The expression of adhesion molecules and the local production of chemotactic cytokines within the epithelium are considered to be key events in neutrophil (PMN) migration at sites of mucosal infections. In their journey toward the gingival sulcus, PMNs have been shown to selectively migrate through the junctional epithelium. Little, however, is known about the molecular mechanisms involved in this key process aimed at the control of subgingival bacterial plaque. This investigation describes the expression of IL‐8 mRNA‐positive cells and the establishment of a gradient of intercellular adhesion molecule‐1 (ICAM‐1) receptors within the junctional epithelium of clinically healthy gingiva. Expression of ICAM‐1 and IL‐8 was topographically associated with the area of PMN migration; i.e., the junctional epithelium. Levels of ICAM‐1 expression increased from the basal cells toward the surface of the junctional epithelium and thus toward areas exposed to bacterial challenges. IL‐8 mRNA‐positive cells were also present at highest density in the most superficial junctional epithelial layers. The combination of the haptotactic stimuli, resulting from the interaction of the PMN's β2 integrin receptors with the gradient of ICAM‐1 expression, and the location of IL‐8 mRNA‐positive cells, consistent with the establishment of a discrete PMN chemotactic source, may play an important physiologic role in efficiently routing PMNs to the gingival sulcus. This process contributes to the maintenance of a local host‐parasite equilibrium and to the limitation of PMN‐associated tissue damage. J Periodontol 1998;69:1139–1147.

In bacterial infections, mononuclear and polymorphonuclear phagocytes are key components of host defenses. Recent investigations have indicated that chemokines are able to recruit and activate phagocytes. In particular, interleukin-8 (IL-8) attracts polymorphonuclear leukocytes (PMNs), while monocyte chemoattractant protein-i (MCP-1) is selective for cells of the monocyte/macrophage lineage. In this investigation, we analyzed the in situ expression of IL-8 and MCP-1 mRNAs in human periodontal infections. Specific mRNA was detected by in situ hybridization using 35S-labeled riboprobes in frozen tissue sections. Phagocytes (PMNs and macrophages) were specifically detected as elastase-positive or CD68+ cells by a three-stage immunoperoxidase technique. Results indicated that expression of phagocyte-specific cytokines was confined to selected tissue locations and, in general, paralleled phagocyte infiltration. In particular, IL-8 expression was maximal in the junctional epithelium adjacent to the infecting microorganisms; PMN infiltration was more prominent in the same area. MCP-1 was expressed in the chronic inflammatory infiltrate and along the basal layer of the oral epithelium. Cells of the monocyte/macrophage lineage were demonstrated to be present in the same areas. The observed expression pattern may be the most economic way to establish a cell-type-selective chemotactic gradient within the tissue that is able to effectively direct polymorphonuclear phagocyte migration toward the infecting microorganisms and modulate mononuclear phagocyte infiltration in the surrounding tissues. This process may optimize host defenses and contribute-to containing leukocyte infiltration to the infected and inflamed area, thus limiting tissue damage.

Tumor necrosis factor-␣ (TNF␣) exists in two biologically active forms, a 26-kDa transmembrane form and a proteolytically cleaved and secreted form. We sequentially inactivated all three known cleavage sites of mouse TNF␣ by mutating the corresponding DNA sequences. A murine T cell hybridoma transfected with the nonsecretable mutant TNF␣ efficiently lysed L929 target cells in a cell contact-dependent manner and induced expression of vascular cell adhesion molecule-1 on mouse endothelioma cells. A genomic mouse TNF␣ clone encoding this mutant was subsequently introduced as a transgene into TNF␣ ؊/؊ lymphotoxin-␣ ؊/؊ mice. The 3 AU-rich regulatory elements of the TNF locus were maintained in the transgene to assure adequate gene regulation. Transmembrane TNF␣ transgenic mice were fully protected from endotoxic shock, and no TNF␣ bioactivity was detectable in the serum after stimulation with lipopolysaccharide. Activated CD4 T cells from these animals, however, lysed L929 cells in a cell contact-dependent way. After administration of lipopolysaccharide, transmembrane TNF␣ transgenic mice produced significantly higher levels of interleukin-12 than wild-type mice or TNF-deficient mice. This indicates that transmembrane TNF␣ may greatly affect the course of a cellular immune responses in vivo and exerts quantitatively and qualitatively distinct functions from secreted TNF␣ in vitro and in vivo.Tumor necrosis factor-␣ (TNF␣) 1 is a pleiotropic cytokine produced by a wide variety of cell types of mostly hematopoietic, but also of nonhematopoietic, origin (for review, see Ref. 1). TNF␣ is instrumental in the immune elimination of various infectious agents such as Candida albicans (2), Listeria monocytogenes (3), or mycobacteria (4) and exerts potent proinflammatory effects, e.g. by inducing the expression of adhesion molecules such as VCAM-1, intercellular adhesion molecule 1 (ICAM-1), or E-selectin on endothelial cells and other cell types (5, 6). Aberrant production of TNF␣, however, has been also implicated in the pathogenesis of various diseases, such as rheumatoid arthritis, insulin-dependent diabetes-mellitus, sialoadenitis, and inflammatory bowel disease, in particular Crohn's disease (7-11).TNF␣ mediates its effects by binding to either TNF receptor 1 (TNFR1) or TNFR2. As revealed by mice deficient for either TNFR1 (12) or TNFR2 (13), these two receptors can mediate distinct effects (14). TNF␣ is synthesized as a 26-kDa precursor that is also expressed as a type II transmembrane molecule. The 26-kDa transmembrane molecule can be cleaved by membrane bound metalloprotease(s), including the TNF␣ converting enzyme (TACE) (ADAM-17) into 17-kDa secreted monomers that form biologically active homotrimers (15)(16)(17)(18)(19). By deleting the DNA sequence encoding the first 12 amino acids of the processed human 17-kDa TNF␣ monomer, Kriegler et al.(19) generated a nonsecretable 26-kDa transmembrane TNF␣ ⌬1-12 mutant (tm TNF␣). This human tm TNF␣ mutant is capable of lysing TNF␣-sensitive target cells (20). Using transfect...

The interpretation of studies aimed at understanding the pathophysiology of periodontal breakdown has been hampered by an insufficient understanding of the physiology of host responses in clinically healthy gingiva. This investigation was aimed at the evaluation of the in situ phenotype and topographic distribution of the inflammatory cells in clinically normal gingiva and peri-implant keratinized mucosa (PIKM). Soft tissue biopsies were obtained from clinically healthy gingiva or PIKM in 14 patients. Acetone fixed, cryostat sections were stained with a panel of monoclonal antibodies with a three stage avidin-biotin immunoperoxidase technique. Numbers of positive cells/mm2 were determined with a calibrated image analysis system. The major findings of the study were: (i) the presence of significantly higher densities of phenotypically characterized mononuclear cells in the ICT than in the JE in both gingiva and PIKM; (ii) the absence of a significant difference in PMN densities between JE and ICT in both gingiva and PIKM; (iii) the absence of statistically significant differences in densities of phenotypically characterized leukocytes associated with gingiva and PIKM; (iv) the presence of regional differences in the relative proportions of immunocompetent cells in both the gingiva and PIKM. It is concluded that inflammatory cells are selectively distributed in gingiva and PIKM. Unique functional compartments could be identified. The observed compartmentalization requires selective regulatory mechanisms.

The tubulin genes of T. brucei are clustered in a tightly packed array of alternating alpha- and beta-genes. The steady state mRNA contains one abundant mRNA species each for alpha- and beta-tubulin, both carrying the identical 35 nt mini-exon sequence at their 5'-ends. We have used in vitro run-on transcription assays to investigate the mechanism of tubulin gene transcription in T. brucei. Our results show that the regions between the individual tubulin genes are transcribed at the same rate as are the genes themselves. On the other hand, transcripts containing the intergenic regions could not be detected by Northern analysis or in vivo labelling experiments. We conclude that putative transcripts from the intergenic regions have a half-life of less than one minute. These results suggest that the tubulin gene cluster is transcribed as a single contiguous transcription unit yielding a primary transcript which is rapidly processed into individual mRNAs by the polyadenylation and mini-exon trans splicing machineries.