Summary
An immunoperoxidase technique was used to examine IP-10 (interferon-gamma inducible protein 10), RANTES (regulated on activation normal T cell expressed and secreted), MCP-1 (monocyte chemoattractant protein-1), and MIP-1alpha (macrophage inflammatory protein-1alpha) in gingival biopsies from 21 healthy/gingivitis and 26 periodontitis subjects. The samples were placed into 3 groups according to the size of infiltrate. MIP-1alpha+ cells were more abundant than the other chemokines with few MCP-1+ cells. The mean percent MIP-1alpha+ cells was higher than the percent MCP-1+ cells (P = 0·02) in group 2 (intermediate size infiltrates) lesions from periodontitis subjects, other differences not being significant due to the large variations between tissue samples. Analysis of positive cells in relation to CD4/CD8 ratios showed that with an increased proportion of CD8+ cells, the mean percent MIP-1alpha+ cells was significantly higher in comparison with the mean percent RANTES+ and MCP-1+ cells (P < 0·015). Endothelial cells were MCP-1+ although positive capillaries were found on the periphery of infiltrates only. Keratinocyte expression of chemokines was weak and while the numbers of healthy/gingivitis and periodontitis tissue sections positive for IP-10, RANTES and MCP-1 reduced with increasing inflammation, those positive for MIP-1alpha remained constant for all groups. In conclusion, fewer leucocytes expressed MCP-1 in gingival tissue sections, however, the percent MIP-1alpha+ cells was increased particularly in tissues with increased proportions of CD8 cells and B cells with increasing inflammation and also in tissues with higher numbers of macrophages with little inflammation. Further studies are required to determine the significance of MIP-1alpha in periodontal disease.
Introduction
Periodontitis results from the inflammatory response to bacterial challenge in the gingival crevicular area [1]. The inflammatory infiltrate which develops in the gingival tissues is dominated by lymphocytes and macrophages. Neutrophils are also present in the inflammatory infiltrate, although the majority migrate through the connective tissues and the epithelium to come in contact with plaque bacteria [2]. Seymour and Greenspan [3] were the first to hypothesize that together with a change in plaque microbial composition, the transition from a stable gingivitis to a progressive periodontitis lesion is accompanied by a shift in lymphocyte populations in the inflammatory infiltrate from a T cell dominant lesion with few B cells to one with an increased proportion of B cells and plasma cells. More recently, Zappa et al. [4,5] reported that the numbers of several different cell populations including plasma cells, lymphocytes and total numbers of inflammatory cells increased in sites undergoing periodontal tissue destruction compared with stable sites, although the lymphocyte subset populations were not determined.
The regulation of leucocyte migration into and through the tissues is determined by the expression of adhesion molecules on firstly endothelial cells and on other cells such as keratinocytes which are induced by pro-inflammatory cytokines as well as to a group of cytokines with chemotactic properties, the chemokines. Chemokines are responsible for the recruitment and subsequent activation of particular leucocytes into inflamed tissues [6] and therefore play a central role in the final outcome of the immune response by determining which subsets of leucocytes form the inflammatory infiltrate.
While few studies on macrophages in periodontal disease have been reported, their numbers have been reported to increase in gingivitis and periodontitis, although, compared with other cell types, their proportion remains low [2]. We have recently demonstrated that one major periodontopathic bacterium, Porphyromonas gingivalis inhibits the production of the chemokine MCP-1 by P. gingivalis-specific T cells, monocytes and B cells [7]. P. gingivalis also inhibits neutrophil chemotaxis [8] and may inhibit the influx and activation of monocytes/macrophages [9] leading to an overall reduction in innate immunity. While there is still controversy over whether particular T cell subsets predominate in periodontal disease, Th1 and Th2 cells do differ in their migratory properties and chemotactic responsiveness [10].
Studies of chemokines are currently being undertaken to further the understanding of their roles played in the pathogenesis of a number of diseases because of their potential use as targets for therapy. Although there have been a number of reports on the expression of adhesion molecules in periodontal disease [11–13], this is not the case for chemokine expression. A study of chemokines in periodontal disease may help to define the particular subsets of lymphocytes which are recruited and then activated locally in the gingival tissues during different stages of the disease. The aim of this study was to investigate the expression of chemokines in gingival tissue biopsies from gingivitis and periodontitis patients exhibiting different degrees of inflammation.
Materials and methods
Patients
Gingival tissue was obtained from 47 subjects undergoing periodontal surgery for disease and nondisease related reasons. Twenty-one biopsies classified as either healthy or gingivitis were taken from patients undergoing surgery for nondisease related reasons such as crown lengthening and displayed minimum periodontal disease (probing depths of < 4 mm) and generally exhibited minimal bleeding upon probing. The description of healthy gingival tissue is a clinical one only and while these subjects did not show clinical signs of disease, histological evidence of inflammation was found in all samples and therefore in this histological study, these healthy tissue sections were grouped together with the gingivitis tissues which displayed both clinical and histological inflammation as a healthy/gingivitis group. Studies using an experimental gingivitis model have clearly demonstrated that histological inflammation was present not only prior to the accumulation of plaque resulting in overt clinical signs of inflammation, but was often of quite a significant extent [14,15]. This grouping of healthy/gingivitis tissue samples in histological studies has been published previously [11,16]. Tissue obtained from 26 patients showed moderate to advanced disease and were classified as a periodontitis group with probing depths > 4 mm, the majority (19/26) of which were > 6 mm. All subjects in this group had previous oral hygiene instruction and scaling and root planing prior to surgery, but continued to demonstrate bleeding on probing. Each of the healthy/gingivitis and periodontitis groups were divided into three subgroups as described under the heading ‘cell analysis’ and all samples in subgroup 3 were taken from sites which exhibited bleeding on probing irrespective of whether they were in the healthy/gingivitis or periodontitis categories. A written explanation of the purpose of the study and signed consent according to the Helsinki Declaration to use tissue which would otherwise have been discarded was obtained at the time of surgery. Institutional ethics review committee approval to carry out the study was also obtained.
Preparation of tissue
Immediately after surgery, the tissue was embedded in OCT, quenched and stored in liquid nitrogen. Six µm serial cryostat sections were cut from each specimen, air dried for 2 h, fixed in equal parts chloroform/acetone for 5 min and stored at −20°C [17].
Immunoperoxidase technique
IP-10, RANTES, MCP-1 and MIP-1alpha were labelled using an avidin-biotin immunoperoxidase method described previously [11,16,18]. Briefly, after rehydration in phosphate-buffered saline pH 7·2 (PBS), the sections were depleted of endogenous peroxidase using 0·3% hydrogen peroxide/0·1% sodium azide in PBS after which the sections were incubated with the following primary monoclonal antibodies, mouse antihuman IP-10, RANTES, MCP-1 and MIP-1alpha (PharMingen, San Diego, CA) at dilutions of 1/20. This was followed by biotinylated rabbit antimouse immunoglobulins (DAKO, Glostrup, Denmark) and finally streptavidin peroxidase (DAKO) both at dilutions of 1/100. The peroxidase was developed using 0·05%, 3,3′-diaminobenzidine (Sigma Chemical Co., St Louis, MO) in Tris-HCl buffer pH 7·6 containing 0·01% hydrogen peroxide. Nuclei were counterstained with Mayer's haematoxylin. The same technique was employed to stain the sections for CD3+ T cells, CD4 and CD8 T cell subsets, macrophages and B cells using mouse antihuman CD3, CD4, CD8, CD14 and CD19 monoclonal bodies, respectively (DAKO). Tonsil sections prepared in the same way as the gingival biopsies were used as positive controls and negative controls included the use of PBS in place of the primary antibody.
Cell analysis
Qualitative assessment was undertaken using haematoxylin and eosin stained sections. Each health/gingivitis or periodontitis biopsy was evaluated on the size of the connective tissue inflammatory infiltrate as first described by Seymour et al.[14]. Biopsies with small infiltrates confined to the upper 1/3 of the section in the region adjacent to the junctional and sulcular epithelium, were placed into group 1. Group 2 contained sections with infiltrates which occupied the upper 2/3 of the section and group 3 infiltrates extended throughout the entire section [11–14,16,19].
High power fields (×400) accounting for the entire area of the infiltrate in each section were viewed. Total numbers of mononuclear cells within each field and the total numbers with positive membrane staining were counted. The percent IP-10+, RANTES+, MCP-1+ and MIP-1alpha+ cells in the inflammatory infiltrates were determined and analysed according to the clinical status and size of infiltrate. The percent CD3+, CD4+, CD8+, CD14+ and CD19+ cells were also determined. Expression of costimulatory molecules was noted on endothelial cells and keratinocytes.
Statistical analysis
Multivariate analysis of variance using the general linear model was used to test for differences between the expression of chemokines in the subgroups of the healthy/gingivitis and periodontitis groups and between the groups compiled according to the CD4/CD8 and CD14/CD19 ratios. Selected pairs of groups were then tested for significance using the student t-test. A significance level of 0·02 was determined to reduce the probability of significant differences occurring by chance. The Minitab statistical package (Mininc., State College, PA) was used to perform the analyses.
Results
Chemokine expression in connective tissue inflammatory infiltrates
Biopsies from healthy/gingivitis and periodontitis individuals were graded into 3 groups on the size of infiltrate as outlined in the materials and methods. The mean percent CD3+ T cells, CD4 and CD8 T cell subsets and CD14+ macrophages did not vary significantly between the tissues from healthy/gingivitis and periodontitis subjects nor between lesions with different sized infiltrates Table 1. However, the percent B cells increased with increasing size of infiltrate from group 1 to group 3 (extensive infiltrates) (P = 0·001) in periodontitis tissues. The percent B cells was also increased in group 3 infiltrates in periodontitis tissues in comparison with healthy/gingivitis tissues (P = 0·004).
Mean percentage (± standard error of the mean) T cells (CD3), CD4 and CD8 T cell subsets, macrophages (CD14) and B cells (CD19) in gingival tissues from healthy/gingivitis and adult periodontitis subjects
| Clinical status | % CD3+ cells | % CD4+ cells | % CD8+ cells | % CD14+ cells | % CD19+ cells |
| H/G–group 1 | 28·36 ± 5·20 | 27·15 ± 6·22 | 9·30 ± 2·69 | 5·11 ± 1·51 | 5·53 ± 2·53 |
| H/G-group 2 | 32·47 ± 4·47 | 21·69 ± 5·14 | 10·62 ± 2·45 | 3·23 ± 1·46 | 14·83 ± 2·59 |
| H/G group 3 | 34·02 ± 5·66 | 28·73 ± 4·84 | 12·26 ± 2·51 | 7·68 ± 3·05 | 7·98 ± 1·24 |
| AP-group 1 | 21·78 ± 5·46 | 16·95 ± 3·81 | 7·63 ± 1·97 | 5·61 ± 2·28 | 5·43 ± 2·14 |
| AP-group 2 | 37·96 ± 7·07 | 28·31 ± 5·37 | 14·03 ± 3·07 | 4·19 ± 1·39 | 16·62 ± 3·90 |
| AP-group 3 | 23·38 ± 3·63 | 19·01 ± 2·64 | 9·36 ± 1·06 | 2·60 ± 0·81 | 20·65 ± 2·84 |
| Clinical status | % CD3+ cells | % CD4+ cells | % CD8+ cells | % CD14+ cells | % CD19+ cells |
| H/G–group 1 | 28·36 ± 5·20 | 27·15 ± 6·22 | 9·30 ± 2·69 | 5·11 ± 1·51 | 5·53 ± 2·53 |
| H/G-group 2 | 32·47 ± 4·47 | 21·69 ± 5·14 | 10·62 ± 2·45 | 3·23 ± 1·46 | 14·83 ± 2·59 |
| H/G group 3 | 34·02 ± 5·66 | 28·73 ± 4·84 | 12·26 ± 2·51 | 7·68 ± 3·05 | 7·98 ± 1·24 |
| AP-group 1 | 21·78 ± 5·46 | 16·95 ± 3·81 | 7·63 ± 1·97 | 5·61 ± 2·28 | 5·43 ± 2·14 |
| AP-group 2 | 37·96 ± 7·07 | 28·31 ± 5·37 | 14·03 ± 3·07 | 4·19 ± 1·39 | 16·62 ± 3·90 |
| AP-group 3 | 23·38 ± 3·63 | 19·01 ± 2·64 | 9·36 ± 1·06 | 2·60 ± 0·81 | 20·65 ± 2·84 |
H/G, healthy/gingivitis; AP, adult periodontitis
Group 1, small infiltrates; Group 2, intermediate infiltrates; Group 3, extensive infiltrates
Mean percentage (± standard error of the mean) T cells (CD3), CD4 and CD8 T cell subsets, macrophages (CD14) and B cells (CD19) in gingival tissues from healthy/gingivitis and adult periodontitis subjects
| Clinical status | % CD3+ cells | % CD4+ cells | % CD8+ cells | % CD14+ cells | % CD19+ cells |
| H/G–group 1 | 28·36 ± 5·20 | 27·15 ± 6·22 | 9·30 ± 2·69 | 5·11 ± 1·51 | 5·53 ± 2·53 |
| H/G-group 2 | 32·47 ± 4·47 | 21·69 ± 5·14 | 10·62 ± 2·45 | 3·23 ± 1·46 | 14·83 ± 2·59 |
| H/G group 3 | 34·02 ± 5·66 | 28·73 ± 4·84 | 12·26 ± 2·51 | 7·68 ± 3·05 | 7·98 ± 1·24 |
| AP-group 1 | 21·78 ± 5·46 | 16·95 ± 3·81 | 7·63 ± 1·97 | 5·61 ± 2·28 | 5·43 ± 2·14 |
| AP-group 2 | 37·96 ± 7·07 | 28·31 ± 5·37 | 14·03 ± 3·07 | 4·19 ± 1·39 | 16·62 ± 3·90 |
| AP-group 3 | 23·38 ± 3·63 | 19·01 ± 2·64 | 9·36 ± 1·06 | 2·60 ± 0·81 | 20·65 ± 2·84 |
| Clinical status | % CD3+ cells | % CD4+ cells | % CD8+ cells | % CD14+ cells | % CD19+ cells |
| H/G–group 1 | 28·36 ± 5·20 | 27·15 ± 6·22 | 9·30 ± 2·69 | 5·11 ± 1·51 | 5·53 ± 2·53 |
| H/G-group 2 | 32·47 ± 4·47 | 21·69 ± 5·14 | 10·62 ± 2·45 | 3·23 ± 1·46 | 14·83 ± 2·59 |
| H/G group 3 | 34·02 ± 5·66 | 28·73 ± 4·84 | 12·26 ± 2·51 | 7·68 ± 3·05 | 7·98 ± 1·24 |
| AP-group 1 | 21·78 ± 5·46 | 16·95 ± 3·81 | 7·63 ± 1·97 | 5·61 ± 2·28 | 5·43 ± 2·14 |
| AP-group 2 | 37·96 ± 7·07 | 28·31 ± 5·37 | 14·03 ± 3·07 | 4·19 ± 1·39 | 16·62 ± 3·90 |
| AP-group 3 | 23·38 ± 3·63 | 19·01 ± 2·64 | 9·36 ± 1·06 | 2·60 ± 0·81 | 20·65 ± 2·84 |
H/G, healthy/gingivitis; AP, adult periodontitis
Group 1, small infiltrates; Group 2, intermediate infiltrates; Group 3, extensive infiltrates
The percent chemokine positive cells was quite low Fig. 1 and 2 and varied greatly between tissue sections regardless of the size of infiltrate. In most cases single stained cells were found although small clusters of positive cells were noted on occasion. The numbers of MCP-1+ cells were notably fewer than other chemokine positive cells with 30/47 of the tissue sections demonstrating less than 1% positive cells. Although the percent MCP-1+ and RANTES+ cells appeared to decrease with increasing inflammation in periodontitis and healthy/gingivitis tissues, respectively, these differences were not significant. MIP-1alpha+ cells Fig. 3 were more abundant than the other three chemokines and the mean percent MIP-1alpha+ cells in group 2 lesions from periodontitis subjects was significantly higher than the percent MCP-1+ cells (P = 0·02) Fig. 2. While the mean percent MIP-1alpha+ cells in group 3 periodontitis lesions appeared to be higher than in groups 1 or 2, there were no significant differences nor in relation to the other three chemokines due to the large variations between tissue samples.
The mean (± standard error of the mean) percentage IP-10+ (░), RANTES+ (▪), MCP-1+ () and MIP-1alpha+ (□) cells in the inflammatory infiltrates of gingival lesions (graded on the size of infiltrate as small in group 1, intermediate in group 2 and extensive in group 3) from healthy/gingivitis (H/G) subjects.
The mean (± standard error of the mean) percentage IP-10+ (░), RANTES+ (▪), MCP-1+ () and MIP-1alpha+ (□) cells in the inflammatory infiltrates of gingival lesions (graded on the size of infiltrate as small in group 1, intermediate in group 2 and extensive in group 3) from periodontitis (AP) subjects.
a, Numerous MIP-1alpha+ cells in an extensive inflammatory infiltrate (group 3) in the gingival tissue from a periodontitis patient (original magnification ×100). b, Virtually no MCP-1+ cells in the area of section.
The results shown in Table 1 highlight the variation in composition of infiltrating cells between individual gingival tissue samples. Additionally, it is not possible to define periodontal lesions as stable or progressing sites. This can only be determined at later dates by the attending clinicians who decide whether the disease has progressed or not. Therefore, as well as analysis of the data with respect to clinical status, the results were analysed additionally with respect to histological parameters of disease status including increased numbers of CD8 cells in relation to CD4 cells and increasing numbers of B cells relative to macrophages regardless of clinical status and size of infiltrate. The tissues were arranged in descending order of their CD4/CD8 and CD14/CD19 ratios regardless of clinical status and size of infiltrate and each list divided again into 3 groups. Group 1 contained sections exhibiting the highest ratios, group 2 had intermediate ratios and group 3 contained lesions with the lowest ratios. The mean CD4/CD8 ratio reduced from group 1 to group 2 (P = 0·000) and group 2 to group 3 (P = 0·000) and with the increased proportion of CD8+ cells, the mean percent MIP-1alpha+ cells in group 3 was significantly higher in comparison with the mean percent RANTES+ cells (P = 0·014) and MCP-1+ cells (P = 0·012) and the mean percent MIP-1alpha+ cells in group 2 was higher than the mean percent MCP-1+ cells although not quite significant at the 0·02 level (P = 0·022) Table 2.
Mean percentage (± standard error of the mean) chemokine positive cells relative to the CD4/CD8 ratios and CD14/CD19 ratios
| Group 1 (n = 15) | Group 2 (n = 16) | Group 3 (n = 16) | |
| CD4/CD8 ratio | 6·13 ± 0·86* | 2·21 ± 0·74* | 1·40 ± 0·05* |
| % IP-10+ cells | 1·72 ± 0·53 | 3·53 ± 1·11 | 2·90 ± 0·90 |
| % RANTES+ cells | 2·35 ± 1·55 | 2·10 ± 0·52 | 1·92 ± 0·53† |
| % MCP-1+ cells | 1·13 ± 0·33 | 0·93 ± 0·34‡ | 1·58 ± 0·59† |
| % MIP-1α+ cells | 6·03 ± 3·60 | 3·81 ± 1·14‡ | 8·19 ± 2·40† |
| CD14/CD19 ratio | 4·24 ± 1·33§ | 0·38 ± 0·06§ | 0·08 ± 0·01§ |
| % IP-10+ cells | 3·29 ± 1·05 | 2·76 ± 0·75 | 2·15 ± 0·85 |
| % RANTES+ cells | 1·39 ± 0·48 | 3·36 ± 1·56 | 1·64 ± 0·39 |
| % MCP-1+ cells | 1·41 ± 0·49 | 1·38 ± 0·46¶ | 0·84 ± 0·35 |
| % MIP-1α+ cells | 4·79 ± 1·63 | 5·37 ± 1·65¶ | 7·68 ± 3·67 |
| Group 1 (n = 15) | Group 2 (n = 16) | Group 3 (n = 16) | |
| CD4/CD8 ratio | 6·13 ± 0·86* | 2·21 ± 0·74* | 1·40 ± 0·05* |
| % IP-10+ cells | 1·72 ± 0·53 | 3·53 ± 1·11 | 2·90 ± 0·90 |
| % RANTES+ cells | 2·35 ± 1·55 | 2·10 ± 0·52 | 1·92 ± 0·53† |
| % MCP-1+ cells | 1·13 ± 0·33 | 0·93 ± 0·34‡ | 1·58 ± 0·59† |
| % MIP-1α+ cells | 6·03 ± 3·60 | 3·81 ± 1·14‡ | 8·19 ± 2·40† |
| CD14/CD19 ratio | 4·24 ± 1·33§ | 0·38 ± 0·06§ | 0·08 ± 0·01§ |
| % IP-10+ cells | 3·29 ± 1·05 | 2·76 ± 0·75 | 2·15 ± 0·85 |
| % RANTES+ cells | 1·39 ± 0·48 | 3·36 ± 1·56 | 1·64 ± 0·39 |
| % MCP-1+ cells | 1·41 ± 0·49 | 1·38 ± 0·46¶ | 0·84 ± 0·35 |
| % MIP-1α+ cells | 4·79 ± 1·63 | 5·37 ± 1·65¶ | 7·68 ± 3·67 |
CD4/CD8 ratio significantly reduced from group 1 to group 2 (P = 0·000) and group 2 to group 3 (P = 0·000);
MIP-1alpha+ cells increased in comparison with RANTES (P = 0·014) and MCP-1 (P = 0·012) positive cells in group 3;
MIP-1alpha+ cells increased compared with MCP-1+ cells in group 2 although not quite significant at the 0·02 level (P = 0·022);
CD4/CD19 ratio significantly reduced from group 1 to group 2 (P = 0·006) and group 2 to group 3 (P = 0·000);
MIP-1alpha+ cells increased compared with MCP-1+ cells in group 2 (P = 0·026), although not significant at the 0·02 level.
Mean percentage (± standard error of the mean) chemokine positive cells relative to the CD4/CD8 ratios and CD14/CD19 ratios
| Group 1 (n = 15) | Group 2 (n = 16) | Group 3 (n = 16) | |
| CD4/CD8 ratio | 6·13 ± 0·86* | 2·21 ± 0·74* | 1·40 ± 0·05* |
| % IP-10+ cells | 1·72 ± 0·53 | 3·53 ± 1·11 | 2·90 ± 0·90 |
| % RANTES+ cells | 2·35 ± 1·55 | 2·10 ± 0·52 | 1·92 ± 0·53† |
| % MCP-1+ cells | 1·13 ± 0·33 | 0·93 ± 0·34‡ | 1·58 ± 0·59† |
| % MIP-1α+ cells | 6·03 ± 3·60 | 3·81 ± 1·14‡ | 8·19 ± 2·40† |
| CD14/CD19 ratio | 4·24 ± 1·33§ | 0·38 ± 0·06§ | 0·08 ± 0·01§ |
| % IP-10+ cells | 3·29 ± 1·05 | 2·76 ± 0·75 | 2·15 ± 0·85 |
| % RANTES+ cells | 1·39 ± 0·48 | 3·36 ± 1·56 | 1·64 ± 0·39 |
| % MCP-1+ cells | 1·41 ± 0·49 | 1·38 ± 0·46¶ | 0·84 ± 0·35 |
| % MIP-1α+ cells | 4·79 ± 1·63 | 5·37 ± 1·65¶ | 7·68 ± 3·67 |
| Group 1 (n = 15) | Group 2 (n = 16) | Group 3 (n = 16) | |
| CD4/CD8 ratio | 6·13 ± 0·86* | 2·21 ± 0·74* | 1·40 ± 0·05* |
| % IP-10+ cells | 1·72 ± 0·53 | 3·53 ± 1·11 | 2·90 ± 0·90 |
| % RANTES+ cells | 2·35 ± 1·55 | 2·10 ± 0·52 | 1·92 ± 0·53† |
| % MCP-1+ cells | 1·13 ± 0·33 | 0·93 ± 0·34‡ | 1·58 ± 0·59† |
| % MIP-1α+ cells | 6·03 ± 3·60 | 3·81 ± 1·14‡ | 8·19 ± 2·40† |
| CD14/CD19 ratio | 4·24 ± 1·33§ | 0·38 ± 0·06§ | 0·08 ± 0·01§ |
| % IP-10+ cells | 3·29 ± 1·05 | 2·76 ± 0·75 | 2·15 ± 0·85 |
| % RANTES+ cells | 1·39 ± 0·48 | 3·36 ± 1·56 | 1·64 ± 0·39 |
| % MCP-1+ cells | 1·41 ± 0·49 | 1·38 ± 0·46¶ | 0·84 ± 0·35 |
| % MIP-1α+ cells | 4·79 ± 1·63 | 5·37 ± 1·65¶ | 7·68 ± 3·67 |
CD4/CD8 ratio significantly reduced from group 1 to group 2 (P = 0·000) and group 2 to group 3 (P = 0·000);
MIP-1alpha+ cells increased in comparison with RANTES (P = 0·014) and MCP-1 (P = 0·012) positive cells in group 3;
MIP-1alpha+ cells increased compared with MCP-1+ cells in group 2 although not quite significant at the 0·02 level (P = 0·022);
CD4/CD19 ratio significantly reduced from group 1 to group 2 (P = 0·006) and group 2 to group 3 (P = 0·000);
MIP-1alpha+ cells increased compared with MCP-1+ cells in group 2 (P = 0·026), although not significant at the 0·02 level.
The mean CD14/CD19 ratios reduced significantly from group 1 to group 2 (P = 0·006) and group 2 to group 3 (P = 0·000) and the mean percent MIP-1alpha+ cells was higher than the percent MCP-1+ cells in group 2 although this was not significant at the 0·02 level (P = 0·026) Table 2.
Costimulatory molecule expression by endothelial cells and keratinocytes
Endothelial cells did not express IP-10, RANTES or MIP-1alpha. However, 11/47 tissue sections (4/8 and 2/6 in groups 1 and 2 healthy/gingivitis tissues, respectively, and 3/8 and 2/9 in groups 1 and 2 periodontitis tissues) demonstrated positive staining for MCP-1 Fig. 4. The numbers of positive tissue samples decreased with increasing size of infiltrate, with no positive endothelial cells being found in group 3. Numerous positive capillaries were demonstrated in positive sections although positive cells were located outside areas containing infiltrating cells and were not found within the infiltrates themselves.
a, Numerous MCP-1+ capillaries in a group 1 tissue section from a periodontitis patient demonstrating little inflammatory activity (original magnification ×200). b, Same area of section showing no MIP-1alpha+ endothelial cells.
Keratinocytes demonstrated weak to moderate cytoplasmic staining of chemokines Fig. 5. Expression occurred predominantly in the basal layers of the junctional and sulcular epithelium although oral epithelial keratinocytes sometimes displayed positive staining particularly in the case of MIP-1alpha.
a, MIP-1alpha+ keratinocytes in a group 1 tissue section from a periodontitis patient demonstrating little inflammatory activity (original magnification ×200). b, The same area in the negative control section.
Only 3/47 tissues demonstrated IP-10 expression of keratinocytes. Similarly 6/47 tissue sections were positive for RANTES and 5/47 for MCP-1. While IP-10 and MCP-1 staining was weak, that of RANTES was in some cases a little stronger. The numbers of tissue sections positive for these 3 chemokines decreased with increasing size of infiltrate Table 2. Keratinocytes in 15 of the 47 tissue sections were positive for MIP-1alpha. In contrast to the other 3 chemokines, the numbers of positive tissues did not decrease with increasing infiltrate Table 3.
Numbers of gingival tissue sections demonstrating chemokine expression of keratinocytes
| Group | IP-10 | RANTES | MCP-1 | MIP-1alpha |
| Group 1 | ||||
| Healthy/gingivitis (n = 8) | 1 | 3 | 2 | 2 |
| Periodontitis (n = 8) | 0 | 1 | 1 | 2 |
| Group 2 | ||||
| Healthy/gingivitis (n = 6) | 0 | 0 | 0 | 3 |
| Periodontitis (n = 9) | 2 | 2 | 1 | 4 |
| Group 3 | ||||
| Healthy/gingivitis (n = 6) | 0 | 0 | 1 | 2 |
| Periodontitis (n = 9) | 0 | 0 | 0 | 2 |
| Group | IP-10 | RANTES | MCP-1 | MIP-1alpha |
| Group 1 | ||||
| Healthy/gingivitis (n = 8) | 1 | 3 | 2 | 2 |
| Periodontitis (n = 8) | 0 | 1 | 1 | 2 |
| Group 2 | ||||
| Healthy/gingivitis (n = 6) | 0 | 0 | 0 | 3 |
| Periodontitis (n = 9) | 2 | 2 | 1 | 4 |
| Group 3 | ||||
| Healthy/gingivitis (n = 6) | 0 | 0 | 1 | 2 |
| Periodontitis (n = 9) | 0 | 0 | 0 | 2 |
Numbers of gingival tissue sections demonstrating chemokine expression of keratinocytes
| Group | IP-10 | RANTES | MCP-1 | MIP-1alpha |
| Group 1 | ||||
| Healthy/gingivitis (n = 8) | 1 | 3 | 2 | 2 |
| Periodontitis (n = 8) | 0 | 1 | 1 | 2 |
| Group 2 | ||||
| Healthy/gingivitis (n = 6) | 0 | 0 | 0 | 3 |
| Periodontitis (n = 9) | 2 | 2 | 1 | 4 |
| Group 3 | ||||
| Healthy/gingivitis (n = 6) | 0 | 0 | 1 | 2 |
| Periodontitis (n = 9) | 0 | 0 | 0 | 2 |
| Group | IP-10 | RANTES | MCP-1 | MIP-1alpha |
| Group 1 | ||||
| Healthy/gingivitis (n = 8) | 1 | 3 | 2 | 2 |
| Periodontitis (n = 8) | 0 | 1 | 1 | 2 |
| Group 2 | ||||
| Healthy/gingivitis (n = 6) | 0 | 0 | 0 | 3 |
| Periodontitis (n = 9) | 2 | 2 | 1 | 4 |
| Group 3 | ||||
| Healthy/gingivitis (n = 6) | 0 | 0 | 1 | 2 |
| Periodontitis (n = 9) | 0 | 0 | 0 | 2 |
Discussion
MCP-1 is a potent chemoattractant for monocytes and macrophages [20] and the results of the present study have shown that fewer leucocytes expressed MCP-1 in the connective tissue infiltrates of gingival lesions regardless of clinical status and size of infiltrate. Recently a study on macrophage populations in periodontal diseased demonstrated that chronic inflammation did not result in increased numbers of macrophages and there was little evidence of activated macrophages [21]. In another study intramuscular injections of P. gingivalis LPS in mice resulted in a reduction or absence of MCP-1 four hours later and few cells in the P. gingivalis LPS-injected muscles expressed MCP-1 mRNA [22]. The results of the present study also demonstrated that endothelial cells expressed MCP-1 however, positive capillaries were located in areas adjacent to infiltrating cells and the numbers of tissue samples exhibiting positive staining decreased with increasing size of infiltrate. This suggests that in the early stages of gingival inflammation, monocytes/macrophages may readily migrate into the tissues via MCP-1+ endothelial cells although with increasing inflammation, expression of MCP-1 is abrogated, although studies would need to determine whether this does lead to a reduction of this cell type in the tissues. In contrast to these results, Yu et al. [23] reported the expression of MCP-1 on endothelial cells as well as monocytes/macrophages in inflamed gingival tissues correlated with the degree of inflammation. The tissue samples in both studies were prepared differently in that Yu et al. [23] used paraformaldehyde fixed cryostat sections while the present study employed cryostat sections which were post fixed in acetone/chloroform. However, since the intensity of MCP-1 expression was relatively strong in both studies, the different methods of tissue processing may not have masked antigenic epitopes. Additionally, a purified monoclonal anti-MCP-1 antibody was used in the present study, whereas Yu et al. [23] detected MCP-1+ cells using a polyclonal antibody raised in a rabbit against purified baboon smooth muscle cell-derived MCP-1 which may partly explain the divergent results in the two studies. Also in the present study, there was a great variability in not only the composition of each infiltrate but the actual numbers of cells in infiltrates in the same groupings and therefore it is possible that the system used to grade the infiltrates in the present study may not have been sufficiently quantitative to result in differences between groups. In another report, Yu et al. [24] reported strong MCP-1 expression by endothelial cells and monocytes/macrophages at the periphery of perivascular infiltrates and by keratinocytes in a number of inflammatory skin diseases. It was proposed that the expression of MCP-1 may explain increased recruitment of monocytes/macrophages in cell-mediated immune responses such as delayed type hypersensitivity reactions in the skin. Reports have also related increased numbers of MCP-1 expressing cells in diseases such as as slowly progressive primary tuberculosis in mice [25] and rheumatoid synovia with active inflammation [26]. Further studies are required to determine whether periodontitis may represent a unique disease where the expression of this chemokine may decrease with increased inflammation due to factors such as its inhibition by P. gingivalis[7,22].
MIP-1alpha, like MCP-1 is a potent chemoattractant for monocytes and macrophages [27] and in the present study, the percent MIP-1alpha+ cells was significantly higher than the percent MCP-1+ cells. MIP-1alpha rather than MCP-1 may therefore play a significant role in the recruitment of monocytes/macrophages into the gingival tissues. However, both these chemokines have also been suggested to influence the accumulation of different T cell subsets. MCP-1 has been shown to be a potent activator of CD8 cytotoxic T cell activity [28], while a number of other studies in animal models have suggested that MCP-1 contributes more to type 2 than type 1 cytokine-mediated inflammation [29–31]. The results of the present study therefore, suggest a possible reduction of Th2 cells in the gingival tissues. Reports on MIP-1alpha have also demonstrated conflicting results. MIP-1alpha has been shown to induce the rapid accumulation of both CD4 and CD8 cells into lymph nodes in the mouse [32], to shift the immune response to a Th2-type response [28], to recruit Th1 cells [10] with a strong correlation between MIP-1alpha expression in the lymph node after HIV infection and a strong IFN-gamma response [33] while microchemotaxis experiments have shown MIP-1alpha to be a potent chemoattractant for B cells and cytotoxic T cells although at higher concentrations, the migration of CD4 cells was enhanced [34]. The increased MIP-1alpha+ cells in comparison with MCP-1, IP-10 and RANTES with increased proportions of CD8 cells suggests a role for these T cells in the elaboration of chemokines which affect the influx of defined populations into periodontitis lesions.
MIP-1alpha [35,36] and MCP-1 [37] have been shown to be involved in the recruitment of neutrophils. However, P. gingivalis has been shown to inhibit the production of MCP-1 by P. gingivalis-specific T cells, B cells and monocytes [7] to inhibit neutrophil migration into P. gingivalis-induced lesions in the mouse model [8,18] and to inhibit neutrophil transmigration through P. gingivalis infected oral epithelial cells [38]. Reduced MCP-1 in gingival tissues therefore may affect neutrophil migration particularly in regards to P. gingivalis infection, however, the increased presence of MIP-1alpha in periodontal disease tissues may still allow effective neutrophil recruitment.
The results of the present study also showed that a greater number of tissue samples demonstrated keratinocyte expression of MIP-1alpha than IP-10, RANTES or MCP-1 and while the expression of the latter 3 chemokines decreased with increasing inflammation, that of MIP-1alpha did not. This suggests a role for MIP-1alpha in the recruitment of leucocytes through the epithelium at early as well as later stages of inflammation. However, Tonetti et al. [39] demonstrated mRNA for MCP-1 by in situ hybridization in tissue biopsies taken from clinically healthy sites, sites undergoing a 21-day experimental plaque accumulation and from untreated and treated periodontitis sites. MCP-1 mRNA was detected in the oral epithelium particularly the basal layer, as well as in the inflammatory infiltrate in periodontal disease tissues. Again these results are at variance with the present study, although surface membrane staining of MCP-1 was not examined. However, it was noted that while MCP-1 expression related to mononuclear phagocytic infiltration in the connective tissues and oral epithelium, macrophages were consistently present in the junctional epithelium in the absence of MCP-1 message indicating the presence of other mediators responsible for the infiltration of these cells.
The role played by RANTES in the migration of T cell subsets and Th1/Th2 cells is controversial. However, IP-10 is specific for activated T cells, selective targetting Th1 cells resulting in the up-regulation of IFN-gamma rather than IL-4 peripheral blood producing T cells [40,41]. In the present study, the percentages of both IP-10 and RANTES in the gingival tissues remained low and stable regardless of infiltrate size or clinical status, suggesting no predominant T cell subset recruitment by these chemokines in gingivitis or periodontitis.
In conclusion, fewer leucocytes expressed the major monocyte/macrophage chemokine MCP-1 in gingival tissue sections. However, the percent MIP-1alpha+ cells was increased in comparison with MCP-1+ cells particularly in tissues with increased proportions of CD8 cells and B cells with increasing inflammation and also in tissues with higher numbers of macrophages with little inflammation. MIP-1alpha release by human adherent peripheral blood mononuclear cells stimulated with LPS has been attributed to the activity of IL-1 and TNF-alpha [42], two proinflammatory cytokines with significant connective tissue and bone destroying properties in periodontal disease [43]. Further studies are required to determine the significance of MIP-1alpha in periodontal disease.
Acknowledgements
This work was supported by the Government Employees Medical Research Fund and the Australian Dental Research Fund Incorporated.
