Tetracyclines have long been considered useful adjuncts in peridontal therapy based on their antimicrobial efficacy against putative periodontopathogens. However, recently these drugs were found to inhibit mammalian collagenases and several other matrix metalloproteinases (MMPs) by a mechanism independent of their antimicrobial activity. Evidence is presented that this property may be therapeutically useful in retarding pathologic connective tissue breakdown, including bone resorption. The experiments leading to this discovery are described and possible mechanisms are addressed, including the specificity of tetracyclines' anti-collagenase activity, the role of the drugs' metal ion (Zn2+, Ca2+)-binding capacity, and the site on the tetracycline molecule responsible for this nonantimicrobial property. Of extreme interest, the tetracycline molecule has been chemically modified in multiple ways, generating a new family of compounds called CMTs (chemically modified tetracyclines) that lack antimicrobial but still retain anti-collagenase activity. The first of these CMTs, 4-de-di-methylaminotetracycline, was found not to produce a major side-effect of antimicrobial tetracycline therapy--its administration to experimental animals did not result in the emergence of tetracycline-resistant microorganisms in the oral flora and gut. Numerous examples of the clinical potential of this non-antimicrobial property of tetracyclines in the treatment of periodontal and several medical diseases (e.g., sterile corneal ulcers, rheumatoid arthritis, skin bullous lesions, tumor-induced angiogenesis and metastasis) are discussed.

Mast cells have been implicated in the pathogenesis of the matrix degradation observed in the cartilaginous and osseous structures of the rheumatoid joint. We previously reported that human mast cell tryptase, a 134-kD granule-associated neutral protease, is present in rheumatoid synovium and can activate collagenase in crude culture medium in vitro. The present study attempts to depict the precise mechanism of this activation. To express full activation of latent collagenase, matrix metalloproteinase 3 (MMP-3) or stromelysin, can be activated by tryptase in a time and dose-dependent manner. Tryptase was not capable of generating active collagenase in the crude media from cultured rheumatoid synoviocytes depleted of proMMP-3 by immunoadsorption. In addition, the function of the tissue inhibitor of metalloproteinases (TIMP) was not altered by tryptase, and SDS-PAGE analysis revealed no degradation of TIMP by tryptase. The tryptase dependent activation of synoviocyte procollagenase thereby appears to be entirely dependent upon its ability to activate proMMP-3.

Doxycycline inhibits neutrophil (PMN)-type matrix metalloproteinases in human adult periodontitis gingiva. J Clin Periodontol 1995; 22: 100^109. © Mutiksgaard, i995.Abstract We previousiy reported tliat low-dose doxycycline (DOXY) therapy reduces host-derived coliagenase activity in gingival tissue of aduit periodontitis (AP) patients. However, it was not ciear wiiether this in vivo effect was direct or indirect. In the present study, iniiamed human gingival tissue, obtained from AP patients during periodontal surgery, was extracted and the extracts partia!!y purified by (NH4)2SO4 precipitation. The extracts were then ana!yzed for coUagenase activity using SDS-PAGE/iluofography/iaser densitometry, and for geiatinase activity using type I geiatin zymography as weii as a new quantitative assay using biotinyiated type I geiatin as substrate. DOXY was added to the incubation mixture at a fina! concentration of 0-!000 ^M. The concentration of DOXY required to inhibit 50% of the gingiva! tissue coiiagenase (IC50) was found to be i6-18 fiM in the presence or absence of i.2 mM APMA (an optimai organomercurial activator of latent procoiiagenases); this IC50 for DOXY was simiiar to that exhibited for coiiagenase or matrix metaiioproteinase (MMP)-8 from polymorphonuclear leukocytes (PMNs) and from gingival crevicuiar fluid (GCF) of AP patients. Of interest. Porphyromonas gingivalis collagenase was also inhibited by similar DOXY levels (IC5o=15 /^M), however the collagenase activity observed in the gingival tissue extracts was found to be of mammalian not bacterial origin based on the production of the specific a'^ (3/4) and a^ (i/4) coliagen degradation fragments. In contrast, the inhibition of coiiagenase purified from cuiture media of human gingival fibroblasts (MMP-1) required much greater DOXY levels (IC5o=280 //M). The predominant molecular forms of gelatinolytic activity present in the AP patients gingivai tissue extracts were found to closeiy correspond to the 92 kD PMN-type geiatinase (MMP-9) although small quantities of 72 kD fibrobiast-type geiatinase (MMP-2), and some other iow moiecuiar weight geiatinases, were also detected. The IC50 of DOXY versus gingivai tissue gelatinoiytic activity was estimated at 30-50 //M measured using either type I geiatin zymography or the biotinylated type I gelatin assay. We conclude that MMPs in inflamed gingival tissue of AP patients, iike those in GCE, originate primarily from infiltrating PMNs rather than resident gingival cells (fibroblasts and epitheha! ce!!s) or monocyte/macrophages, and that their patho!ogica!!y-elevated tissue-degrading activities can be directly inhibited by pharmaco!ogic !eve!s of doxycyc!ine.

No abstract

Recently, a new mechanism was proposed to explain the therapeutic efficacy of tetracyclines in periodontal disease— that specifically these antibiotics directly inhibit the activity of collagenase (and possibly other collagenolytic enzymes) produced by the host tissues resulting in a reduced rate of collagen breakdown (Golub et al. 1983). In the current studies, the effect of tetracyclines (minocycline, doxycycline, tetracycline) on collagenolytic enzyme activity was examined in human periodontal pockets and in several animal systems. In the clinical studies, gingival crevicular fluid (GCF) was collected on filter paper strips from individual periodontal pockets, the volume measured with the Periotron 6000, and the Gingival Index and pocket depth scored immediately thereafter at the same sites. The strips were incubated with [3H‐methyi] collagen maintained in solution (or with 14C‐glycine labeled collagen Fibrils), and the radiolabeled collagen degradation products measured in a liquid scintillation spectrometer, Using the techniques of SDS‐PAGE and fluorography, the GCF was observed to generate collagen degradation products characteristic of those produced by mammalian collagenase, a finding consistent with previous studies. Both minocycline and tetracycline therapy reduced GCF collagenolytic activity about 70% during the initial weeks of treatment; the clinical parameters also showed improvements at the same time periods. This reduction in collagenolytic activity had largely regressed by 5 weeks after terminating tetracycline therapy and by 19 weeks after minocycline therapy, in the absence of other forms of periodontal treatment. In a preliminary study, 50 days of doxycycline treatment also reduced GCF collagenase activity, an effect which persisted for at least 2 months (but less than 1 year) after terminating its administration. In several studies involving experimental animals, the tetracyclines (1) in vivo, were found to reduce the pathologically excessive collagenase activity in gingiva and skin of diabetic rats; and (2) in vitro, directly inhibited the activity of collagenases from rat PMNL's and rabbit chondrocytes. The data in the current and our previous studies (Golub et al. 1983, Gomes, Golub & Ramamurthy 1984) continue to support the hypothesis that tetracyclines, by directly inhibiting collagenolytic enzyme(s) activity, may be therapeutically useful in treating diseases, such as periodontitis, characterized by excessive collagen breakdown.

No abstract

Both Type I and Type II diabetes mellitus (DM) have been associated with unusually aggressive periodontitis. Accordingly, rat models of both types of DM were used to study (i) mechanisms mediating this systemic/local interaction and (ii) new pharmacologic approaches involving a series of chemically modified tetracyclines (CMTs) that have lost their antimicrobial but retained their host-modulating (e.g., MMP-inhibitory) properties. In vitro experiments on tissues from Type I DM rats demonstrated that several of these CMTs were better matrix metalloproteinase (MMP) inhibitors than was antibacterial doxycycline (doxy), except for CMT-5, which, unlike the other MMP inhibitors, was found not to react with zinc. Data from in vivo studies on the same rat model generally supported the relative efficacy of these compounds: the CMTs and doxy were found to inhibit MMP activity, enzyme expression, and alveolar bone loss. To examine other long-term complications such as nephropathy and retinopathy, a Type II (ZDF) model of DM was studied. Treatment of these DM rats with CMT-8 produced a 37% (p < 0.05), 93% (p < 0.001), and 50% (p < 0.01) reduction in the incidence of cataract development, proteinuria, and tooth loss, respectively; whereas the doxy-treated ZDF rats showed little or no effect on these parameters. CMT treatment decreased mortality of the Type II ZDF diabetic animals, clearly indicating that CMTs, but not commercially available antibiotic tetracyclines (TCs), may have therapeutic applications for the long-term management of diabetes.

Background-Acute lung injury (ALI) after cardiopulmonary bypass (CPB) results from sequential priming and activation of neutrophils. Activated neutrophils release neutral serine, elastase, and matrix metalloproteinases (MMPs) and oxygen radical species, which damage alveolar-capillary basement membranes and the extracellular matrix, resulting in an ALI clinically defined as adult respiratory distress syndrome (ARDS). We hypothesized that treatment with a potent MMP and elastase inhibitor, a chemically modified tetracycline (CMT-3), would prevent ALI in our sequential insult model of ALI after CPB. Methods and Results-Anesthetized Yorkshire pigs were randomized to 1 of 5 groups: control (nϭ3); CPB (nϭ5), femoral-femoral hypothermic bypass for 1 hour; LPS (nϭ7), sham bypass followed by infusion of low-dose Escherichia coli lipopolysaccharide (LPS; 1 g/kg); CPBϩLPS (nϭ6), both insults; and CPBϩLPSϩCMT-3 (nϭ5), both insults plus intravenous CMT-3 dosed to obtain a 25-mol/L blood concentration. CPBϩLPS caused severe lung injury, as demonstrated by a significant fall in PaO 2 and an increase in intrapulmonary shunt compared with all groups (PϽ0.05). These changes were associated with significant pulmonary infiltration of neutrophils and an increase in elastase and MMP-9 activity. Conclusions-All pathological changes typical of ALI after CPB were prevented by CMT-3. Prevention of lung dysfunction followed an attenuation of both elastase and MMP-2 activity. This study suggests that strategies to combat ARDS should target terminal neutrophil effectors. (Circulation. 1999;100:400-406.)