Recently, it has been shown that the NK cell constitutes between 3-7% of the total mononuclear cell infiltrate in periodontitis (Kopp. 1988. J. Clin. Periodontol. 15:595-600). These cells seem to be associated with active, B-cell lesions and are virtually absent in healthy gingivae (Wynne et al., 1986. J. Periodontol. 57: 699-702). The NK cell has important immunoregulatory functions (for example, the production of IFNg). The role of the NK cell in periodontal disease is as yet, unknown.
Cytokines are expressed by many cell types, including monocytes, T-cells, B-cells, NK cells, platelets, fibroblasts, bone cells, keratinocytes (in essence, all nucleated cells and platelets). However, certain cells are much more active at producing certain cytokines than other cells. For example, IL-1 is a cytokine which is produced in one form or another by virtually all nucleated cells. IL-1b is produced by activated B-cells, monocytes, and macrophages. Keratinocytes generate interleukin-1a (IL-1a). T-cells and NK cells are important sources of interferon g (IFNg).
Lymphocyte-monocyte-cytokine system effects on bone. Bone, like all connective tissues, is continually remodelling. Remodelling is due to two processes: osteoclastic bone resorption and osteoblastic bone formation. Usually, these processes are coupled. The lymphocyte-monocyte-cytokine system cytokines may affect both processes and may stimulate either net bone loss, net bone formation, increase both processes to increase the rate of remodelling, or decrease both processes to decrease the rate of remodelling. When there are net decreases or increases in bone, it is said that the processes of bone formation and resorption have been uncoupled.
Cytokines and net bone loss. Dental scientists first revealed that IL-1b and "osteoclast activating factor (OAF)" were the same protein (Dewhirst et al., 1985J. Immunol. 135: 2562-2568). IL-1b, TNFb, and TNFa can induce osteoclastic bone resorption, and IL-1b and TNFa interact synergistically in that process. As mentioned in the chapter on monocytes, IL-1b does not interact directly with osteoclasts, but probably stimulates osteoblastic cells which subsequently may induce osteoclasia. IL-1b appears to inhibit the synthesis of collagen, alkaline phosphatase, and osteocalcin by osteoblasts (Bertolini et al., 1986. Nature (London) 319: 516-518). Interestingly, short term exposure to IL-1b, TNFb, and TNFa can induce osteoblastic bone formation, whereas prolonged exposure to these cytokines inhibits osteoblastic bone formation.
Clinically, IL-1b is known to be elevated in the crevicular fluid and gingival tissues at sites of periodontal lesions (Masada et al., 1990. J. Periodont. Res. 25: 156-163; Jandinski et al., 1991. J. Periodontol. 62: 36-43). At periodontitis sites, bone resorptive and inhibitory cytokines (IL-1a, IL-1b, TNFa) reach levels of 0.3, 11.7, and 0.4 ng/ml, respectively. Clinically healthy sites exhibit much lower levels of these cytokines: 0.07, 3.1, and 0.03 ng/ml, respectively (Stashenko et al., 1991. J. Periodontol. 62: 504-509). In terms of potency, IL-1b is15 times more potent than IL-1a and 500 times more potent than TNFa in stimulating bone resorption. Given its higher gingival levels and greater bone resorptive potency, it is clear that IL-1b is likely to be important in the bone resorptive aspects of periodontal disease. In addition, TNFa may affect bone formation; it potently down-regulates osteocalcin transcription and protein levels by osteoblast-like cell lines. Both IL-1a and IL-1b decrease osteocalcin levels but have no effect on transcription. Osteoblast-like cells do have receptors for IL-1, as evidenced by the increase of PGE2 production in the presence of IL-1 (Stashenko et al., 1994. In: Genco et al. (eds) Molecular Pathogenesis of Periodontal Disease Amer. Soc. Microbiol. Washington DC, pp171-181). Osteocalcin is a noncollagenous protein made by osteoblasts that appears to directly regulate bone formation and, indirectly (?) bone resorption.
Both macrophages and activated B-cells produce IL-1b (Fig. 14). Macrophages also release prostaglandin E2 (PGE2, see Fig. 15), a short range inflammatory agent which stimulates with osteoclasts. IL-1b can also induce the production of collagenase and PGE2 from gingival fibroblasts (Richards et al., 1988. Arch. Oral Biol. 33:237-243; Richards et al., 1990. J. Periodont. Res. 25: 222-229).A role for Interleukin-6 (IL-6) in bone resorption. Recently, IL-6 has been shown to share bone resorptive effects with IL-1b (Mundy, 1991. J. Periodontal Res. 26:213-217). IL-6 is produced by many cells, including Th2 CD4+T-cells, bone marrow stromal cells and osteoblasts, and may play a central role in bone resorption. Inhibition of IL-6 production abrogates the bone resorptive effects of IL-1.
PGE2 and bone resorption. As described above, PGE2 can mediate short range bone resorption. Activation of inflammatory cells causes a local release of arachidonic acid from the plasma membranes of cells. This activity is catalyzed by phospholipase A2 (Fig. 15). Inflamed periodontal tissues exhibit a substantial elevation of free arachidonate (El Attar et al., 1986. J. Periodontal Res. 21: 169). The free arachidonate is the metabolized oxidatively by mononuclear cells via the cyclooxygenase pathway and the potent bone resorbing hormone, PGE2 is produced. PGE2 has been shown to be elevated in periodontitis lesions (Goodson et al., 1974. Prostaglandins 6: 81) and elevated PGE2 levels in the gingival crevicular fluid appear to correlate with periods of periodontal disease activity (ie; destructive phase of connective tissue remodelling) (Offenbacher et al., 1986. J. Periodont. Res. 21: 101-112). Functional receptors for PGE2 have been detected in isolated osteoclast-like cells but not osteoblast-like cells (Dziak, R. M. et al., 1983. Calcif. Tissue Int. 35:243-249). PGE2 is released by mononuclear cells in the presence of bacterial lipopolysaccharides (Garrison et al., 1988) or complement component C6. At present, it is unclear to what extent PGE2 contributes to bone resorption in periodontitis. However, several studies suggest that the blockade of the cyclooxygenase activity by aspirin and nonsteroidal antiinflammatory drugs (NSAIDs) such as indomethacin can reduce alveolar bone loss up to 28% (Howell and Williams, 1992. Pharmacologic blocking of host responses as an adjunct in the management of periodontal diseases: a research update. Position Paper, Am. Acad. Periodontol.) Other NSAIDs, such as topical "substituted oxaolopyridine derivative," topical ibuprofen, topical piroxicam, topical meclofenamic acid, and p.o. flurbiprofen or p.o. naproxen appear to slow alveolar bone loss or reduce bleeding on probing. These observations are not only of clinical significance, but also suggest that a percentage of alveolar bone loss in periodontal diseases is due to cyclooxygenase products of mononuclear cells. From a clinical perspective, it is also interesting that certain aromatic oils (eugenol, guaiacol, thymol, creosol, and capsaicine) can block the formation of prostaglandins in vitro by up to 50% (El Attar, 1978. J. Oral Pathol. 7: 175-207). Eugenol (an oil from cloves) is particularly common in periodontal dressings and a variety of endodontic applications. Eugenol is also thought to have antiseptic and anesthetic properties. Lymphocyte and monocyte connective tissue cytotoxicity (TNFb and TNFa). O'Neill et al., 1982 demonstrated that the gingival mononuclear cells exhibited functional activity distinct from peripheral blood lymphoid cells (PBL). Using unseparated mononuclear cells isolated from gingival connective tissue via collagenase extraction, they showed that the GMC in periodontitis were more cytotoxic for fibroblasts than were PBL. It is possible that tissue destruction may occur as a result of the activation of lymphoid cells by pathogenic bacteria (Lindemann et al, 1988). Cytotoxicity may be mediated by a number of factors including TNFb (or "lymphotoxin" if it is present as a TNFb-containing heterodimer) from lymphocytes and TNFa from monocytes (Fig. 16).Angiogenesis. Both IL-1b and TNFa bridge the gap between inflammation and healing, participating in both processes as angiogenic factors. Although TNFa is known for its cytotoxicity, the molecule can also induce endothelial proliferation. This angiogenic factors are made by cells of the monocyte lineage.
Fibrogenic cytokines. IL-1b and IL-1a are also involved in inducing fibroblast proliferation and collagen synthesis. This effect is not direct; and in vitro, they appear to have very little capacity to stimulate fibroblasts. The addition of inflammatory cells seems to lead to in vitro fibrogenesis (Kovaks. 1991. Immunol. Today 12: 17-23). It has been proposed that this is due to the production of PGE2 or secondary cytokines, especially, platelet-derived growth factor (PDGF) and Transforming growth factor (TGFb). PDGF is a dimeric 30 kdal cationic protein complex (consisting of combinations of aa, ab, and bb chains) which was originally isolated from the a granules of platelets. The a subunit is encoded on chromosome 7 and the b subunit is encoded on chromosome 22. Other fibrogenic cytokines which may play a role include fibroblast growth factor (FGF), TGFa, and TNFa. The fibrogenic cytokines are produced mainly by cells of the monocyte lineage.
Gingival hyperplasia and fibrogenic cytokines. Phenytoin (PHT), a drug used to control seizures, and cyclosporin A (CsA), a drug used to suppress specific T-cell responses, have both been associated with gingival connective tissue hyperplasia in about 50% of users. It has been shown that PHT increases the expression of PDGF in cultured monocytes and that CsA increases PDGF in gingival tissues independent of the inflammatory state of the tissues (Dill et al., 1993. J. Periodontol. 64: 169-173; Nares et al., 1996. J. Periodontol. 67: 271-278). The increased expression of such a fibrogenic cytokine may explain the association of these two drugs with gingival connective tissue hyperplasia. CsA has also been associated with production of the pro-inflammatory cytokine, IL-6, possibly by gingival fibroblasts, suggesting an additional way in which CsA may perturb the chronic immune system and lead to gingival hyperplasia (Williamson et al. 1994 65: 895-903)
Negative regulators of inflammation. At some point, the immune system must shut down of "pro-inflammatory" processes such that anti-inflammatory processes dominate. Certain factors are thought to transmit the signal to initiate this conversion, including IL-1 receptor antagonist (IL-1ra), TGFb, and IFNg (Genco, 1992. J. Periodontol. 63: 338-355).
There are at least three ways in which the immune system (along with other tissues, such as bone tissue itself) can induce bone healing. The first way is to prevent osteoclast activation by cytokines and the second is to block osteoclast formation. A third possible mechanism involves the activation of osteoblasts.
Blocking cytokine induced activation. NK-cells and Th1 T-cells can secrete interferon-g (IFNg). IFNg can inhibit osteoclast differentiation and proliferation (Mundy, 1991). The main effect of IFNg appears to be inhibition of IL-1 and TNFa induced osteoclast activation. Another cytokine which inhibits osteoclastic bone resorption includes interleukin-1-receptor antagonist (IL-1ra). IL-1ra is produced by monocytes and monocyte-derived cells, and is structurally related to IL-1a, IL-1b, and TNFa. IL-1ra antagonizes the osteoclastic effects of all three.
Blocking osteoclast formation. Osteoclasts appear to remain active for about 10 days before disappearing, possibly by splitting into mononuclear cells. TGFb is a potent inhibitor of osteoclast formation. Therefore, by blocking osteoclast formation, it is possible to cause the a marked decrease in osteoclastic activity within 10 days. Although monocytes can serve as a source of TGFb, so can the bone matrix itself. The bone matrix contains TGFb which is released by osteoclastic resorption (Mundy, 1991).
Activation of osteoblasts. FGF, PDGF, and insulin-like growth factors I and II (IGF-I and IGF-II) are potent bone cell mitogens. The insulin-like factors induce osteoblast growth, differentiation, synthesis of Type-1 collagen, and generation of alkaline phosphatase. Intriguingly, as single exposure to IGF-I plus PDGF has been used therapeutically in dogs to induce bone regeneration at periodontal lesions (Lynch et al., 1989. J. Clin. Periodontol. 16: 545-548).
Given that the MHC gene products are the molecules which dictate which antigens will be presented to T-cells by antigen-presenting cells, the MHC gene products should control immune response pleomorphism in the periodontium. That is, in terms of our antigen-specific immunologic responses, none of us should behave in exactly the same fashion. Thus, there has been interest in determining whether there is a correlation between MHC gene products and periodontal disease.
MHC Class I molecules and periodontal diseases. Many of the studies with regard to periodontal disease were performed in the 1970s, when only serological assays (lymphocytotoxicity tests: killing of lymphocytes by antisera plus complement) were used to determine HLA-typing. These reactions primarily detect differences among the HLA-A, HLA-B, and HLA-C phenotypes (MHC Class I molecules). The student should bear in mind that the MHC class I molecules are primarily for the presentation of intracellular antigens. Reinholdt et al. (1977. J. Dent. Res. 56:1261-1263) detected a positive association of localized juvenile periodontitis with HLA-A9, HLA-A28, and HLA-BW-15. Both Reinholdt et al., 1977 and Terasaki et al. (1975.Tissue Antigens 5: 286-288) detected a possible association between decreased frequency of HLA-A2 and LJP. The Reinholdt data is shown (Table 7).
| Table 7: Association of MHC Class I molecules with Periodontal Diseases | |||||||||
| Group | n | HLA-A2 | HLA-A9 | HLA-A28 | HLA-Bw15 | ||||
| % | (p) | % | (p) | % | (p) | % | (p) | ||
| LJP | 39 | 43.6 | (=.13) | 38.5 | (=.003) | 23.1 | (=.03) | 38.5 | (=.003) |
| AP | 29 | 58.6 | (>>.05) | 6.9 | (>.05) | 10.3 | (>>.05) | 17.2 | (>>.05) |
| Controls | 1967 | 53.6 | 17.3 | 10.0 | 17.9 | ||||
| Note: Data was collected in Copenhagen. Reinholdt et al. 1977. J. Dent. Res. 56: 1261-1263. | |||||||||
Denmark has much more genetic homogeneity than the United States. Thus, Reinholdt et al. could detect relatively small variations in their study population. From a clinical point of view, these data are fairly useless (they have little predictive value); thus, they have been de-emphasized in dental research. From a biological point of view, it is extremely interesting that certain immune response phenotypes associate with certain diseases, especially HLA-A9 and HLA-Bw15 with LJP. These data further suggest that CD8+ T-cell processes are important in determining susceptibility to LJP. That is, perhaps CD8+ T-cells important in suppression are not responding properly in controlling immune responses in LJP. It is also possible that these markers are in linkage disequilibrium with some other, more critical gene within the MHC.
In London, Cullinan et al. (1980. J. Periodont. Res. 15: 177-184) examined African (Zambian) and Caucasian (unspecified nationality) populations. In Zambians, no significant difference in HLA-A or HLA-B alleles could be detected between individuals without LJP and individuals with LJP. In Caucasian populations, no significant association of LJP with differing frequencies in HLA-A2, HLA-A9, HLA-A28 or HLA-Bw15 was detected. The Cullinan et al., 1980 study was inconclusive, because the sample size was very small (12 Caucasians with LJP, 18 Africans with LJP).
Saxén and Koskimies (1984. J. Periodont. Res. 19: 441-444), in Helsinki, took a different approach. They decided to determine whether the HLA-phenotype within a family dictated the presence or absence of LJP. They found that both healthy and diseased siblings expressed HLA-Bw15 and HLA-A9. They suggested that therefore, the "gene for LJP" was not linked to the HLA complex. Now, we may view this conclusion as a bit simplistic. We know that "periodontal disease" is a result of both an opportunistic infection (failure of the acute immune system) and a host response to that infection (response of the chronic immune system, often involving MHC molecules).
MHC Class II molecules and periodontal disease. The MHC Class II molecules are more closely associated with immune responses to infection by extracellular pathogens. Since periodontal disease is envisioned to be a result of extracellular pathogens, it seems reasonable that MHC class II alleles are more likely to influence the outcome of the bacteria-host interaction than are MHC class I alleles, comparing individuals with periodontitis and normal controls. Tantalizing data suggests that there may be some association between rapidly progressing periodontitis (RAP) and the HLA-D phenotype (Katz et al., 1987. J. Periodontol. 58: 607-610). They found an extremely strong association between HLA-DR4 and RAP (in Israel with both Ashkenazi and non-Ashkenazi groups), using a B-cell enriched, complement-dependent lymphocytotoxicity system which could determine DR, DQ, and DP differences (This HLA-DR4 phenotype is also found in high proportions among rheumatoid arthritis). Unfortunately, only 10 subjects were examined in the Katz study.
The polymerase chain reaction - restriction fragment length polymorphism (PCR-RFLP) method has revolutionized the way in which alleles of a given gene are detected. These methods have a far greater resolution than previous serologically-based lymphocytotoxicity methods (ie., they can "see differences" not distinguished by serological methods) and has led to the understanding that the pleomorphism of the MHC has been greatly underestimated (Chapter 11). Such resolving power is critical, since variation within the specificity pockets of MHC molecules may go undetected by serology yet greatly influence the types of antigens which are bound and presented by the MHC. In Japan, PCR-RFLP has been used to study the class II alleles which may lead to susceptibility to EOP (Ohyama et al., 1996. J. Periodontol. 67: 888-894). In this study, no single allele was found which correlated with EOP; however, certain alleles of HLA-DQ were found with greater frequency in EOP. These included primarily members of the DQ1 family (serologically defined as DQ5 and DQ6; PCR-RFLP-defined as the DQ-b subunit alleles DQB1*0503 and DQB1*0602 --- aarrrggghhh!!!!). In Japan, DQB1*0503 and DQB1*0602 are in tight linkage disequilibrium with the HLA-DR alleles DRB1*1401 and DRB1*1501, respectively. The authors speculated that the findings suggest that either "DQ1-ness" or the "DR14/15-ness" can be associated with EOP, but exactly which ones cannot be ascertained due to the tight linkage disequilibrium.
Insulin dependent diabetes mellitus (IDDM). IDDM is an autoimmune disease which has been associated with periodontal disease when normalized for age and plaque indices (Cianciola et al., 1982. J. Am. Dent. Assoc. 104:653-660), and it is now recognized that periodontitis is one of the six main complications of IDDM (Genco, 1994. Compendium (suppl) 18: S678-S683). HLA-DR3 or DR4 are found in >90% of Scandinavian children with IDDM (type 1 diabetes) (Michelsen et al., 1990. Scand. J. Immunol., 31: 405-413). HLA-DQ3.2 has also been associated strongly with IDDM. At present, it is unknown why there is an association between HLA phenotype, periodontal disease, and IDDM. One possibility is that IDDM leads to host defects which predispose the subject to periodontal disease (for example, neutrophil dysfunction). A second possibility is that the HLA phenotype results in a predisposition for both autoimmunity and the presentation of antigens (including microbial superantigens) which can then lead to inappropriate responses by the chronic immune system. An interesting third possibility is that this association may represent a linkage between MHC class II alleles and other genes involved in MHC class I antigen presentation. Thus, it has also been proposed that the ATP-binding cassette proteins (Tap-1 and Tap-2) or certain proteasome-proteases (Lmp-1 and Lmp-2) encoded within the MHC class II region may also be involved in the autoimmune destruction associated with IDDM and IDDM-associated periodontal disease (Faustman et al., 1991. Science 254: 1756-1761). It should be pointed out that as yet, no one has demonstrated autoimmune destruction of the periodontium in periodontal disease. At any rate, the association between IDDM and periodontitis is presently one of the "hottest" topics in periodontal research.
Why should MHC class II haplotypes be associated with periodontal disease? It is attractive to believe that HLA-D phenotypes may correspond to periodontal disease entities, since we largely view periodontal disease as a consequence of extracellular infection. However, the student should be warned that even if there were an association between HLA-D region genetics and periodontal disease, that the precise mechanism would still be unknown. For example, HLA-DR4 may lead to inadequate responses to certain microbial epitopes as well as increased responses to host epitopes (hence the relationship with type 1 diabetes and rheumatoid arthritis). Remember that Lavine et al., 1979 found an association between RAP and cell-directed inhibitors of chemotaxis (CDI; which are immunoglobulins that appear to attack phagocyte function via the Fc rather than via the Fab portion). It is possible that HLA-DR4 may present peptides which stimulate autoimmunity or CDI.
In this chapter, a simplistic paradigm was provided for the role of the immune system in periodontal disease. In this paradigm, we considered the acute and chronic immune phases to be temporally, spatially, and functionally distinct. We applied our understanding of the role of leukocytes in the natural history of inflammation to develop a model consistent with observations that the acute phase cells appear to be protective and the chronic phase cells appear to be destructive, with respect to the periodontium. Undoubtably, this is a gross simplification, but it should give you a perspective with which to understand periodontal research until more understanding is developed (just like Newtonian physics held-up for centuries until Einstein came along -- well, not quite as important, I suppose!). Please remember, this paradigm is not shared by everyone (anyone?) in the USA. The really good scientists are too cautious to accept the paradigm at present even though they probably have thought it many times before (it's too speculative). But I do think you students need a point of view; otherwise, the periodontal research literature will seem rather daunting.
The paradigm is easily summarized. Neutrophils protect our gingival crevice from untoward shifts in the microbial ecology. Monocytes and lymphocytes engage in chronic immunologic functions (including tissue remodelling) within the gingival connective tissues. Periodontal disease is diagnosed after many events in which remodelling leads to a net loss of tissue. And check out the Venn diagram at the beginning of this chapter, what do you think?
Do you wish to take this course for Continuing Education credit? Yes
Top of Page | Page [ 1 | 2 | 3 | 4 ]