well). The cultures were stimulated with Pokeweed mitogen and with sonicates of B. intermedius and B. gingivalis. Culture plates were incubated for 7 days at 37°C, in a humidified 95% air, 5% C02 atmosphere. Cultures were set up in quadruplicate wells for Immunoglobulin production. These cultures were terminated by collecting the supernatants of the four wells. The amount of IgM produced was determined by ELISA. In each assay a standard curve was obtained by using five different dilutions of a normal human serum with a known concentration of IgM. The amounts of IgM in supernatants of cultured lymphocytes were expressed as nanograms per culture well. The proliferative responses of the PBL, cells and non-T cells, stimulation with sonicates of B. intermedius and B. gingivalis were determined by the incorporation of 3H thymidine. Harvesting of cells and counting of incorporated radioactivity were performed as described.
I: Estimation of optimal antigen concentration o/B. intermedius (B!) and B. gingivalis (BG) bacterial extracts. Both donors respond to B. gingivalis, but one does not respond to B. intermedius. As to the antigen doses, the optimum is rather broad. A concentration of about 1:256 appears to be optimal. However, at the concentration routinely used (1:64), a positive response (if present ) is not missed.
Figure
BACTEROIDES INTERMEDIUS
200*10 ly
RESULTS 100*10
Assessment of Conditions and PBL Responses Lymphocytes were isolated from heparinized venous blood from two individuals to assess optimal culture conditions. PBL were cultured in the presence of different antigen concentrations (Fig. 1), during various culture periods and at different numbers of lymphocytes per well (Figs. 2 and 3). As a positive control the lymphocytes were cultured in the
presence of PHA. This led to 30xl03 cpm and 24xl03 cpm for the two donors. Optimal antigen concentrations were estimated using serial dilutions of the bacterial extracts. From data shown in Figure 1 it can be concluded that both B. intermedius and B. gingivalis sonicates can be diluted to 1:256 to obtain optimal lymphocyte responses. The dose-response curve reveals a rather broad range. From experiments using different numbers of lymphocytes (Figs. 2 and 3), it can be concluded that relatively high numbers of lymphocytes (200
ly
80*10
ly
40*10
ly
days
Figure 2: Time course of the lymphocyte proliferative response to B. intermedius. Different numbers of lymphocytes from one donor were measured for 5 to 9 days as shown.
220
RABER-DURLACHER, ZEIJLEMAKER, MEINESZ, ABRAHAM-INPLJN
cpm*10
BACTEROIDES GINGIVALIS
25-
20-
200-10
ly
15-
10H
100-103ly
.80-1031 y
-40-10 ly
days
Figure 3: Time course of the lymphocyte proliferative response to B. gingivalis. Different numbers of lymphocytes from one donor were cultured from 5 to 9 days as shown.
103/well) are required to obtain optimal responses when these cells are cultured in the presence of B. intermedius and B. gingivalis. It can also be concluded from data shown in Figures 2 and 3 that, at 200 x 103 cells/well, the responses were optimal on day 8. PBL of 12 individuals were cultured under these conditions in the presence of B. intermedius and B. gingivalis, to determine whether proliferation to these periodontal pathogens did occur in this periodontally healthy population. PBL of all subjects did show proliferation to these antigens; responses to B. intermedius were higher than to B. gingivalis (data not shown). Also in Figures 2 and 3, the response to B. intermedius appears to be higher than the response to B. gingivalis. Cultures of Umbilical Cord Blood Lymphocytes Umbilical cord blood lymphocytes (UCL) are thought to be relatively immunologically naive cells.11 Therefore, we decided to determine whether UCL would display a proliferative response to B. intermedius and B. gingivalis preparations. If these bacterial antigens would act as polyclonal activators, they should be able to induce proliferation of UCL exceeding background stimulation values.
J Periodontol
April 1990
The results shown in Table 1 indicate that B. intermedius and B. gingivalis antigen preparations apparently do not act as polyclonal activators. From these data it is clear that the UCL are strongly stimulated by the non-specific mitogens PHA, ALS, and Con A, (response measured at day 3 of culture). When cells were cultured in the presence of specific antigens (PPD, tetanus toxoid, diphtheria toxoid) the 3H-thymidine incorporation was assessed at day 6 of culture. It can be seen that at day 6 the UCL show a considerable background 3Hthymidine incorporation in the absence of any added stimulant. When cultured in the presence of the antigens, the 3H-thymidine incorporation did not exceed this background value. When UCL were cultured in the presence of B. intermedius and B. gingivalis antigens, 3H-thymidine incorporation was assessed at day 8 of culture (Table 1). At that time, background 3H thymidine incorporation was still higher than at day 6. In cultures in the presence of B. intermedius and B. gingivalis antigens 3H thymidine incorporation never exceeded background values.
Estimation of Antigen-Reactive Lymphocytes To investigate whether the lymphocyte proliferative response to a preparation of B. intermedius is antigen-specific (as opposed to polyclonal), limiting dilution experiments were performed as outlined in the Materials and Methods section above. The results of one experiment are given in Figure 4. From these data it can be seen that individual culture wells differ greatly in their 3H-thymidine incorporation. At each cell number the proportion of negative (nonresponding) cultures (i.e., with a 3H-thymidine incorporation of less than 150 cpm) was assessed. From this, the proportion of negative cultures was obtained for each set of 48 replicate cultures. These data were plotted in a semilogarithmic plot against the number of non-irradiated lymphocytes per cultures (Fig. 5). According to Poisson statistics the finding of a straight line in this plot indicates singlehit kinetics; i.e., only one cell type is limiting under the conditions used. From this graph, fitted by the least squares method, the number of cells was determined at which a fraction 0.37 of the cultures were non-responding (Fig. 5). The data from three experiments with lymphocytes from three different donors are shown in Table 2. Role of and Non-T Cells in the Response to B. intermedius and B. gingivalis antigens In order to assess the role of and lymphocytes, and non-T cells were separated and their proliferative responses analyzed. PBLs were obtained from three periodontally and otherwise healthy donors and cells were separated as described above. The interphase cells, depleted of cells, contained lymphocytes, non-T/non-B cells, and monocytes. Responses to PWM and B. intermedius and B. gingivalis appeared to be monocyte dependent. The cell populations cultured in the absence of antigenpresenting cells did not show significant proliferation. When
Volume 61 Number 4
IN VITRO LYMPHOCYTE STIMULATION BY BACTEROIDES
Table 1: Proliferative
Response of UCL to
Several
Mitogens
and
Antigens
Day 3 cord blood lymphocytes 2 10s cells/well
Experiment
A
C D E
221
Dipht
neg
8 1:64 BI
1798 8723 3641 1773
3093 3302 5545 9318 5291
3936 5724 6090 8952 6309
Day
Day 6
neg
PHA
ALS
Con A
neg
PPD
Tet
357 501 1318 972 734
8213 44885 32507 10696 29629
5583 55880 46528 13099 18692
21506 15736 42138 38626 27268
1640 2194 8255 3304 2236
1100 1822 5748 3974 1804
1361 1933
Experiments A through E were performed with UCL obtained from different deliveries.
NT
=
8746 5595 3040
1:64 BG
2936 4105 3638 2320 NT
not tested.
cpm*103
25x103ly
20
12-
1110· 9-
876-
Figure 5: Statistical evaluation of the limiting dilution experiment of Figure 4. Experimental data plotted according to Poisson analysis.
5-
Frequency of Reactive Units in Limiting Dilution Experiments Performed on Peripheral Blood Lymphocytes Obtained Table 2: The
From 3 Donors Donor A C
0^ 0
781
1562
.
3125
""" 6250
12500
25000
lymphocytes
Figure 4: Lymphocyte proliferative response to B. intermedius under limiting dilution culture conditions. Experimental set-up as described in Materials and Methods. Each point represents an individual culture well. irradiated non-T cell fractions containing substantial numbers of monocytes were added to the cultures, proliferation did occur, as measured by the 3H thymidine incorporation. Proliferation of the non-T cells when cultured in the presence of B. intermedius and B. gingivalis is shown in Table 3. Since the responses never exceeded background values, there appeared to be little or no contribution of cells in the proliferative response to B. intermedius and B. gingivalis. In other experiments, cell differentiation was measured by means of IgM production. At termination of the cultures, supernatants were assayed for IgM by means of an ELISA.
Frequency of Reactive
Units
1:4000 1:6800 1:8200
Preliminary
studies showed that optimal IgM production occurred on day 7. Therefore, all culture supernatants were tested after 7 days of incubation. Data obtained at the maximum stimulatory dose of activator are shown in Table 4. A response of cells to antigens requires the presence of antigen presenting cells (APC). Indeed, in our experiments cells by themselves did not proliferate in the presence of B. intermedius and B. gingivalis, except for one instance stimulated by B. intermedius. DISCUSSION The present studies indicate that the in vitro stimulation of human peripheral blood lymphocytes by preparations of B. intermedius and B. gingivalis reflects an antigen-specific cell response.The question whether preparations of bacteria induce an antigen-specific response, an aspecific polyclonal response, or both specific and aspecific lymphoblastic in
Table 3: Role of - and Non-T
Lymphocytes
in the Proliferative
3H thymidine incorporation (cpm)
Lymphocytes PBL non-T
BG
4698 215 2979 4994
non-T +
Table 4: Introduction of
IgM production (mg/well) Number of Lymphocytes PBL
Non
20xl03 40 80 lOxlO3 20 40 80
100
TR"
=
BI
BG
1778 261 444 1043
5933 845 1790 6472
2024 306 1095 2263
IO3
neg 44 120 881 988
Donor A PMW BI 186 6 16 996 1627 1963 1245 1757
977 3693 4272
75 153 75 195
55 94 187 563
3430 4290 4803 5035
67 72 196 285
6 761 1430 2129
75
2642
75
120
BG 183
(BI)
and B.
gingivalis (BG).
_
Cell Differentiation by B. intermedius (BI) and B. gingivalis in the presence of B. intermedius (BI) and B. gingivalis (BG)
160
NonT50xl034-TR"
to B. intermedius
neg
_
903 85 503 995
Response
_Donor
_Donor A BI
_Donor C
neg
BI
BG
neg
1762 353 1086 2500
4941 608 2341 6050
1150 944 1099 3136
950 1273 1706 417
(BG).
Donor PWM 75 342 631 915
BI 75 183 191 191
BG 75 75 155 208
neg 75 81 291 1362
619 463 1166 1071
1513 1185 1139 1672
75 75 132 245
75 163 88 126
264 183 1147 2149
75
1750
75
121
535
Donor C PWM BI 600 75 75 806 254 2174 4174 1138
BG 600 600 600 656
60000
75 155 227 560
122 168 223 328
38923
412
544
14682 17696 43400
cells irradiated with 4000 rad.
vitro lymphocyte responses may be addressed in different ways: 1. Following the papers published by Ivanyi and colleagues12'13 in the early 70s, there have been attempts to relate the periodontal status and in vitro proliferative responses to oral bacteria. It was postulated that a direct correlation between these parameters would indicate antigen-specificity, whereas the lack of such a relationship would indicate non-specificity. A preparation of B. gingivalis elicited statistically more frequent positive responses in patients with destructive Periodontitis as compared to subjects with
gingivitis, suggesting an antigen-specific blastogenic response to this periodontally pathogenic microorganism.1 However, in vitro blastogenic PBL responses to mitogens and to preparations of oral bacteria show great inter- and
intra-individual variations. Therefore, it is difficult to draw conclusions from PBL blastogenic responses and clinical data with respect to the specific versus aspecific nature of the response. Nevertheless, the individual variability of the responses to B. intermedius and B. gingivalis provides an argument in favor of the antigen specific nature of response: some donors respond strongly to these two antigens, whereas others respond only to one of these two antigens (Fig. 1). Such data were obtained in a group of 10 other donors (data not
J Periodontol April 1990
RABER-DURLACHER, ZEIJLEMAKER, MEINESZ, ABRAHAM-INPJJN
222
shown).
2. An alternative approach to study the issue of antigen specificity is the use of human umbilical cord blood lymphocytes (UCL) which can be regarded immunologically as a relatively naive population. In our study UCL showed an appreciable background proliferation in non-stimulated cultures; we were unable to detect any proliferative response
to B. intermedius and B.
gingivalis, using UCL from five different donors. It has been suggested that the non-responsiveness of UCL might be due to the presence of a high proportion of suppressor cells,11,14 and that the cells might respond at lower numbers.2 However, the UCL we tested responded well to polyclonal stimulants such as PHA, Con A, and ALS, but not to B. intermedius and B. gingivalis. Although we did not characterize the UCL, these data argue against B. intermedius and B. gingivalis as polyclonal activators. 3. The question whether the lymphocyte proliferative response to antigenic extracts of oral bacteria is specific or polyclonal was also addressed by means of limiting dilution experiments. Our results show that only a small proportion of lymphocytes determines the reactivity to the bacterial antigens of the sonicate of B. intermedius; in different individuals we measured frequencies of reactive cells of 1:4000, 1:6800, and 1:8200. This finding strongly argues in favor of the antigen specific nature of the response. If a bacterial extract would merely act as a polyclonal stimulator, a much higher frequency of reactive cells was to be expected.15 Unfortunately, we have been unable to perform limiting dilution experiments with B. gingivalis, due to a lack of donor cells. Together with the fact the bacterial extracts were unable to increase the proliferation of UCL, our findings indicate antigen-specificity of the responses. 4. The antigen-specific responding cells in the. limiting dilution experiments are very likely to be cells. To ascertain this, we analyzed the reactivity of separated T- and non-T cells to preparations of B. intermedius and B. gingivalis. Proliferation of lymphocytes cultured in the près-
Volume 61 Number 4
of these antigens appeared to be strictly dependent on monocytes and cells. This is in agreement with the results of Stashenko et al.,16 who have reported that separated cells in the presence of a well-defined number of monocytes responded to a preparation of B. gingivalis. In addition to the assessment of the nature and the cell types involved in the proliferative PBL responses to preparations of B. intermedius and B. gingivalis, it is relevant cell to analyze the nature of the cell types involved in activation to proliferation and differentiation. Antigen-specific cells are capable of fulfilling a helper function for primed cells, resulting in the production of specific antibodies. However, antigen-specific stimulation of cells may also be followed by non-specific polyclonal activation of lymphocytes to the production of Immunoglobulins. cell independent cell Another possibility would be stimulation. It should be realized that bacterial cell walls cell activation are complex antigens, which may induce to different With the requirement of respect by pathways. cells in the culture system, our results are in agreement with those of Carpenter et al.4 These investigators reported that immunoglobin synthesis when lymphocytes were cultured in the presence of B. intermedius (formerly B. melaninogenicus subspecies intermedius) did not occur in the absence of cells. In their experiments the polyclonal nature of the Immunoglobulin response to Actinobacillus actinomycetemcomitans was clearly demonstrated by means of adsorption experiments. However, such experiments were not performed to investigate the nature of the antibody response to preparations of other oral bacteria. In studies using the plaque-forming cell assay to assess polyclonal antibody production, B. intermedius and B. gingivalis appeared to be rather weak polyclonal activators as compared to other oral microorganisms.3 In our studies we did not determine the specificity and the subclass distribution of the antibodies produced in vitro as a result of peripheral lymphocyte stimulation with preparations of B. intermedius and B. gingivalis. However, there is convincing evidence from recent literature that systemic antibodies to B. intermedius and B. gingivalis in patients infected with these microorganisms are the result of specific stimulation of the immune system.17 Taken together, our data provide support for the notion that antigen preparations of B. intermedius and B. gingivalis are capable of inducing specific lymphocyte activation, followed by cell dependent IgM production by lymphocytes. enee
Acknowledgements
The authors thank Dr. C. de Groot for his critical comments, Dr. J. C. Dijkstra for his help in obtaining human umbilical cord blood, and Ms. Marijke Roos for skillful
IN VITRO LYMPHOCYTE STIMULATION BY BACTEROIDES
223
technical assistance. The editorial assistance of Ms. Stephanie Weinreich is gratefully acknowledged. This study was supported by the Dutch Praeventiefonds grant 28-1176. REFERENCES 1. Patters MR, Chen P, McKenna J, Genco RJ. Lymphoproliferative 2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
responses to oral bacteria in humans with varying severities of periodontal disease. Infect Immun 1980; 28:777. Donaldson SL, Ranney RR, Tew JG. Evidence of mitogenic activity in periodontitis-associated bacteria, infect Immun 1983; 42:487. Bick PH, Carpenter AB, Holdeman LV, et al. Polyclonal B-cell activation induced by extracts of Gram-negative bacteria isolated from periodontally diseased sites. Infect Immun 1981; 34:43. Carpenter AB, Scully EC, Ranney RR, PH Bick. T-cell regulation of polyclonal B-cell activation induced by extracts of oral bacteria associated with periodontal diseases. Infect Immun 1984; 43:326. Van der Veiden U. Probing force and the relationship of the probe tip to the periodontal tissues. J Clin Periodontol 1979; 6:106. Van Winkelhoff AJ, Carlee AW, de Graaff J. Bacteroides endodontalis and other black-pigmented Bacteroides species in odontogenic abscesses. Infect Immun 1985; 49:494. Baker JJ, Tondreau SP. The stimulation of human peripheral blood lymphocytes by oral bacteria: Macrophage and T-cell dependence. / Dent Res 1985; 64:909. Redinbaugh MJ, Turley RB. Adaptation of the bicinchoninic acid protein assay for use with microtiter plates and sucrose gradient fractions. Anal Biochem 1986; 153:267. Waldmann H, Cobbold S, Lefkovits I. Limiting dilution analysis. In: Klaus GGB ed. Lymphocytes, a Practical Approach, Oxford: IRL Press; 1987:163-188. Van Oers MHJ, Zeijlemaker WP, Schellekens PTA. Separation properties of EA-rosette forming lymphocytes in humans. Eur J Immunol 1977; 7:143. Chen P, Doroszczak N. Differences in lymphoproliferative responses to the bacterium Actinomyces viscosus in various mammalian species. Arch OralBiol 1982; 27:319. Ivanyi L, Lehner T. Stimulation of lymphocyte transformation by bacterial antigens in patients with periodontal disease. Arch Oral Biol
1970; 15:1089. 13.
14.
15.
16.
17.
Wilton JMA, Lehner T. Cell-mediated immunity in periodontal disease: cytotoxicity, migration inhibition and lymphocyte transformation studies. Immunology 1972; 22:141. Baker JJ, Tondreau SP. Solubilized dental plaque is mitogenic for nylon wool-purified human cord blood lymphocytes. / Periodont Res 1987; 22:94-122. Moretta A, Pantaleo G, Moretta L, Mingari MC, Cerottini JC. Quantitative assessment of the pool size and subset distribution of cytolytic lymphocytes within human resting or alloactivated peripheral blood cell populations. / Exp Med 1983; 158:571. cell reStashenko P, Resmini LM, Haffajee AD, Socransky SS. sponses of periodontal disease patients and healthy subjects to oral microorganisms. / Periodont Res 1983; 18:587. Ebersole JL, Holt SC. Serum antibodies to periodontopathic microorganisms: Specific induction. In: Periodontology Today. International Congress. Guggenheim, B, ed. Zürich. 169, Basel, Karger, 1988.
Ivanyi L,
Send reprint request to: Dr. J.E. Raber-Durlacher, Department of General Pathology and Internal Medicine, Academic Center for Dentistry Amsterdam, (ACTA), Louwesweg 1, 1066 EA Amsterdam, The Netherlands. Accepted for publication October 14, 1989.
