complement is best known and perhaps the most impor­ tant. It has recently been demonstrated by Miller et a l . , M a r y and F o l k e , and Mary, et a l . that human dental plaque chemotactically attracts human P M N cells. Tem­ pel et a l . have reported similar findings using rabbit granulocytes. Concomitant in vivo studies have shown a correlation between plaque accumulation and increasing numbers of neutrophils in the gingival crevices of dogs. The purpose of this investigation was to evaluate the chemotactic characteristics of human polymorphonu­ clear leukocyte cells towards autologous and pooled plaque and towards certain plaque fractions; and to analyze if any difference exists in the chemotactic response to autologous plaque between individuals with a healthy periodontium and individuals with advanced periodontal disease.

Chemotactic Ability of Dental Plaque upon Autologous or Heterologous Human Polymorphonuclear Leukocytes*

15

16

1 7

1 8

19

by

RICHARD L . MILLER, D.D.S., M.S.D. LARS E. A . FOLKE, D.D.S., PH.D.f C . ROBERT UMANA, M.D., PH.D.

MATERIALS AND METHODS

IT SEEMS ESSENTIAL that efforts be made to explore the possible individual differences between periodontally diseased and periodontally healthy individuals utilizing realistic parameters. The cellular response of the host to autologous dental plaque would constitute one such parameter. It has yet to be scientifically documented whether oral bacterial factors directly attract polymorphonuclear leu­ kocytes ( P M N s ) , or whether bacterial antigens and nonantigenic components of plaque activate the comple­ ment system which in turn chemotactically attract the P M N s . If such reactions occur in the crevicular environ­ ment, the immune system along with the organisms may play a role in the initiation and persistence of periodontal inflammation. The P M N response and associated phagocytic activity have been proposed as a limiting factor in the rapid spread of gingivitis, because intact and disintegrating P M N s are found within the crevice or crevicular epithe­ lium and adjacent connective tissue . Yet, we have no clear understanding of why these cells are present or what their function is. Many substances act as chemotactic factors, and it has been proposed that they be classified according to their mode of action. Cytotaxins have a direct chemotactic effect on cells, while cytotaxigens induce formation of cytotaxins in the presence of fresh s e r u m . The interaction of complement with endotoxins has recently been investigated, and it has been demonstrated that the generated low molecular weight product, C , is chemotactic for polymorphonuclear l e u k o c y t e s . Thus, it is apparent that the chemotactic properties of many substances depend on serum factors, of which

A. Studies with Pooled Plaque and Heterologous Polymorphonuclear Leukocytes Preparation of Plaque Suspensions. Dental plaque samples were obtained from several individuals with no reference to their periodontal condition and pooled. Plaque suspensions were made with a Dounce homogenizer by adding sterile 0.14 M N a C l to obtain a final concentration of 10, 20, and 40 m g / m l . Complete homogenization was then accomplished by using both loose and tight pestles for 30 seconds each. Chemotactic studies were then conducted to observe the effect of these different suspensions. A supernatant and an ultrafiltrate of a plaque suspen­ sion (20 mg/ml) were also tested for chemotactic ability. The supernatant fraction was obtained after centrifugation of the plaque suspension at 0 ° C for 15 minutes at 2500 rpm (1800 x g). The ultrafiltrate was acquired after filtering the previous supernatant through a 0.45-µ Millipore filter. Suspension, supernatant, and ultrafil­ trate were stored in ice baths prior to chemotactic analysis. Preparation of Polymorphonuclear Leukocyte Sus­ pension. Human polymorphonuclear leukocyte cells were collected by a technique essentially described by Gewurz et a l . in which a leukocyte-rich plasma was prepared. A 30-ml venous blood sample was collected from each individual with a sterile 35-cc Monoject syringe containing 45 units of heparin. This sample was centrifuged twice at approximately 1200 rpm to obtain heparinized homologous plasma to be used as the complement source. A second 30-ml sample was col­ lected with a syringe containing heparin and 9 ml of 6% dextran. The presence of dextran permitted the collection of a leukocyte-rich plasma after a 1-hour gravity sedi­ mentation at room temperature. A Neubauer (AO Instrument C o . , Buffalo, N . Y . ) hemocytometer, with a nuclear staining media as the diluting fluid (0.1 gm of crystal violet, 10.0 ml of glacial acetic acid, and 390.0 ml

1

2,8

9

20

10,11

10,12

5 a

13,14

* This investigation was supported by Public Health Service Grant D E 03174-02 from the National Institute of Dental Research. Principal Investigator Lars E . A . Folke. † Division of Periodontology, School of Dentistry, University of Minnesota, Minneapolis, M i n n . 55455.

409

410 Miller, Folke, Umana

J. Periodontol. July, 1975

of distilled water), was used to make an initial count of leukocytes in this suspension. After the initial count, the cell concentration was adjusted by centrifuging the leukocyte-rich plasma at 800 rpm for 10 minutes. After centrifugation, plasma was pipetted off, the pellet resus­ pended, and the counting repeated until a concentration of 10 x 10 P M N / m l was obtained. Two milliliters of this cell suspension were then transferred to a test tube containing 8 ml of L-15 (Grand Island Biological Co., Grand Island, N . Y . ) culture medium fortified with glucose (1 mg/ml). Penicillin streptomycin (100 fig/ml) and 2% human serum albumin were also added to the medium. This initial cell concentration of 10 x 10 P M N / m l was diluted with L-15 medium to a final concentration of 2 x 10 P M N / m l . 6

6

6

B. Studies with Individual Plaque and Polymorphonuclear Leukocytes

Autologous

The periodontal condition of 14 males were scored utilizing a modified Ramfjord periodontal disease index ( P D I ) . These patients were paired on the basis of identical age and high and low P D I scores (0 to 5.5). The individuals' ages ranged from 36 to 65 years. Sub- and supragingival plaques were randomly col­ lected from each subject, homogenized, and adjusted to a 40 m g / m l concentration as previously described. The negative control suspension consisted of 0.6 ml of sterile saline. Four negative control experiments were con­ ducted for each subject, and the average value was expressed as the "random P M N migration" during the time of the assay. A solution containing an antigen-anti­ body complex with chemotactic properties was used as positive control to monitor proper function of the in vitro system. Two chambers with bovine serum albumin ( B S A ) and anti-bovine serum albumin complex, precipi­ tated at the zone of equivalence (approximately 3 mg protein per ml), were used for each individual. The positive control suspension consisted of 0.1 ml of B S A and anti-BSA, 0.5 ml of complement source, and L-15 medium. Determination of Chemotaxis. Four chambers were used for each individual plaque test. The lower compart­ ment of each chamber was Filled with 0.1 ml of a whole plaque suspension, 0.5 ml of complement source, and L-15 medium. When individual plaque suspensions were used, the complement source and P M N cells were always from the same donor. The material to be tested for its chemotactic properties was introduced into the lower compartment of the chambers (a) in Figure 1. The P M N cell suspension was then pipetted into the upper compartment of the chamber (b). The prepared chambers (Fig. 2) were then incubated in air for 1.5 hours at 37°C. The filters were subsequently removed, placed in a custom-made filter holder, and immersed in sterile saline to prevent dehydration. The filters were stained in hemotoxylin and mounted on regular microscopic slides with the side previously ex­ posed to the plaque suspension facing the cover glass. 21

FIGURE 1. Diagrammatic representation of the modified Boyden chamber. Plaque suspension, complement source and L-15 medium were introduced into the lower compartment (a) to a level x. When the filter (b) was "moistened," the PMN sus­ pension and L-15 medium were simultaneously pipetted into the upper compartment (c) and upper extension (d), respec­ tively.

FIGURE 2. Photograph demonstrating the chamber filled with the plaque suspension, complement source, and L-15 medium in the lower compartment and the PMN suspension in the upper compartment. The chemotactic property of the various suspensions tested were determined by counting all the cells that had migrated completely through the filter (Fig. 3). Using a "blind" technique (Miller 1970), 10 contiguous Fields on the Filter surface were counted employing a 40x objective and a Zeiss C-10X grid. The number of cells recorded was expressed as the chemotactic index. The chemotactic ability of whole plaque, its fractions, or different concen­ trations were expressed as a corrected chemotactic index (CCI). This index was obtained by subtracting the "ran­ dom P M N migration" count from the chemotactic in-

Volume 46 Number 7

Chemotactic Ability of Human PMN Cells 411

dex. Statistical analysis was performed using the W i l ­ coxon signed rank sum test. OBSERVATIONS

A . Studies with Pooled Plaque and Heterologous Polymorphonuclear Leukocytes Dental plaque collected from randomly selected pa­ tients was consistently chemotactic for human P M N s at concentrations of 20 m g / m l . The response observed in the negative control experiments can be attributed to passive and random migration, while the chemotaxis induced by the plaque and the positive control sus­ pensions indicate active directional migration (Fig. 3). The supernatant and ultrafiltrate of whole plaque suspension revealed statistically significant differences (P = 0.016) in chemotactic ability. The plaque supernatant was found to be approximately 60% less chemotactic than whole plaque, while plaque ultrafiltrate induced negligible P M N attraction (Table 1). The chemotactic response to whole plaque suspensions increased with greater plaque concentration. However, the chemotactic activity did not bear a direct relationship to the plaque concentrations tested per unit time (Fig. 4).

FIGURE 4. Chemotactic properties of various concentrations of whole plaque suspensions. T A B L E 2. Corrected Chemotactic Index for Plaque Derived from Subjects with Low and High PDI Scores Autologous P M N s

Pair

FIGURE 3. Positive control. The chemotactic response ob­ tained towards the BSA /anti-BSA complex. T A B L E 1.. Corrected Chemotactic Index (CCI) of Plaque Suspension, Supernatant, and Ultrafiltrate in Reference to Heterologous Human Polymorphonuclear Leukocytes*

Experiment

Whole plaque

1 2 3 4 5 6

245 280 149 162 373 298

99 135 114 48 89 68

Mean C C I

235 (/> = 0.016)

92

Supernatant

Ultrafiltrate 39 37 50 16 1 2 24 (P = 0.016)

* Sub- and supragingival plaques were collected at random from different subjects and pooled.

A B C D E F Mean

Subject no. 1 3 5 7 9 11

Low PDI

Subject no.

78 221 335 220 238 162 206

2 4 6 8 10 12

High PDI 141 114 32 142 217 329 163

(P > 0.05)

B. Studies with Individual Plaque and Polymorphonuclear Leukocytes

Autologous

Individual plaque suspensions (40 mg/ml) from all the subjects studied were consistently found to be chemotac­ tic for autologous P M N s regardless of their periodontal disease state (Table 2). N o significant differences could be detected between subjects with high or low P D I scores (P > 0.05). However, subjects with a high P D I demon­ strated a trend for a reduced chemotactic response, particularly subject N o . 6. C . Negative and Positive Control

Experiments

Evaluation of the negative and positive controls of the 12 subjects studied showed that the positive control

412

J. Periodontol. July, 1975

Miller, Folke, Umana

values were approximately 10 times those of the negative controls (Table 3). The range in random P M N migration (1 to 97) indicates the variation in passive leukocyte migration from individual to individual. When the indi­ vidual C C I was contrasted with the amount of random P M N migration, no correlations could be found (Table 4). The variability of the P M N response to the negative and positive control solutions was studied in one subject at four different time points within a 2-month period (Table 5). It was noted in this individual that the chemo­ tactic responses in the negative and positive control ex­ periments were always low and consistently less than the average values of the other subjects. From these data, it appears that the P M N s of this individual behave simi­ larly from week to week not only in their random migra­ tion, but also in their chemotactic response towards the antigen-antibody complex. The great intersubject variability in the positive control experiments (range 145 to 668) suggests a great deal of individual differences in this particular immunologic response (Table 3). However, regardless of the P D I , all individuals had a C C I below the mean of the positive control. Subjects with high P D I scores presented larger nega­ tive (P > 0.05) and positive (P = 0.047) control values than subjects with low P D I scores (Table 6). However, as previously noted among the subjects with high P D I

T A B L E 4. Comparison of "Random PMN Migration" in Each Individual Studied with Corrected Chemotactic Index Random migration

Subject

Subject

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

Positive control

25 22 51 42 29 22 16 21 79 91 72 72 46 49 79 97 10 13 23 28 1 3 17 24

376 314 612 450 668 491 658 470 145 164 191 151 405 403 608 342 296 317 385 328 282 165 573 611

24 47 26 19 85 72 48 88 12 26 2 21

221 114 335 32 220 142 238 217 162 329

Mean

39

185

T A B L E 5. Response of PMNs from One Individual to Negative and Positive Control Solutions at Four Different Time Points Negative control

Day

1 12 26 54

39 1-97

392 145-668

Mean Range

Positive control

22 13 30 15 15 11 4 10

184 226 309 161 184 312 317 257

15 4-30

243 161-317

scores (Table 2), subject N o . 6 showed a comparatively low P M N response to both the positive control and the autologous plaque solutions (Table 6). When the subjects were arbitrarily divided into groups according to their low (1.0) plaque indices, the observations suggest a greater P M N response to autologous plaque among individuals with high plaque indices (Table 7). However, the differ­ ences in cellular response between the two groups was not statistically significant (P > 0.05). DISCUSSION

Dental plaque suspensions were shown to be chemotac­ tic in vitro for human polymorphonuclear leukocytes. This result is in accord with recent reports that specific oral bacteria are capable of chemotactically attracting rabbit P M N cells. The different chemotactic ability of whole plaque suspension, supernatant, and ultrafiltrate when tested with human P M N cells agrees with the findings of Tempel et al. When the supernatant was tested over the same period of time and compared with the whole plaque suspension, it was found to be less than half as 18

18

Mean Rkge

78 141

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

T A B L E 3. PMN Response to Saline (Negative Control) and to Antigen-Antibody Complex (Positive Control) Negative control

CCI

Chemotactic Ability of Human PMN Cells 413

Volume 46 Number 7

TABLE 6. Response of Individual PMNs to Negative and Positive Control Solutions in Reference to Low and High PDI Scores High P D I

Low P D I Subject

Negative control

Positive control

1 3 5 7 9 11

24 26 85 48 12 2

345 580 155 404 307 224

Mean

33

336

Subject

2 4 6 8 10 12

TABLE 7. PMN Response (CCI) to Autologous Plaque in Individuals with Regard to Low or High Plaque Indices High Plaque Index

Low Plaque Index

(>1.0)

K1.0)

335 32 238 220 142 217 329

221 78 162 141 114

Mean

(P > 0.05)

active, suggesting that the bacteria themselves rather than soluble enzymes are primary chemotactic factors. The negligible chemotactic ability of the ultrafiltrate, which contained no bacteria, substantiated this conten­ tion. Chemotaxis induced by bacteria is supported by previous work, and subsequent investigations have revealed that viable bacteria produce cytotaxic fac­ tors. Such a factor has recently been isolated from a strain of Staphylococcus aureus and found cyto­ taxic both in vivo and in vitro. Our observations that the bacterial portion of the whole plaque suspension in­ duces the highest chemotaxis implicates cytotaxic fac­ tors on the basis of recent immunologic studies. How­ ever, cytotaxigenic components, related to plaque, cannot be disregarded as endotoxins have been observed to induce P M N migration in vivo and specifically interact with complement to generate a C factor. Since complement, as well as endotoxin producing organ­ isms, can exist in the gingival a r e a , this cytotaxigenic mechanism may be relevant particularly in reference to the concept held by Weismann and Thomas that endo­ toxins release lysosomal enzymes which in turn are capa­ ble of cleaving complement factor C into chemotactically active fragments. The plausible presence of antigen and antibodies in dental plaque may be another factor responsible for its chemotactic ability since such im­ munologic complexes are known to attract P M N cells. 22

10,11,

2 3 ,

Positive control

47 19 72 88 26 21

525 564 171 475 308 592

46

439

Although there were no statistical differences in the chemotactic response to autologous plaque when com­ paring subjects with low and high P D I scores, a tendency to depressed chemotaxis was observed among subjects with severe periodontal disease. It is conceivable that this altered chemotactic response is related to plasma inhibi­ tory factors recently reported by M a r y et al. The differences in chemotaxis among the subjects studied, to the positive control solution ( B S A / a n t i - B S A complex) seems to represent an individual immunologi­ cal response to this antigen-antibody complex. This tentative conclusion received support from the repetitive experiments performed in one individual at four different time points. The results indicated that the chemotactic index remained quite similar across time when the P M N cells were challenged with the same control solution. The observations that two subjects with high P D I scores had a chemotactic index related to the positive control solution that was greater than that of the low P D I subjects and that they had a comparatively lower C C I in response to autologous plaque could be of clinical importance (Tables 2 and 6). The preliminary data also tends to indicate that the chemotactic response by human P M N cells directly or indirectly may be modi­ fied by the bacterial composition or other chemotactic mediators of plaque. This hypothesis would require additional scientific support. When all the patients were evaluated without reference to the P D I , there did appear to be a relationship between the amount of plaque present on the teeth and the migration rate of the P M N s . A plaque index of 1.0 or more seemed to produce a higher C C I with autologous P M N cells (Table 7). Whether this chemotactic effect is inherently related to the composition of plaque from higher plaque formers and whether present observations apply to a larger panel of subjects has yet to be determined. 17

216

143

Negative control

2 4

25

24

4

5

27

27,28

29

5

30

31

SUMMARY

The chemotactic effect of pooled human plaque sus­ pension, supernatant, and ultrafiltrate upon heterologous human P M N s was investigated using the Boyden cham-

J. Periodontol. July, 1975

414 Miller, Folke, Umana ber technique. It was observed that pooled plaque suspensions (20 mg/ml) were consistently chemotactic for heterologous human P M N cells. Whole plaque suspensions were most chemotactic, and the supernatant was approximately half as active, while the bacteria-free ultrafiltrate induced a negligible chemotactic response. Chemotactic assays of individual plaque suspensions were also performed. Twelve male subjects were paired according to age and P D I scores to assess whether the P M N cells of certain individuals responded differently to their autologous plaque. When comparing subjects with high or low P D I scores, there were no significant differ­ ences in the chemotactic responses. However, a trend of reduced chemotaxis was observed in most subjects with a high P D I . When the subjects were arbitrarily divided into groups with high and low plaque indices, a greater overall chemotactic response was generated by the higher plaque formers. The differences between the two groups, how­ ever, were not statistically significant. ACKNOWLEDGMENT

The authors gratefully acknowledge the technical assistance of Miss Bonnie Smith Carver. ADDEDUM

Upon completion of this manuscript others have reported that whole plaque is chemotactic over divergent molecular weight fractions, with major activity in fractions «5000 molec­ ular weight, (Hellden, L., Ericson, T., and Lindhe, J.: Scand. J Dental Res 81: 276, 1973). The coincidence of severe perio­ dontal disease with a serum derived substance which neutral­ izes chemotactic factors of plaque has recently been observed in our laboratory. (Gothier, Gaumer, Pihlstrom and Folke: J Periodont Res 10: (In Press), 1975). REFERENCES

1. Waerhaug, J.: Anatomy, physiology and pathology of the gingival pocket. Belg Tijds Tandh 21: 9, 1966. 2. Schroeder, H . E., and Theilade, J.: Electron microscopy of normal human gingival epithelium. J Periodont Res 1: 95, 1966. 3. Freedman, H . L., Listgarten, M . A., and Taichman, N . S.: Electron microscope features of chronically inflamed human gingiva. J Periodont Res 3: 313, 1968. 4. Jensen, S. B., Theilade, E., and Jensen, J. S.: Influence of oral bacteria and endotoxin on cell migration and phagocytic activity. J Periodont Res 1: 129, 1966. 5. Zachrisson, B. U.: A histological study of experimental gingivitis in man. J Periodont Res 3: 293, 1968. 6. Levy, H . E., Taylor, A . C , and Bernick, S.: Relationship between epithelium and connective tissue in gingival inflamma­ tion. J Dent Res (suppl) 48: 625, 1969. 7. Porteous, J. R.: Architectural changes in gingivitis and some implications of the plasma cell response. Pathol Micro­ biol 33: 308, 1969. 8. Löe, H . E., Theilade, E., and Jensen, S. B.: Experimental gingivitis in man. J Periodontol 36: 177, 1965. 9. Keller, H . , and Sorkin, E.: Chemotaxis of leucocytes. Experientia 24: 641, 1968. 10. Keller, H . , and Sorkin, E.: Studies on chemotaxis. VI.

Specific chemotaxis in rabbit polymorphonuclear leucocytes and monocular cells. Int Arch Allergy 31: 575, 1967. 11. Keller, H., and Sorkin, E.: Studies on chemotaxis. VII. Cytotaxins in rabbit serum. Experientia 23: 549, 1967. 12. Keller, H., and Sorkin, E.: The significance of normal sera for chemotaxis induced by various agents. Immunochemistry 9: 441, 1965. 13. Snyderman, R., Gewurz, H . , and Mergenhagen, S. E.: Interactions of the complement with endotoxic lipopolysaccharide: generation of a factor chemotactic for polymorphonuclear leukocytes. J Exp Med 128: 259, 1968. 14. Snyderman, R., Shin, H . S., Phillips, H . , Gewurz, H . , and Mergenhagen, S. E.: A neutrophilic chemotactic factor derived from C upon interaction of guinea pig serum with endotoxin. J Immunol 103: 413, 1969. 15. Miller, R. L., Umana, R. C , and Folke, L. E. A.: Chemotactic ability of dental plaque upon homologous and heterologous human PMNs J Dent Res 50 (suppl): 244 (ab­ stract) 1971. 16. Mary, G., and Folke, L. E. A.: Chemotactic properties of dental plaque J Dent Res 52 (suppl): 373 (abstract) 1972. 17. Mary, G. G., Haggan, G., and Folke, L. E. A.: The effect of host specific factors, plaque and autologous microor­ ganisms upon the chemotactic response of human P M N cells. J Periodont Res. (suppl.), 10: 23, 1972. 18. Tempel, T. R., Snyderman, R., Jordan, H . V., and Mergenhagen, S. E.: Factors from saliva and oral bacteria chemotactic for polymorphonuclear leukocytes: Their possible role in gingival inflammation.J Periodontol 41: 71, 1970. 19. Attstrom, R., and Egelberg, J.: Presece of leukocytes within the gingival crevice during developing gingivitis in dogs. J Periodont Res 6: 110, 1971. 20. Gewurz, H . , Page, A. R., Pickering, R. J., and Good, R. A.: Complement activity and inflammatory neutrophil ex­ udation in man. Int Arch Allergy 32: 64, 1967. 21. Ramfjord, S. P.: Indices for prevalence and incidence of periodontal disease. J Periodontol 31: 51, 1959. 22. Harris, H.: Role of chemotaxis in inflammation. Physiol Rev 34: 529, 1954. 23. Keller, H., and Sorkin, E.: Studies on chemotaxis. V. On the chemotactic effect of bacteria. Int Arch Allergy 31: 505, 1967. 24. Ward, P. A . , Lepow, I. H . , and Newman, L. J.: Bacterial factors chemotactic for polymorphonuclear leuko­ cytes. Am J Pathol 52: 725, 1968. 25. Walker, W. S., Barlet, R. L., and Kurtz, H . M . : Isolation and partial characterization of a staphylococcal leukocyte cytotaxin. J Bacteriol 97: 1005, 1969. 26. Brandtzaeg, P.: Local factors of resistance in the gingival area. J Periodont Res 1: 19, 1966. 27. Gibbons, R. J., Socransky, S. S., DeAraujo, W. C , and Van Houte, J.: Studies of the predominant cultivable microbiota of dental plaque. Arch Oral Biol 9: 365, 1964. 28. Bladen, H . , Hageage, G., Pollock, F., and Harr, R.: Plaque formation in vitro on wires by gram-negative oral microorganisms (veillonella). Arch Oral Biol 15: 127, 1970. 29. Weismann, G., and Thomas, L.: Studies on lysosomes. I. The effects of endotoxins, endotoxin tolerance, and cortisone on the release of acid hydrolases from a granular fraction of rabbit liver. J Exp Med 116: 433, 1962. 30. Ward, P. A., and Hill, J. G.: C chemotactic fragments produced by an enzyme in lysosomal granules of neutrophils. J Immunol 104: 535, 1970. 31. Boyden, S. V.: The chemotactic effect of mixtures of antibody and antigen on polymorphonuclear leukocytes. J Exp Med 115: 453, 1962. 5

5

Chemotactic ability of dental plaque upon autologous or heterologous human polymorphonuclear leukocytes.

The chemotactic effect of pooled human plaque suspension, supernatant, and ultrafiltrate upon heterologous human PMNs was investigated using the Boyde...
5MB Sizes 0 Downloads 0 Views