Human body's immune response.

Immune Response: The Key to Bone Resorption in Periodontal Disease Martin A. Taubman,* Paloma Valverde,† Xiaozhe Han,* and Toshihisa Kawai*

Periodontal disease infection with oral biofilm microorganisms initiates host immune response and signs of periodontitis, including bone resorp- tion. This review delineates some mechanisms underlying the host im- mune response in periodontal infection and alveolar bone resorption. Activated T lymphocytes have been historically implicated in experimen- tal periodontal bone resorption. An experimental rat adoptive transfer/ gingival challenge periodontal disease model has been demonstrated to require antigen-specific T lymphocytes and gingival instillation of antigen and LPS for bone resorption. Interference with costimulatory interactions between T cells and antigen-presenting cells abrogated bone resorption, further emphasizing the significance of immune response in periodontal disease. Receptor activator of nuclear factor kB ligand (RANKL), a critical osteoclast differentiation factor, is expressed on T lymphocytes in human periodontal disease as determined by immunohistochemical and confo- cal microscopic analyses. Interference with RANKL by systemic adminis- tration of osteoprotegerin (OPG), the decoy receptor for (and inhibitor of) RANKL, resulted in abrogation of periodontal bone resorption in the rat model. This finding indicated that T cell-mediated bone resorption is RANKL-dependent. In additional experiments, treatment of T cell-trans- ferred rats with kaliotoxin (a scorpion venom potassium channel inhibi- tor) resulted in decreases in T-cell RANKL expression, diminished induction of RANKL-dependent osteoclastogenesis, and abrogation of bone resorption, implicating an important role of immune response/ RANKL expression in osteoclastogenesis/bone resorption. In other ex- periments, adoptive transfer of antigen-specific, RANKL-expressing B cells, and infection with the antigen-bearing Actinobaccillus actinomy- cetemcomitans gave rise to periodontal bone resorption, indicating that B cells also have the capacity to mediate bone resorption, probably via RANKL expression. In humans, prominent T lymphocytes have been identified in periodontal disease, and diseased tissues showed elevated RANKL mRNA expression, as well as decreased OPG mRNA expression. Mononuclear cells from periodontal lesions involving T cells and B cells of patients induced osteoclastogenesis in vitro. In summary, a biofilm inter- face initiates immune cell infiltration, stimulating osteoclastogenesis/ bone resorption in periodontal disease. This resorption can be amelio- rated by inhibition of RANKL activity or by diminishing immune cell stim- ulation. These two procedures, if localized, have the potential to lead to the prevention or therapeutic management of periodontal disease and therefore require further study. J Periodontol 2005;76:2033-2041.

KEY WORDS

B lymphocytes; osteoprotegerin; periodontal disease; T lymphocytes.

P eriodontitis is associated with a constellation of oral microorganisms that in-

fect the gingival crevice.1 These polymicrobial infections cause gingival inflammation and re- sorption of alveolar bone. Host- mediated immune responses (both innate and adaptive) to these microorganisms lead to the destruction of periodontal tissues. Recent studies have suggested that the host im- mune response can contribute to protective and/or destructive effects in periodontal disease. In particular, attention has been focused on receptor activator of NF-kB ligand (RANKL), a member of the tumor necrosis factor ligand family, because of the requirement of this key regulatory molecule for osteo- clastogenesis.2,3 Bone remod- eling in vertebrates is the process of maintaining a con- stant bone mass by coupling the actions of the bone-pro- ducing cells, osteoblasts, with bone-resorbing cells, osteo- clasts. It has been suggested that RANKL is involved in alve- olar bone remodeling during orthodontic tooth movement4

and inflammatory alveolar bone resorption in periodontal dis- ease.5-7 Helper T lymphocyte (CD4+ T cells)-mediated immu- nity is an important arm of the adaptive immune response. These cells may differentiate

* Department of Immunology, The Forsyth Institute, Boston, MA. † Division of Oral Biology, Tufts University School of Dental Medicine, Boston, MA.

J Periodontol • November 2005 (Suppl.)

2033

into two major subsets of effector T cells: type 1 helper T cells (Th1 cells; responsible for cell-mediated immunity) and type 2 helper T cells: (Th2 cells; help for antibody synthesis). It has been demonstrated that these cell types were involved in the cellular immune response to bacteria in periodontal disease patients.8,9

The purpose of this manuscript is to delineate the critical role of host responses involving immune cells in the progression of periodontal disease.

HOST RESPONSE TO BACTERIA IN PERIODONTAL DISEASE

Ebersole and colleagues10 clearly indicated the in- volvement of host response to bacterial antigens in periodontitis. Host response to bacteria is involved in periodontal disease and can be detected as serum IgG antibody to particular bacteria and/or their anti- gens.11 The IgG antibody is elevated significantly above the mean level of subjects without periodonti- tis. This is illustrated in Table 1, which demonstrates that the organism (from a battery of 18 oral microor- ganisms that were tested to which there was an abnor- mally elevated serum antibody response) can be detected colonizing the oral cavity of approximately 85% of periodontitis patients.10 Thus, bacteria associ- ated with periodontal disease elicit an immune re- sponse.11 The host response to the ‘‘disease active’’ process was further investigated. The relationship be- tween serum antibody to an organism and detection of that organism (homologous microorganism) in ‘‘dis- ease active’’ versus ‘‘disease inactive’’ sites was ex- amined and compared (see Table 2). We compared disease-active sites (two to nine per patient) with dis- ease-inactive sites (one to 18 per patient) in the same group of 34 patients with chronic periodontal disease (chronic periodontitis, 22; generalized aggressive periodontitis, seven; and localized aggressive peri- odontitis, five). The organism to which there was an elevated response was de- tected in disease-active sites in 56% of the cases, whereas only 18% of inactive sites contained the elevated im- mune response and the ho- mologous microorganism. The difference in the recoveries of bacteria from diseased as op- posed to non-diseased im- mune response in these sites was highly significant (P <0.001 using chi square statis- tical analyses). Most im- portantly, these findings offer strong support that the host response to oral bacterial species arises in those micro-

organisms associated with an active disease lesion in human periodontal disease.

CELLULAR IMMUNE RESPONSE IN PERIODONTAL DISEASE

In addition to the humoral immune response de- scribed above, it was clear that there was also a cellu- lar immune response in periodontal disease patients, which seemed to be related to the severity of disease, as indicated by bone resorption.12 The signs of peri- odontitis can be related to immune cells and their products involved in the response, as shown in early studies from our laboratory, initiated by Yoshie and colleagues,13 implicating activated T lymphocytes in periodontal disease pathogenesis. The findings are summarized14 in Figure 1. Figure compares one group of rats (I) immunized with A. actinomycetemco- mitans (Fig. 1A), with another group (SI) that consisted of sham-immunized and non-immunized

Table 1.

Relationship of Elevated Serum Antibody Response to a Particular Organism and Oral Colonization by the Homologous Microorganism*

N Elevated

Serum Antibody

Responses

Detection of

Homologous

Microorganisms

34 patients 58 49

% homologous organisms detected of total number of elevated antibody responses

— 85%

* Ebersole et al. 10

Table 2.

Relationship Between Serum Antibody and Detection of Characterized Homologous Microorganism in Disease-Active Versus Disease-Inactive Sites*

Disease Active Disease Inactive

N Sites

Tested

(2-9/patient)

N Sites

With Organisms

Detected

N Sites

Tested

N Sites

With Organisms

Detected

34 patients 242 132 152 27

% sites with organisms detected of total sites tested

— 56% — 18%

* Ebersole et al. 10

The Immune Response in Periodontal Disease Volume 76 • Number 11 (Suppl.)

2034

animals (Fig. 1B). Delayed type hypersensitivity was demonstrated in the A. actinomycetemcomitans- sensitized rats (Fig. 1A). These rats were also infected with A. actinomycetemcomitans and the sensitized rats demonstrated significantly elevated bone resorp- tion (Fig. 1B; P <0.05; t test). Similar results were ob- served in an experimental mouse model, suggesting that T cells and their responses to oral infection with P. gingivalis help to push bone remodeling in the di- rection of net loss of bone.15,16 This indicated that the same mechanism could be applicable with other bacteria during the cellular immune response in periodontal disease. There are also reports that immunization of non-human primates with killed P. gingivalis; inhibits progression of periodontal tissue destruction, however, the cellular response was not elucidated.17 Although it is clear in several models that antibody can interfere with an experimental infec- tion, it is also clear in a myriad of situations described below that immune responses are the key to bone resorption in periodontitis.

The findings of T cell involvement were further sup- ported by adoptive transfer of sensitized A. actinomy- cetemcomitans-specific T cells or non-sensitized T cells to recipient rats infected with A. actinomycetem- comitans.13 Recipients of the antigen-activated T cells clearly demonstrated delayed-type hypersensi- tivity and significantly increased periodontal bone re- sorption (Fig. 2A), whereas the sham-immunized T cell-transferred and non-transferred control groups

demonstrated neither detectable delayed hypersensi- tivity nor increased periodontal bone resorption (Fig. 2B). Experiments of Yamashita and colleagues18

firmly supported the association of T lymphocytes with bone resorption by adoptive-transfer of T cells using characterized antigen-specific (A. actinomyce- temcomitans) T cell clones (Th2)19 followed by A. ac- tinomycetemcomitans infection. Recipient animals demonstrated markedly elevated levels of serum and gingival antibody to A. actinomycetemcomitans corresponding to an amelioration of periodontal bone resorption. Further study of adoptive transfer of antigen-specific Th2 clone cells to congenitally athy- mic recipients that were subsequently infected with A. actinomycetemcomitans also demonstrated mark- edly elevated antibody to A. actinomycetemcomitans and significant reductions in periodontal bone resorp- tion compared to appropriate controls.20 These

Figure 1. A) Foot pad swelling 24 hours after test injection of A. actinomycetemcomitans sonicate immunized or sham- immunized germ-free rats (6 to 8/group). Bars indicate the mean foot pad swelling, and the bracket encloses the standard error of the mean. The immunized group (I) foot pad swelling was significantly elevated on days 0, 10, 22, and 30 compared to sham-infected group (SI) (* P <0.05; Student t test). The foot pad lesion consisted of greater than 50% monocytes.14 B) Horizontal bone resorption in germ-free rats (N = 7 to 8/group) immunized and infected with A. actinomycetemcomitans or sham-immunized and infected. The bars indicate the mean bone loss, and the bracket encompasses the standard error. *Bone resorption was significantly elevated in the immunized group; P < 0.05; Student t test.14

Figure 2. A) Delayed-type hypersensitivity foot pad swelling to A. actino- mycetemcomitans in rats receiving A. actinomycetemcomitans sensitized T (AaT) cells (2 x 108), sham-sensitized T (ST) cells (2 x 108), and no cell transfer control group (C). Bars indicate the mean swelling – standard error (brackets). In this assay, more than 23% (24 hours) or 16% (48 hours) swelling was statistically significant when compared to controls.13 B) Bone loss in A. actino- mycetemcomitans-sensitized T cell transfer (AaT), sham-immunized T cell transfer (ST), and no transfer (C) rat group. Bars indicate the mean bone resorption + standard error (see bracket) on the surfaces indicated. Differences significantly greater in the AaT group (For A and B, *one-way analysis of variance; P at least 0.03 compared to control group).13

J Periodontol • November 2005 (Suppl.) Taubman, Valverde, Han, Kawai

2035

combined results verified previous studies with adop- tive transfer of antigen-specific Th2 clone cells into conventional syngenic rats.

B LYMPHOCYTES ARE ALSO ASSOCIATED WITH PERIODONTAL BONE RESORPTION

While our major thrust was the investigation of the role of T lymphocytes in the immune response and their association with bone biology, the mechanism of this association with bone resorption remained unknown. However, data were also accumulating in support of B cell involvement in bone resorption, using multiple myeloma as a model. It was evident that B cells seemed to be associated with the bone resorption ob- served in multiple myeloma.21 Importantly, Harada and colleagues22 developed a system where anti- gen-specific memory B cells could be generated and isolated after binding to antigen as an absorbent. Such B cells were specific for A. actinomycetemcomi- tans, and adoptive transfer of these B cells (80% to 85% B lymphocytes) resulted in marked elevations of serum IgG and IgM and gingival IgG antibody. Ad- ditionally, marked elevation in A. actinomycetemco- mitans-binding spleen cells and in spleen cells that could recognize antigens from A. actinomycetemco- mitans (spot-forming cells viewed by ELISPOT assay, an immunological assay to determine the number of activated antigen-specific T-cells producing a particu- lar cytokine or specific antibody secretion by B cells in a mixture of cells [In this case, specific antibody secre- tion by B cells was elucidated]) was observed in the B cell-transferred animals. Most importantly, the animals receiving transferred B cells also had signifi- cantly elevated periodontal bone resorption. In sum- mary, A. actinomycetemcomitans-specific B cells were adoptively transferred into normal syngenic re- cipients that were infected with A. actinomycetemco- mitans. The recipient rats demonstrated elevated serum antibody to A. actinomycetemcomitans, ele- vated A. actinomycetemcomitans antibody-forming cells in spleen, elevated antibody in gingival wash, and markedly elevated bone resorption. These fea- tures were interpreted to mean that the transferred B cells or their progeny had survived in the specific B cell recipients, and that these B cells were, in some way, responsible for the bone resorption.

TISSUE MONONUCLEAR IMMUNE CELL COMPOSITION IN DISEASED PERIODONTAL TISSUES

Since the immune cells seemed to be involved in peri- odontal disease bone resorption, we investigated the composition of cells extracted from diseased human gingival/periodontal tissues.23 We demonstrated that CD4+ mononuclear cells were relatively prominent (constituting approximately 27% of recoverable

mononuclear cells) and that the CD8+ cells were rel- atively enriched compared to the composition of peripheral blood or normal gingival tissues.23,24 The combination of findings suggested that the sensitized T lymphocytes (later considered activated Th1-type T cells)25 were in some way involved in the induction of bone resorption. However, at that time, no specific mechanisms were known or elucidated for the associ- ation of T cells with bone resorption. It was more than a decade before this association was clearly ex- plained. Thereafter, most significantly, Kawai and col- leagues26 were able to demonstrate the criteria for antigen-specific Th1 and Th2 clone cells to home to infected gingival and periodontal tissues. The criteria included the presence of antigen and lipopopolysac- charide (LPS).

RODENT MODEL FOR T CELL-MEDIATED PERIODONTAL BONE RESORPTION

Considering the fact that antigen and LPS could drive T cells to gingival tissues,26 we developed an experi- mental periodontal disease model in rats that could faithfully reproduce T cell-mediated bone resorption. The model required adoptive transfer of A. actinomy- cetemcomitans antigen-specific (omp29) T lympho- cytes9,26 and gingival instillation of antigen and LPS to generate periodontal bone resorption. A key finding in this model is the presence of abundant osteoclasts on the alveolar bone crest of the animals receiving T lymphocytes, but not on the crest of the ‘‘no T cell transfer’’ group or on the ‘‘no gingival challenge’’ rats. In this model, Th1 cells, in particular, were involved in periodontal bone resorption. This bone resorption could be abrogated by the administration of a fusion protein (CTLA4Ig) that interferes with costimulatory interactions between T lymphocytes and antigen-pre- senting cells (CD28 and B7), further implicating the immune response in the induction of bone re- sorption.9

DISCOVERY OF THE KEY FACTOR LINKING IMMUNE CELLS AND BONE RESORPTION (RANKL)

At about this time, a significant discovery was made by Kong and colleagues.3 A new member of the TNF family was found. The protein factor, which can be present on the surface of osteoclast precursor cells, is called receptor activator of nuclear factor kB (RANK). Its ligand (RANKL) can be found on osteo- blasts, stromal cells, and the T lymphocyte cell sur- face, and has recently been demonstrated on B cells.3,27 Receptor activator of RANKL, RANK (its cel- lular receptor), and the decoy receptor (OPG) were identified as the key molecular regulation system for bone remodeling.3,28 RANK-RANKL interaction, with macrophage colony stimulating factor [MCSF], on

The Immune Response in Periodontal Disease Volume 76 • Number 11 (Suppl.)

2036

osteoclast precursors is necessary and sufficient for the differentiation of functional osteoclasts. RANKL on osteoblasts, stromal cells, or osteoblast progeni- tors by ligation to RANK, expressed on osteoclast precursor cells with permissive MCSF, gives rise to activation and differentiation of osteoclasts and bone resorption (shown schematically in Fig. 3).

THE RODENT MODEL OF T CELL-MEDIATED PERIODONTAL BONE RESORPTION IS RANKL-DEPENDENT

In the rat adoptive transfer/challenge model of peri- odontal disease, we have demonstrated that the bone resorption is osteoclast-related.9 Others with similar models have inhibited the bone resorption with a fu- sion protein of human OPG and human IgG-Fc (OPG-Fc), suggesting that the resorption in our model

might also be RANKL-dependent.5,28 A major finding in our rat model was the discovery that bone resorp- tion could be abrogated by osteoprotegerin fusion protein (OPG-Fc).29 In order to investigate RANKL dependence of the bone loss in our model, we admin- istered OPG-Fc systemically on days 0, 2, and 4 after adoptive transfer. Alveolar bone resorption after treat- ment was reduced by 90% (N = 5; P <0.05) with re- spect to the control group injected with the control protein L6-Fc (a fusion protein of irrelevant peptide L6 and human Fc fragment). These experiments, along with the inhibition of bone resorption by OPG- Fc in other similar rodent models of bone resorptive disease,5 indicated a crucial role of T cell-derived RANKL in inducing bone resorption in the experimen- tal adoptive transfer/antigen challenge model used herein. Further studies showed that many of the

antigen-specific T lymphocytes (from the gingiva of T lym- phocyte adoptively transferred rats demonstrating bone re- sorption) expressed RANKL.30

Thus, Th1 T cell-mediated peri- odontal bone resorption in the rat adoptive transfer/antigen- challenge periodontal disease model is associated with ex- pression of RANKL by T cells, and is RANKL-dependent. This was further demonstrated in ex- periments in the rat model where T cell activation was markedly reduced and subse- quent RANKL expression was also greatly reduced, as was bone resorption.6

ABLATION OF T CELL STIMULATION AND RANKL EXPRESSION BY KALIOTOXIN

Since RANKL (the key mole- cule for bone remodeling) is expressed by antigen-activated T cells,3,28 and subsequent mo- lecular events are partly regu- lated by calcium-dependent signal transduction,3,31 we hy- pothesized that the specific blocking of Kv1.3 potassium channels which balance Ca2+

cell entry by kaliotoxin would diminish RANKL expression measured as RNA message and also as RANKL protein. We, therefore, investigated

Figure 3. Periodontal bone resorption mediated by infiltrating immune cells in gingival periodontal tissues. Macrophages stimulated by LPS from Gram-negative organisms commonly associated with periodontal disease microflora give rise to tumor necrosis factor (TNF)-a, interleukin (IL)-1 and IL-6 synthesis among other cytokines, along with other calcitropic factors such as vitamin D-3, prostaglandin, and E-2 glucocorticoid; these cytokines and factors induce RANKL expression on osteoblasts. RANKL binding to RANK expressed on hematopoietic progenitors of osteoclasts (osteoclast precursor) activates a signal transduction cascade that leads to osteoclastogenesis in the presence of survival factor MCSF. OPG produced by osteoblasts is a decoy receptor for RANKL, which inhibits osteoclastogenesis and osteoclast activation by binding to RANKL. Similarly, activated T cells can affect bone physiology by producing cytokines such as TNF-a, IL-11, and IL-17 that lead to RANKL expression on osteoblasts (T cell-predicated indirect stimulation of bone resorption). Activated macrophages (e.g., macrophage MHC Class II + antigen) activate T cells that express and produce RANKL which directly induces osteoclast formation and activation. Recently, B cells have also been shown27 to directly express RANKL in cell-bound and soluble forms. OPG (soluble decoy receptor for RANKL) blocks all pathways (indirect and direct T cell or B cell [osteoblast/stromal cell]). Both types of RANKL can combine with RANK to give rise to osteoclastogenesis. The reactions depicted above indicate the importance of both types of immune cells in the genesis of osteoclastogenesis. These reactions occur when infiltrates of immunologically activated cells enter the gingival/periodontal tissues and produce osteoclasis.

J Periodontol • November 2005 (Suppl.) Taubman, Valverde, Han, Kawai

2037

the effects of interfering with antigen-specific T cell stimulation by the treatment of T lymphocytes with kaliotoxin (a potassium channel inhibitor derived from scorpion venom) in vitro.6 (Table 3). Such in vitro treatment resulted in dramatically inhibited antigen- stimulated Th1 cell proliferation, marked decreases in T cell RANKL expression, and in T cell-mediated induction of RANKL-dependent osteoclastogenesis. We monitored and demonstrated the requirement for antigen-specific activation of T cell clones, for Kv1.3 expression (voltage-gated potassium channels), and on the mRNA and protein levels in two different Th1 T-cell clones. Further antigen-specific proliferative acti- vation of T clone cells could be dramatically inhibited by the Kv1.3 selective blocker, kaliotoxin.6 This finding has also recently been demonstrated for encephalitogenic myelin-basic protein-specific rat T clone cells.32 Such a selective blockage was shown to improve experi- mental autoimmune encephalitis.32,33

Importantly, to investigate the role of Kv1.3 in po- tential regulation of RANKL expression by antigen- activated T cells, we studied the effects of Kv1.3 blockade with kaliotoxin on T cell-dependent bone re- sorption in vivo (Table 3).6 The study demonstrated that kaliotoxin could not only downregulate T cell activation parameters (including antigen-specific proliferation and antibody production of the IgG2a isotype), but also could dramatically reduce Th1-me- diated bone resorption in the rat antigen challenge/ adoptive transfer model of periodontal disease. A highly significant 84% decrease in periodontal bone resorption compared to controls was observed.34

In addition, we demonstrated that both RANKL and OPG were reduced in vivo and in vitro after treatment with kaliotoxin at the mRNA and protein levels.6 How- ever, the normalized RANKL/OPG ratios were also markedly reduced after treatment with kaliotoxin. These findings suggested that Kv1.3 potassium chan- nels can contribute to regulation of RANKL expression and potentially to disease-associated periodontal bone resorption. This was tested on antigen-activated

Th1 clone cells and we demon- strated that kaliotoxin or OPG could inhibit multinucleated tar- trate-resistant acid phospha- tase-positive (TRAP+) cells, and reduce bone resorption pit for- mation on osteologic disks in a like manner.26 In the rodent model, activated T lymphocytes, LPS, and antigen (bacterial), are all required for periodontal bone resorption, which is accompanied by an upregulation of RANKL expression on immune cells in gingiva and subsequent osteo-

clastogenesis.6 Potassium channels have been dem- onstrated to be involved in a wide variety of functions, including regulation of membrane potential, signal trasduction, insulin secretion, and hormone and im- mune responses.8 Therefore, the linkage between po- tassium channel blockade and RANKL-dependent osteoclastogenesis implicated in periodontal disease progression and/or inflammatory bone resorption calls for further characterization. However, it is important to note that inhibition of potassium channels and subse- quent reduction in T cell activation led to clearly dimin- ished RANKL expression, increased OPG, and reduced osteoclastogenesis, in vivo and in vitro. The effects of lo- cal or systemic administration of blockers may reveal underlying mechanisms of bone resorption.

THE RODENT MODEL OF B CELL-MEDIATED PERIODONTAL BONE RESORPTION IS RANKL DEPENDENT

We recently performed studies of adoptive transfer of antigen-specific B cells into T cell-deficient (congen- itally athymic) rats in order to investigate whether RANKL-expressing B cells can also produce bone re- sorption in the absence of T cells.35 The purpose of the study was to evaluate A. actinomycetemcomitans- responsive B lymphocytes on their level of RANKL expression and their effects on periodontal bone re- sorption. Normal Rowett rats were used as donors af- ter intraperitoneal immunization with formalin-fixed A. actinomycetemcomitans and spleen B cells were greatly enriched using fluorescent-activated cell sort- ing (95% to 98% purity). Congenitally athymic Rowett rats received injections of formalin-fixed A. actinomy- cetemcomitans into the gingival papillae and donor B cells were transferred via tail vein injection. We demonstrated that B cells from A. actinomycetemco- mitans-immunized animals had greater levels of RANKL expression and induced a significantly higher level of osteoclast differentiation in vitro. Im- portantly, antigen specificity was a necessary compo- nent of the transferred B cells. Antigen-specific

Table 3.

Effects of Kaliotoxin Blockade on T Cell-Mediated Experimental Bone Resorption6

A. Downregulation of antigen-specific cell proliferation and level of IgG2a isotype antibody.

B. Reduction of RANKL and OPG at the mRNA and protein levels.

C. Reduction of normalized RANKL/OPG ratios at the mRNA and protein levels.

D. Inhibition of multinucleated tartrate-resistant acid phosphatase-positive (TRAP+) cells and reduction of bone resorption pit formation on osteologic disks.

E. Dramatic reduction (84%) in periodontal bone resorption compared to controls.

The Immune Response in Periodontal Disease Volume 76 • Number 11 (Suppl.)

2038

RANKL-expressing B cells could be recovered from the gingival tissues of recipient rats transferred with A. actinomycetemcomitans-binding B cells (but not from recipients of phosphate buffered saline, non-immune B cells, or A. actinomycetemcomitans non-binding B cells). Recipients of antigen-specific, RANKL-expressing B cells exhibited increased osteo- clast formation on the alveolar bone surface and sig- nificant periodontal bone resorption. This effect was antagonized by the injection of the RANKL decoy re- ceptor osteoprotegerin (OPG Fc) into the local gingi- val tissues.

In summary, animals receiving antigen-specific RANKL-expressing B cells demonstrated induction of in vitro osteoclastogenesis, and significantly ele- vated experimentally induced bone resorption, in ac- cordance with the emergence of osteoclasts adjacent to the alveolar crest. Control animals receiving B cells with far less RANKL expression showed little or no bone resorption. This suggests that B lymphocytes can contribute to increased periodontal bone resorp- tion in the absence of T lymphocytes, and that the effect is associated with the upregulation of RANKL expression. These and previous experiments demon- strate quite clearly that periodontal disease bone re- sorption signs can be related to RANKL expression by B cells and/or by T cells, and subsequent osteo- clastogenesis. In many respects, periodontal disease is an inflammatory disease mediated by immune cells and their products (Fig. 3).

RANKL AND OPG IN HUMAN PERIODONTAL DISEASE TISSUES

Recently, we investigated RANKL and OPG mRNA levels in healthy gingival and periodontal tissues and in tissues from patients with chronic periodonti- tis.34 Gingival tissues of patients with chronic peri- odontitis demonstrated qualitative elevation of RANKL mRNA levels and depressed OPG mRNA lev-

els. By contrast, this finding can be compared to the marked reductions in RANKL/OPG ratios observed in the kaliotoxin experiments.6 The RANKL/OPG ratio seems to regulate bone loss or bone apposition.3

The reduced ratios strongly implicated the immune response and T cell RANKL expression in osteoclasto- genesis in rat T cell-induced periodontal bone resorp- tion and now in human periodontal disease. We also investigated the expression of RANKL in immune cells in gingival and periodontal tissues from chronic peri- odontitis patients by confocal microscopy.36 These additional studies, involving immunohistochemical analyses and confocal microscopy of the periodontal lesions, indicated abundant RANKL expression on T cells identified in the lesion37 and also B lymphocytes in the lesion with little or no expression of RANKL by macrophages of patients with chronic periodontitis. In- terestingly, the percentage of RANKL-positive T cells (50% to 60%) in diseased periodontal lesions was sub- stantially less than the percentage of RANKL-positive B cells (>90%). It is not clear if this indicates that B cells can contribute more extensively to bone resorp- tion than T cells in the lesions, or if this is not indicative of any cellular effect and unrelated to bone resorption potential. Furthermore, we could demonstrate that gingival mononuclear cells from periodontal lesions of patients induced osteoclastogenesis in vitro giving rise to multinucleated TRAP+ osteoclast-like cell dif- ferentiation from mouse osteoclast precursor cells (MOCP-5). These cells were also capable of demon- strating the formation of resorption pits in osteogenic models. Therefore, a compelling case can be made that immune cells infiltrating periodontal lesions can stimulate osteoclastic differentiation and give rise to subsequent periodontal bone resorption, the key sign of periodontal disease. The significance of these find- ings in periodontal disease is far-reaching, consider- ing that from such discoveries, the actual nature of periodontal disease can be challenged on this basis.

Table 4.

Summary of Evidence That Periodontal Bone Resorption Is Mediated by Host Immune Response

Rodent Studies

A. T helper lymphocytes are involved in periodontal bone resorption.26

B. In vivo inhibition of RANKL function with the decoy receptor OPG diminishes alveolar bone destruction.5

C. Local antigen-specific activation of Th1-type T cells by B7 costimulation triggers inflammatory bone resorption.9

D. Inhibition of T cell activation by potassium channel blockade reduces ratio of RANKL/OPG expression and decreases experimental bone resorption.6

Human Investigations

A. Increased expression of RANKL by T lymphocytes in diseased periodontal tissues.36

B. Increased expression of RANKL by B lymphocytes in diseased periodontal tissues (unpublished observations).

C. Human lymphocytes from diseased periodontal tissues induce osteoclast differentiation (unpublished observations).

J Periodontol • November 2005 (Suppl.) Taubman, Valverde, Han, Kawai

2039

The most compelling evidence now indicates that periodontal disease is not a conventional infectious disease, but is in fact an inflammatory disease, triggered by host immune response to a constellation of periodontal biofilm-associated microorganisms. These inflammatory cells can infiltrate gingival tissues in an antigen-specific manner.26 Activation of these cells gives rise to expression of RANKL within the peri- odontal tissues, which leads to the induction of osteo- clastogenesis. Maturation of osteoclasts adjacent to the alveolar crest gives rise to substantial periodontal bone resorption (Fig. 3).

CLINICAL IMPLICATIONS

Both human and experimental animal studies support the hypothesis that there is considerable potential for intervention or interference with periodontitis by inter- ference with the host immune system (Table 4). As we begin to gradually understand the critical role of host responses involving immune cells in the progres- sion of periodontal disease, efforts are being made to evaluate strategies aimed at interference with the det- rimental effects of T and B cell activation to ameliorate periodontal bone resorption. These may include: 1) interference with T cell antigen-presenting cell inter- actions; 2) stimulation of protective T cell responses; 3) interference with RANKL-RANK interaction; and 4) interference with T cell activation by decreasing RANKL expression and RANKL/OPG ratio. Interfer- ence with these processes, in a variety of ways as de- scribed above, should contribute to the amelioration of periodontal bone resorption and to prevention of periodontal disease progression.

ACKNOWLEDGMENT

This study was supported by grants DE03420 and DE014551 from the National Institute of Dental and Craniofacial Research.

REFERENCES 1. Socransky SS, Haffajee AD. The nature of periodontal

diseases. Ann Periodontol 1997;2:3-10. 2. Yasuda H, Shima N, Nakagawa N, et al. Osteoclast

differentiation factor is a ligand for osteoprotegerin/ osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA 1998;95: 3597-3602.

3. Kong YY, Feige U, Sarosi I, et al. Activated T cells regulate bone loss and joint destruction in adjuvant arthritis through osteoprotegerin ligand. Nature 1999; 402:304-309.

4. Oshiro T, Shiotani A, Shibasaki Y, Sasaki T. Osteo- clast induction in periodontal tissue during experimen- tal movement of incisors in osteoprotegerin-deficient mice. Anat Rec 2002;266:218-225.

5. Teng YT, Nguyen H, Gao X, et al. Functional human T-cell immunity and osteoprotegerin ligand control alveolar bone destruction in periodontal infection. J Clin Invest 2000;106:R59-R67.

6. Valverde P, Kawai T, Taubman MA. Selective block- ade of voltage-gated potassium channels reduces inflammatory bone resorption in experimental peri- odontal disease. J Bone Miner Res 2004;19:155-164.

7. Okahashi N, Sakurai A, Nakagawa I, et al. Infection by Streptococcus pyogenes induces the receptor activator of NF-kB ligand expression in mouse osteoblastic cells. Infect Immun 2003;71:948-955.

8. Hille B. Ionic Channels of Excitable Membranes. Sinauer Associates, Sunderland, MA. 2001; 814 p.

9. Kawai T, Eisen-Lev R, Seki M, Eastcott JW, Wilson ME, Taubman MA. Requirement of B7 costimulation for Th1-mediated inflammatory bone resorption in experimental periodontal disease. J Immunol 2000; 164:2102-2109.

10. Ebersole JL, Taubman MA, Smith DJ, Frey DE, Haffajee AD, Socransky SS. Human serum antibody responses to oral microorganisms. IV. Correlation with homologous infection. Oral Microbiol Immunol 1987; 2:53-59.

11. Taubman MA, Ebersole JL, Smith DJ. Association between systemic and local antibody and periodontal disease. In: Mergenhagan SE, ed. Host Parasite Inter- actions in Periodontal Diseases. Washington, DC: American Society for Microbiology; 1982:283-298.

12. Ivanyi L, Lehner T. Stimulation of lymphocyte trans- formation by bacterial antigens in patients with peri- odontal disease. Arch Oral Biol 1970;15:1089-1096.

13. Yoshie H, Taubman MA, Olson CL, Ebersole JL, Smith DJ. Periodontal bone loss and immune characteristics after adoptive transfer of Actinobacillus-sensitized T cells to rats. J Periodontal Res 1987;22:499-505.

14. Taubman MA, Buckelew JM, Ebersole JL, Smith DJ. Periodontal bone loss in ovalbumin sensitized germ- free rats fed antigen-free diet with ovalbumin. J Periodontal Res 1983;18:292-302.

15. Baker PJ, Garneau J, Howe L, Roopenian DC. T-cell contributions to alveolar bone loss in response to oral infection with Porphyromonas gingivalis. Acta Odon- tol Scand 2001;59:222-225.

16. Baker PJ, Evans RT, Roopenian DC. Oral infection with Porphyromonas gingivalis and induced alveolar bone loss in immunocompetent and severe combined immu- nodeficient mice. Arch Oral Biol 1994;39:1035-1040.

17. Persson GR, Engel D, Whitney C, et al. Immunization against Porphyromonas gingivalis inhibits progression of experimental periodontitis in nonhuman primates. Infect Immun 1994;62:1026-1031.

18. Yamashita K, Eastcott JW, Taubman MA, Smith DJ, Cox DS. Effect of adoptive transfer of cloned Actino- bacillus actinomycetemcomitans-specific T helper cells on periodontal disease. Infect Immun 1991;59:1529- 1534.

19. Eastcott JW, Yamashita K, Taubman MA, Smith DJ. Characterization of rat T-cell clones with bacterial specificity. Immunology 1990;71:120-126.

20. Eastcott JW, Yamashita K, Taubman MA, Harada Y, Smith DJ. Adoptive transfer of cloned T helper cells ameliorates periodontal disease in nude rats. Oral Microbiol Immunol 1994;9:284-289.

21. McDonald DF, Schofield BH, Prezioso EM, et al. Direct bone resorbing activity of murine myeloma cells. Can- cer Lett 1983;19:119-124.

22. Harada Y, Taubman MA, Eastcott JW, Smith DJ. Generation of B cells specific for Actinobacillus actino- mycetemcomitans. J Dent Res 1990;69(Spec. Issue): 143(Abstr. 279).

The Immune Response in Periodontal Disease Volume 76 • Number 11 (Suppl.)

2040

23. Taubman MA, Stoufi ED, Ebersole JL, Smith DJ. Phenotypic studies of cells from periodontal disease tissues. J Periodontal Res 1984;19:587-590.

24. Stoufi ED, Taubman MA, Ebersole JL, Smith DJ. Preparation and characterization of human gingival cells. J Periodontal Res 1987;22:144-149.

25. Mosmann TR, Coffman RL. TH1 and TH2 cells: Different patterns of lymphokine secretion lead to different functional properties. Annu Rev Immunol 1989;7:145-173.

26. Kawai T, Shimauchi H, Eastcott JW, Smith DJ, Taubman MA. Antigen direction of specific T-cell clones into gingival tissues. Immunology 1998;93: 11-19.

27. Choi Y, Woo KM, Ko SH, et al. Osteoclastogenesis is enhanced by activated B cells but suppressed by activated CD8(+) T cells. Eur J Immunol 2001;31: 2179-2188.

28. Theill LE, Boyle WJ, Penninger JM. RANK-L and RANK: T cells, bone loss, and mammalian evolution. Annu Rev Immunol 2002;20:795-823.

29. Taubman MA, Kawai T. Involvement of T-lympho- cytes in periodontal disease and in direct and indirect induction of bone resorption. Crit Rev Oral Biol Med 2001;12:125-135.

30. Makihira S, Kawai T, Hosokawa Y, Karimbux N, Eastcott JW, Taubman MA. Role for TACE in lympho- cyte RANKL expression in periodontitis. J Dent Res 2004;83:1459.

31. Wang R, Zhang L, Zhang X, et al. Regulation of activation-induced receptor activator of NF-kappaB

ligand (RANKL) expression in T cells. Eur J Immunol 2002;32:1090-1098.

32. Beeton C, Wulff H, Barbaria J, et al. Selective block- ade of T lymphocyte K(+) channels ameliorates ex- perimental autoimmune encephalomyelitis, a model for multiple sclerosis. Proc Natl Acad Sci USA 2001;98: 13942-13947.

33. Beeton C, Barbaria J, Giraud P, et al. Selective block- ing of voltage-gated K+ channels improves experi- mental autoimmune encephalomyelitis and inhibits T cell activation. J Immunol 2001;166:936-944.

34. Valverde P, Kawai T, Taubman MA. Potassium chan- nel blockers as therapeutic agents to interfere with periodontal disease bone resorption. J Dent Res 2005; 84:488-499.

35. Han X, Kawai T, Eastcott JW, Taubman MA. Bacterial- responsive B lymphocytes induce periodontal bone resorption. J Dent Res 2005;84:2838.

36. Matsuyama T, Kawai T, Taubman MA. OPG-L expres- sion by T lymphocytes in periodontal diseased tissues. J Dent Res 2001;80:171.

37. Matsuyama T, Kawai T, Izumi Y, Taubman MA. Expression of major histocompatibility complex class II and CD80 by gingival epithelial cells induces activation of CD4+ T cells in response to bacterial challenge. Infect Immun 2005;73:1044-1051.

Correspondence: Dr. Martin A. Taubman, Department of Immunology, The Forsyth Institute, 140 The Fenway, Boston, MA 02115. E-mail: [email protected].

Accepted for publication July 15, 2005.

J Periodontol • November 2005 (Suppl.) Taubman, Valverde, Han, Kawai

2041