MedicineLakex

 

Subantimicrobial-dose doxycycline modulates gingival crevicular fluid biomarkers of periodontitis in postmenopausal osteopenic women

J Periodontol • August 2008 Subantimicrobial-Dose DoxycyclineModulates Gingival Crevicular FluidBiomarkers of Periodontitisin Postmenopausal Osteopenic WomenLorne M. Golub,* Hsi Ming Lee,* Julie A. Stoner,† Timo Sorsa,‡ Richard A. Reinhardt,§Mark S. Wolff,*i Maria E. Ryan,* Pirkka V. Nummikoski,¶ and Jeffrey B. Payne§ Background: We recently demonstrated that a 2-year subantimicrobial- dose doxycycline (SDD) regimen (double-masked, placebo-controlled clinicaltrial) in postmenopausal (PM) women exhibiting mild systemic bone loss(osteopenia) and local bone loss (periodontitis) reduced the progression of peri-odontal attachment loss (intent-to-treat analysis) and the severity of gingivalinflammation and alveolar bone loss (subgroups) without producing antibioticside effects. We now describe SDD effects on biomarkers of collagen degrada-tion and bone resorption in the gingival crevicular fluid (GCF) of the same vul- More than 2 decades ago, Golub et al.1 nerable subjects.
Methods: GCF was collected from SDD- and placebo-treated PM subjects other groups2-4 confirmed, (n = 64 each) at the baseline and 1- and 2-year appointments; the volume was determined; and the samples were analyzed for collagenase activity (us- such as doxycycline and ing a synthetic peptide as substrate), relative levels of three genetically dis- tinct collagenases (Western blot), a type-1 collagen breakdown product/ host-derived matrix metallo- bone resorption marker (a carboxyterminal telopeptide cross-link fragment proteinases (MMPs), such as of type I collagen [ICTP]; radioimmunoassay), and interleukin-1b (enzyme- collagenases and gelatin- linked immunosorbent assay). Statistical analyses were performed using ases, and by a mechanism generalized estimating equations; primary analyses were intent-to-treat.
unrelated to the antibiotic Results: Collagenase activity was significantly reduced by SDD treatment activity of these drugs. The relative to placebo based on intent-to-treat (P = 0.01). ICTP showed a similar first mechanism identified pattern of change during SDD treatment, and GCF collagenase activity and was the ability of TCs to ICTP were positively correlated at all time periods (P <0.001). Matrix metallo- directly inhibit already acti- proteinase (MMP)-8 accounted for ;80% of total collagenase in GCF, with vated MMPs by binding the much less MMP-1 and -13, and SDD reduced the odds of elevated MMP-8 metal ions, calcium and by 60% compared to placebo (P = 0.006).
zinc, in the catalytic domain Conclusion: These observations support the therapeutic potential of long- of the enzymes.1-3,5 Addi- term SDD therapy to reduce periodontal collagen breakdown and alveolar tional pleiotropic mecha- bone resorption in PM women; effects on serum biomarkers of systemic bone loss in these subjects are being analyzed. J Periodontol 2008;79:1409-1418.
apparent, such as the abilityof these drugs to downregu- late the expression of inac- Clinical trial; collagenases gingival crevicular fluid; osteopenia; and to block the activationof these zymogens. Twostrategies were pursued to * Department of Oral Biology and Pathology, School of Dental Medicine, Stony Brook University, Stony Brook, † Currently, Department of Biostatistics and Epidemiology, College of Public Health, University of Oklahoma Health Sciences Center, Oklahoma City, OK; previously, Department of Biostatistics, College of PublicHealth, University of Nebraska Medical Center, Omaha, NE.
property of TCs into new ‡ Department of Oral and Maxillofacial Diseases, Helsinki University Central Hospital, Helsinki, Finland.
therapies to inhibit patholog- § Department of Surgical Specialties, College of Dentistry, University of Nebraska Medical Center, Lincoln, NE.
ically excessive connective i Department of Cariology and Comprehensive Care, New York University College of Dentistry, New York, NY.
¶ Longitudinal Radiographic Assessment Facility, University of Texas Health Science Center at San Antonio, San Antonio, TX.
SDD and GCF Biomarkers in Postmenopausal Women Volume 79 • Number 8 tissue destruction, including bone resorption. One 3) long-term SDD therapy can produce these effects strategy was to chemically modify the TC molecule in PM women exhibiting bone loss locally and system- to eliminate its antibiotic properties (i.e., bacteri- ically, whereas previous studies2,3,19,20 on this topic of ostatic) but to retain (even enhance) its MMP-inhibitory host-modulation therapy did not target subjects with properties, i.e., chemically-modified tetracyclines this important systemic factor, estrogen deficiency as- (CMTs) 1 through 10.2,3 The second strategy was to sociated with the menopause.
titrate downward the oral dose of doxycycline toproduce blood levels of the drug too low to produceantibiotic activity (and, thus, eliminate side effects of MATERIALS AND METHODS antibiotic administration) but which still produced The details of this clinical trial, as well as the methods MMP-inhibitory effects and clinical improvement in used for clinical, radiographic, and microbiologic mea- patients with periodontitis.2-4 Early in these studies, surements, were described in our earlier articles.16-18 TCs and CMTs were found to inhibit bone resorption In brief, the study was a two-center, double-masked, in organ culture6,7 and in animal models of bone- placebo-controlled clinical trial with each of 128 PM deficiency diseases, including the estrogen-deficient subjects randomly assigned to take placebo (n = 64 (ovariectomized) osteoporotic aged female rat8 and the subjects) or SDD (doxycycline hyclate, 20 mg; n = diabetes-induced osteopenic rat.9 These effects were 64) tablets twice daily for 2 years. Subjects were associated, in part, with the MMP-inhibitory properties recruited and randomized between June 2002 and of these drugs. These drugs also ‘‘normalized'' patho- October 2003. The last subject completed the clinical logic bone turnover by inhibiting osteoclast activity and trial in October 2005. All subjects received calcium bone resorption and by enhancing osteoblast activity, (600 mg) and vitamin D (200 IU) supplements twice type I collagen synthesis, and bone formation.10-12 daily with instructions for use and were scheduled to Because estrogen deficiency in postmenopausal receive periodontal maintenance therapy every 3 to (PM) women is the most common cause of osteo- 4 months, all of which was provided at no cost to the porosis, involves accelerated bone resorption over- participants during the 2-year protocol. Enrolled sub- powering the rate of bone formation, and has been jects were 45 to 70 years of age, PM for ‡6 months, di- associated with increased tooth loss and oral bone agnosed as osteopenic (not osteoporotic, because this loss,13-15 we hypothesized that subantimicrobial-dose disease would have required treatment with a United doxycycline (SDD), by a non-antimicrobial mecha- States Food and Drug Administration–approved med- nism, can reduce bone loss and improve clinical mea- ication, e.g., a bisphosphonate) based on dual-energy sures of periodontitis in these vulnerable subjects.
x-ray absorptiometry (DEXA; i.e., T scores of -1.0 As a result, we recently completed a double-masked, to -2.5 inclusive) of the lumbar spine or femoral placebo-controlled clinical trial on PM women who neck, had moderate to advanced periodontitis, and exhibited mild systemic bone loss (osteopenia) and were undergoing periodontal maintenance therapy.
periodontitis and who were administered a 2-year reg- Additional enrollment criteria were described by us imen of SDD or placebo adjunctive to periodontal previously.16 However, once enrolled, subjects were maintenance therapy and calcium and vitamin D not removed from the trial if they did not adhere to supplements. Our data demonstrated that SDD sig- the protocol (e.g., started bisphosphonate therapy nificantly reduced the progression of periodontal at- or chronic non-steroidal anti-inflammatory drug ther- tachment loss (intent-to-treat analysis) and reduced apy) based on an intent-to-treat paradigm. These the severity of gingival inflammation and alveolar occurrences of non-adherence to the protocol were bone loss (in subgroups of these subjects), without recorded and addressed during data analysis (see producing side effects associated with antibiotic ther- Statistical Analysis). All subjects provided written in- apy.16-18 We now present our findings, in the same formed consent to participate in the study. The study clinical trial, describing the effect of SDD on bio- protocol was reviewed and approved by the Stony chemical ‘‘markers'' of collagen degradation and bone Brook Institutional Review Board and the University resorption in the gingival crevicular fluid (GCF) from of Nebraska Medical Center Institutional Review this vulnerable population. To the best of our knowl- edge, this study is the first to show that 1) SDD can re- Computer-assisted densitometric image analysis duce collagenase levels and activity over a prolonged of oral posterior bite-wing radiographs and DEXA period of time (previous studies2-4 described effects scans of the lumbar spine and femoral neck to assess on collagenases over several weeks to 3 months, local and systemic bone loss, respectively, as well as which did not preclude the subsequent potential loss clinical measurements of periodontal disease and of drug effect), 2) effects on collagenase are positively subgingival plaque samples for microbiologic analy- correlated with a biomarker of bone resorption in the sis, were taken at regular intervals over 2 years; these same GCF samples over a long period of time, and data were described previously.16-18 J Periodontol • August 2008 Golub, Lee, Stoner, et al.
Collection of GCF Samples At each of three appointments (baseline and 1 and 2 Gln-dArg)‡‡ that served as the enzyme substrate. Fol- years), GCF samples were collected from two pocket lowing incubation at 37C, the reaction mixture was sites (5 to 9 mm in depth) per subject identified at a quenched with l,10-phenanthroline (a zinc chelator previous screening appointment. The GCF collection that binds this cation in the collagenase molecule), technique and measurement of GCF volume were de- the tripeptide breakdown product was separated by scribed by us previously.19,20 In brief, the identified high-performance liquid chromatography§§ using a pocket sites were isolated with cotton rolls and gently reverse-phase C18 column (4.6 · 75 mm, 3.5-mm air dried. Supragingival plaque was carefully removed macroporous spherical support), and the eluate using periodontal curets, then precut presterilized fil- was monitored at 375 nm for quantifying the DNP- ter paper strips# were inserted into each isolated peri- labeled peptides. The collagenase activity mea- odontal pocket until slight resistance was felt. The sured by this assay was further characterized as a filter strips were left in place for 10 seconds, and the host-derived collagenase based on its response in volume absorbed onto the paper strip was im- vitro to several different proteinase inhibitors and mediately determined in a calibrated GCF flow me- activators19,20 and was scored on a scale of 0% to ter.** GCF samples visually contaminated with 100% hydrolysis of the synthetic octapeptide.
blood were discarded. Immediately after measure- ICTP and IL-1b Analyses ment, the GCF samples were placed into a microfuge As described previously, l00-ml aliquots were taken tube on ice at chairside and stored frozen at -80C and analyzed by radioimmunoassay for ICTP20 using within 10 minutes of collection. GCF collection pre- a commercial kit,ii and duplicate 50-ml aliquots were ceded any clinical measurements.
analyzed for IL-1b using an enzyme immunoassay¶¶ Assay Methods for GCF Biomarkers based on a double-antibody sandwich technique.21 The frozen GCF samples (one pooled sample per sub- Western Blot Analysis of MMP-1, -8, and -13 ject/appointment) were thawed (4C) for 15 minutes.
In brief, lyophilized GCF extracts (100 ml containing Then, 400 ml 50 mM Tris/0.2 M NaCl/5 mM CaC12 10 to 20 mg protein) were treated with Laemmli buffer buffer (pH 7.6) containing a proteinase-inhibitor (pH 7.0) containing 5 mM dithiothreitol and heated for cocktail (which blocked serine, cysteine, and thiol 5 minutes at 100C. High- and low-range prestained proteinases, but not MMPs), consisting of antipain sodium dodecyl sulfate (SDS)-polyacrylamide gel (1 mg/l), aprotinin (1 mg/l), N-ethylmaleimide (125 electrophoresis standard proteins were used as mo- mg/l), leupeptin (1 mg/l), and 50 mg/l detergent,†† lecular weight markers. The samples were electro- were added to the pooled GCF samples. The two strips phoresed on 7.5% SDS-polyacrylamide gels and (pooled) containing the GCF were exhaustively mixed then electrophoretically transferred to nitrocellulose and extracted (1 hour, 4C), and aliquots were taken membranes, and Western blot analysis was carried for analysis of the following: collagenase (MMP) activ- out as described by us previously.20,22 ity, the only type of proteinase that can degrade the Specific immunoreactivity was visualized as dark triple-helical collagen molecule under physiologic bands against a clear background, and the mem- conditions;3,4 a carboxyterminal telopeptide cross- branes were scanned with an imaging densitometer## link fragment of type I collagen [ICTP], a degradation using a program*** that corrects for background fragment of type I collagen and a bone resorption values. The densitometric units were measured in marker;20 relative protein levels of the three different the linear range of immunoreactivity for each of the collagenases (MMP-1, -8, and -13) in GCF;20 and in- three MMPs; purified human MMP-1, -8, and -13 were terleukin (IL)-1b, a proinflammatory cytokine that used as positive controls.
can induce osteoclastic activity and bone resorp-tion.21 If one of the two teeth selected for GCF sam- Statistical Analysis pling was extracted before the 2-year protocol Statistical analytical procedures were described by us ended, the GCF collected on a filter strip from the re- in detail.16-18 The method of generalized estimating maining tooth was eluted in 200 ml instead of 400 ml equations, with a working exchangeable correlation buffer, and the aliquots for each of the assays below structure, was used.23 For the collagenase, ICTP, and were reduced by half. These assays were carried outas follows.
Periopaper, Proflow, Amityville, NY.
** Periotron 6000, Proflow.
Total Collagenase Activity †† Zwittergent, Calbiochem-Novabiochem, La Jolla, CA.
The details for measuring GCF collagenase activity ‡‡ Bachem, King of Prussia, PA.
§§ Waters Alliance 2695 System, Waters Alliance, Milford, MA.
were described by us previously.19 Seventy microliters Immunodiagnostic Systems, Fountain Hills, AZ.
of GCF extract were transferred to a microfuge tube ¶¶ Biosource, Camarillo, CA.
## Bio-Rad Model GS-700, Bio-Rad, Hercules, CA.
containing a synthetic, collagenase-susceptible octa- *** Analyst, Bio-Rad, Hercules, CA.
SDD and GCF Biomarkers in Postmenopausal Women Volume 79 • Number 8 IL-1b measures, a linear regression model was fit, forwhich the outcome was the natural log-transformedfollow-up measure and the baseline biochemical value,a time effect (12- or 24-month), and a study drug ef-fect; randomization stratification factors (study centerand baseline smoking status) were independent varia-bles. The models were adjusted for a batch effect(assays were run in three different batches, and allsamples were analyzed in the same batch for a givensubject; batches were well-balanced by treatmentgroup), along with all two- and three-way interactionterms among treatment, batch, and time. Non-significantinteraction terms among time, batch, and treatmentwere dropped, and the model was refit. Among all sub- jects in the intent-to-treat analyses, interactions in- GCF collagenase activity in PM women with chronic periodontitis: effect volving treatment and the time or batch terms were of placebo and SDD administration. The median % lysis of a not significant; therefore, the treatment effects are col agenase-susceptible octapeptide per pool of two GCF samples persubject at baseline (B), 1 year (1-YR), and 2 years (2-YR) is summarized and reported across time periods and represented by the bar height; whiskers are drawn between the 25th batches. The influence of extreme data points, defined and 75th percentiles. The estimated effect on median col agenase as falling more than three standard deviations away activity levels was a 22% reduction (95% CI: 37% lower to 5% lower; from the mean, was investigated by refitting regres- P = 0.01), comparing combined 1- and 2-year values between SDD sion models without such points. Because MMP distri- and placebo after adjustment for baseline levels.
butions were highly skewed, the measures were codedinto two or three categories based on the median valueor tertiles and were analyzed using a similar modeling two pockets per subject. Using techniques we pub- approach as described above with a binomial (logistic lished previously,19,20 the SDD-treated PM women link) or multinomial (cumulative logit link) regression showed ;50% reduction in GCF collagenase activity model, respectively. A Pearson correlation coefficient over the 2-years compared to their own baseline was calculated to summarize the association between values. In contrast, the placebo values appeared to collagenase and ICTP measures.
decrease only slightly. Moreover, based on linear re- The primary analysis was intent-to-treat; data were gression analysis, the SDD-treated group showed a analyzed from all randomized subjects, regardless of statistically significant 22% reduction in median protocol adherence. As a secondary analysis, only GCF collagenase activity compared to placebo-trea- measurements from subjects up to the time when a ted subjects over the study period, based on intent-to- lack of protocol adherence occurred (e.g., initiation treat analysis (95% confidence interval [CI]: 37% lower of significant concomitant medications, such as bis- to 5% lower; P = 0.01), and a 29% reduction in median phosphonates, or subject adherence to study medica- GCF collagenase activity compared to placebo sub- tions or calcium/vitamin D below an 80% threshold) jects based on the per-protocol analysis (95% CI: were analyzed (per-protocol analysis). Reasons for 48% lower to 4% lower; P = 0.02) after adjusting for exclusion from per-protocol analysis were described baseline values. When the GCF collagenase data were previously;16,17 overall, the SDD group had a slightly expressed as enzyme activity per microliter GCF, the larger per-protocol subset (n = 32) than the placebo greater reduction over time for SDD compared to pla- group (n = 27). Placebo and SDD groups exhibited cebo was not statistically significant based on intent- similar characteristics, including age, ethnicity, race, to-treat (P = 0.2), but it was significant based on the years following estimated onset of menopause, smok- per-protocol analysis (P = 0.05; data not shown).
ing, number of teeth, and probing depths at base- For subgroup analyses, the effect of SDD seemed to line.16 The effect of SDD compared to placebo also depend on smoking status (P = 0.05), and there was was investigated for subgroups defined by smoking a significant interaction between time and treatment status, time since onset of menopause, adherence for non-smokers (P = 0.02). At 1 year, median levels to study medications, and significant concomitant of collagenase activity per pool of GCF were 40% medication use, using tests of interactions in the re- lower for SDD subjects compared to placebo subjects gression models as described in detail previously.16,17 in the non-smoking group, which was statistically sig-nificant (95% CI: 53% lower to 22% lower; P <0.0001).
However, the 17% reduction for SDD-treated subjects The data in Figure 1 show the effect of SDD therapy on compared to placebo in the non-smokers at 2 years collagenase activity expressed per pool of GCF from was not statistically significant (P = 0.2). The smoking J Periodontol • August 2008 Golub, Lee, Stoner, et al.
cations, or use of concomitant medications based onregression modeling.
Because the initiation of the degradation of the na- tive triple-helical collagen molecule is mediated bycollagenases under physiologic conditions2-4 and col-lagen degradation is a key event in bone resorption, thecorrelation between the values for collagenase activityand ICTP in the GCF of these subjects, at all time pe-riods, was determined and summarized across placeboand SDD groups (Fig. 3). The data for GCF collage-nase activity and GCF ICTP levels were converted toa log value that demonstrated that the collagenase ac-tivity and the ICTP in the GCF were linearly related with positive correlation coefficients (r) of 0.62, 0.52, and GCF ICTP levels in PM women with chronic periodontitis: effect of 0.50 for baseline, 1 year, and 2 years, respectively; placebo and SDD administration. The median ICTP per subject all three r values were highly statistically significant (expressed as picograms ICTP per pool of two GCF samples) atbaseline (B), 1 year (1-YR), and 2 years (2-YR) is represented by the (P <0.001). In general, the higher the values for colla- bar height; whiskers are drawn between the 25th and 75th percentiles.
genase activity per pool of GCF, the greater was the The estimated effect on median ICTP levels was a 16% reduction level of bone collagen breakdown products (ICTP).
(95% CI: 31% lower to 2% higher; P = 0.08, comparing combined In addition to measuring total collagenase activity 1- and 2-year values between SDD and placebo after adjustment for in the GCF (Figs. 1 and 3) of these PM women, the rel- baseline levels.
ative protein levels of the three genetically distinct col-lagenases, previously identified in human GCF,20,22 group did not show significant reductions with SDD were also assessed (Table 1). Using the Western blot compared to placebo (P = 0.3), and no other subgroup technique, MMP-1 (collagenase-1), -8 (collagenase- effects were significant.
2), and -13 (collagenase-3) were detected in the A similar pattern of change over time was seen for GCF samples. Then, after densitometrically scanning ICTP expressed per pool of GCF (Fig. 2) collected the electrophoretic gels for the different molecular from the placebo- and SDD-treated PM subjects. As forms of each type of collagenase, the data were ex- described by us previously,20 ICTP is a breakdown pressed as a percentage of the total collagenase pro- product of type I collagen, and this collagen makes tein. Regardless of whether the data were expressed up >90% of the organic matrix of bone. Thus, as a mean or median value, MMP-8 (which included ICTP measurements in GCF, blood, and urine have 65- to 75-kDa leukocyte and 45- to 55-kDa mesen- been considered a diagnostic biomarker of bone re- chymal isoforms of this enzyme22) was the predomi- sorption24 and are believed to reflect (at least in part) nant collagenase type in the GCF, accounting for collagenase-mediated breakdown of the triple-helical ;80% of the total. This was followed by MMP-13 at collagen molecule. Once again, placebo treatment 0% to 18% (expressed as 25th to 75th percentiles) had no effect. In contrast, SDD therapy over the study and MMP-1, which was detected at only very low period seemed to reduce the median ICTP levels per levels (0% to 9%). Focusing on changes in the domi- pool of GCF by ;30% compared to this group's own nant type of collagenase, MMP-8, in the GCF of these baseline values. Using linear regression analysis, PM women (Fig. 4), and based on intent-to-treat anal- the SDD-treated group showed a 16% reduction in ysis, SDD therapy reduced the odds of elevated MMP-8 median GCF ICTP levels compared to placebo-treated values (across the ordered categories of 0 to 1.00, subjects, after adjusting for baseline values (P = 0.08).
1.001 to 2.5, and >2.5 units) by 60% compared to pla- However, when three extreme baseline values were cebo during the 2-year study period. This treatment excluded (two values in the placebo group and one effect was highly statistically significant (odds ratio in the SDD group), the SDD effect was statistically sig- [OR] = 0.40; 95% CI: 0.21 to 0.77; P = 0.006). Consis- nificant, with a median follow-up measure for SDD tent with this pattern, SDD therapy increased the odds subjects that was 19% lower than for placebo subjects of lower values (among the ordered categories of 0 (95% CI: 33% lower to 2% lower; P = 0.03). Among the to 1.00, 1.00l to 2.5, and >2.5 units) for this type of per-protocol subset, SDD was associated with a 16% collagenase, compared to placebo therapy, over the reduction in median ICTP levels compared to placebo, study period. Based on per-protocol analysis, this which was not significant (P = 0.2). With regard to sub- effect was even more dramatic because the odds of group analyses, no significant effects were seen higher values for MMP-8 in SDD-treated subjects were between SDD- and placebo-treated subjects for 78% lower than in those receiving placebo tablets smoking status, years PM, adherence to study medi- (OR = 0.22; 95% CI: 0.07 to 0.66; P = 0.007).

SDD and GCF Biomarkers in Postmenopausal Women Volume 79 • Number 8 not use concomitant medications (P =0.0002), whereas this effect was not sig-nificant in subjects who used concomi-tant medications (P = 0.7). Reductionsin GCF collagenase activity and MMP-8 immunoreactive levels due to SDDtherapy were not complete (i.e., resid-ual collagenase activity and MMP-8 pro-tein levels could be detected at 1 and 2years; the therapeutic advantage of thisless-than-complete reduction is ad-dressed in the Discussion). No other sig-nificant associations between treatmentand MMP levels were observed.
Regarding the levels of IL-1b in the GCF samples from placebo- and SDD-treated subjects over the 2-year timeperiod, the pattern of change was simi-lar to that seen for collagenase andICTP, although the reduction in IL- 1b levels was not significant except for Correlation between natural log–transformed col agenase activity (percentage lysis of a subgroup of subjects (see below). In collagenase-susceptible substrate) and natural log–transformed ICTP in GCF of PM women general, based on intent-to-treat and with chronic periodontitis over the 2-year clinical protocol for SDD and placebo subjects on per-protocol analyses, the SDD sub- combined. Linear regression lines are drawn for each time point. 1-YR = 1 year; 2-YR = jects exhibited ;20% and 33% lower median values, respectively, for IL-1b over the study time period compared to placebo subjects, after adjusting for different base-line values, but these data were not statistically signif- Distribution of MMP-1 (collagenase-1), icant (P >0.2 for each; Fig. 5). Regarding subgroup MMP-8 (collagenase-2), and MMP-13 analyses, the data for IL-1b were not statistically sig- nificant for subjects within 5 years of menopause (P =0.1). However, for those subjects beyond 5 years of menopause, subjects administered SDD showed a statistically significant 51% lower median value forIL-1b per pool of GCF than placebo-treated subjects (OR = 0.49; 95% CI: 76% lower to 1% lower; P = 87.5 (39.2 to 98.5) 0.05). When the data were expressed per microliterof GCF, a similar pattern of change was seen, and re- sults from the intent-to-treat and per-protocol analyses Data reported as the percentage of total collagenase protein in GCF.
were similar, but these data were not statistically signif-icant (data not shown).
Moreover, 1) the reduction of MMP-8 in SDD-treated subjects likely was driven by differences in the higher The rationale for the current interventional (i.e., long- term administration of SDD) human clinical trial on (PMN)-type MMP-8 (65 to 75 kDa), with a 38% reduc- PM women exhibiting local (periodontitis) and mild tion in the odds of higher PMN values for SDD subjects systemic (osteopenia) bone loss was two-fold. First, compared to placebo subjects (P = 0.1), and a much organ and cell culture, in vivo animal, and human smaller effect on the mesenchymal-type MMP-8 (45 studies1-3,8-11,16,20,25-29 over the past 25 years dem- to 55 kDa) where the odds of non-zero values were onstrated beneficial, non-antibiotic effects of TC com- 12% lower for SDD subjects compared to placebo pounds (e.g., SDD and CMTs) on pathologic local and subjects (P = 0.8), and 2) subgroup analysis demon- systemic bone loss. Second, it is increasingly being strated that the dramatic reduction in MMP-8 levels recognized that patients and experimental animals due to SDD therapy reflected an 83% lower odds of with systemic bone-deficiency disease, particularly high values for this collagenase in subjects who did PM osteoporosis (but also other disorders such as

J Periodontol • August 2008 Golub, Lee, Stoner, et al.
mone, prostaglandin E2, or endotoxin.6,7Non-TC antibiotics were ineffective inthis system. Mechanisms included theability of TCs to inhibit MMPs expressedby osteoblasts and osteoclasts (for re-views, see Golub et al.2,3). More recently,TCs were found to enhance bone forma-tion and inhibit bone resorption.10,25-28As examples, using ultracytochemistry, autoradiography, and dynamic histo- The effect of SDD (versus placebo) administration on the risk of low, medium, or high levels morphometry on the osteoporotic bones of MMP-8 (leukocyte-type collagenase) in the GCF of PM women with chronic periodontitis of diabetic and ovariectomized (surgi- over the 2-year clinical protocol: 0 to 1.00, 1.001 to 2.5, and >2.5 represent low, medium, cally induced menopausal) rats, TCs and high levels of MMP-8, respectively. Data were available for 64, 59, and 57 placebo were found to enhance osteoblast activ- subjects and for 63, 55, and 51 SDD subjects at the baseline, 1-year (1-YR), and 2-year(2-YR) visits, respectively. The odds of higher MMP-8 values were reduced by 60% in SDD ity, type I collagen synthesis, and bone subjects compared to placebo subjects after adjustment for baseline levels (OR = 0.40; formation. In the estrogen-deficient oste- 95% CI: 0.21 to 0.77; P = 0.006).
oporotic rat, the increased production ofnew bone as a result of TC treatment in-creased the connectivity of the resorbed discontinuous trabeculae in long bones,28 and, in anarthritic rat model, these drugs increased bone biome-chanical strength and resistance to experimental frac-ture.29 Of interest, both alveolar bone loss andsystemic (skeletal) bone loss benefited from thesetherapeutic effects of TCs in an animal model of PMosteoporosis.8 Two earlier clinical studies also provided a rationale for the current 2-year clinical trial. In a small pilotstudy, Payne et al.31 found that PM women diagnosedwith periodontitis and systemic bone loss (osteopeniaor osteoporosis) and treated with a 1-year cyclicalSDD regimen showed less alveolar bone height loss and alveolar bone density loss (based on computer- GCF IL-1b levels in PM women with chronic periodontitis: effect ofplacebo and SDD administration. The median GCF IL-1b per subject assisted densitometric image analysis) than pla- (expressed as picogram GCF IL-1b per pool of two GCF samples) at cebo-treated subjects. In an earlier study, Golub baseline (B), 1 year (1-YR), and 2 years (2-YR) is represented by the et al.20 monitored biomarkers of collagen and bone bar height; whiskers are drawn between the 25th and 75th percentiles.
destruction, as well as bone formation and turnover, The estimated effect on median IL-1b levels was a 20% reduction in GCF of male and female subjects with chronic peri- (95% CI: 54% lower to 39% higher; P = 0.4), comparing combined1- and 2-year values between SDD and placebo after adjustment for odontitis (who were not diagnosed with osteopenia or baseline levels.
osteoporosis); a 2-month regimen of SDD (adjunctiveto scaling) significantly reduced collagenase activityand ICTP, with no change in osteocalcin. Osteocalcinwas originally viewed as a biomarker of bone forma- diabetes-induced osteopenia), can exhibit acceler- tion because this Gla protein (rich in gamma-carboxy ated local (alveolar) bone loss beyond that induced glutamic acid) is expressed only by osteoblasts, and, by subgingival periodontopathogens, and all of these once secreted, small quantities are released into the diseases might also benefit from treatment with TC bloodstream where they can be measured in serum samples. However, more recently, osteocalcin has Regarding these rationales, soon after the MMP- been considered a biomarker of bone turnover (not inhibitory activity of TCs was discovered, the rele- just bone formation) because, after secretion by oste- vance of this non-antimicrobial property of TCs to bone oblasts, this highly anionic matrix protein binds to resorption was explored. Using standard aseptic organ Ca++ in mineralized bone; during bone resorption, culture systems, traditional and chemically modified the calcium-bound osteocalcin is released into the cir- TCs were found to inhibit bone resorption, regardless culation.32 Because the earlier study20 on subjects of whether the loss of the mineral and organic matrix with chronic periodontitis found that SDD decreased constituents of bone was induced by parathyroid hor- the biomarker of bone resorption (ICTP) but did not SDD and GCF Biomarkers in Postmenopausal Women Volume 79 • Number 8 affect the levels of the bone turnover biomarker (oste- for these same interstitial collagenases in an earlier ocalcin), and the rate of bone turnover reflects a com- study20 on GCF from adult females and males with bination of bone formation plus bone resorption, these chronic periodontitis). The highly significant reduction findings suggested that this TC treatment suppressed in MMP-8 during the 2-year regimen of SDD reflected bone resorption and may have enhanced bone forma- decreased levels of 65- to 75-kDa leukocyte-type col- tion, with the net effect being no change in the bone lagenase, because the smaller molecular weight (45 to turnover marker osteocalcin (placebo treatment had 55 kDa) mesenchymal forms of this proteinase did not no effect on either GCF biomarker: ICTP or osteocalcin).
seem to be affected. The ;50% reduction of collagenase These clinical results are consistent with earlier stud- activity (compared to its own baseline) was measured ies10,25-28 using cell culture and animal models of by a functional assay using, as a substrate, an octa- bone-deficiency disease in which TCs, such as doxy- peptide with the mammalian collagenase–susceptible cycline and minocycline, and the chemically modified Gly–Ile peptide bond. The similar (less than complete) TC derivative (CMT-1) increased bone formation and reduction of MMP-8 (collagenase-2) protein levels inhibited bone resorption. Although it has long been was measured using a polyclonal antibody to this genet- assumed that elevated levels of collagenase activity ically distinct type of collagenase. As we described pre- in humans likely reflects pathologic/ongoing collagen viously,4,33 complete inhibition of MMPs (in contrast to a breakdown, based on well-established biologic con- reduction of just the pathologically excessive levels of cepts (i.e., the ability of only collagenase[s], but not MMPs and their activity) may not be desirable because other neutral proteinases, to degrade the undenatured these neutral proteinases have physiologic functions triple-helical collagen molecule under physiologic such as processing of anti-inflammatory cytokines conditions of pH and temperature), to the best of and chemokines, which are needed for host defense.
our knowledge, this was the first study20 to directly In this regard, the current clinical trial also demonstrated link these two biochemical events in subjects in situ.
that SDD (relative to placebo) significantly reduced The current, more definitive clinical trial confirmed GCF IL-1b (a proinflammatory and bone-resorbing this link by demonstrating, in the same pooled GCF cytokine) and alveolar bone height loss16 in the subjects samples, a strong statistically significant linear rela- who were PM for >5 years. The amplitude and duration tionship between the level of collagenase activity and (2 years) of these effects of SDD therapy in the absence the ICTP degradation products of type I collagen, pre- of significant adverse events (AEs)16-18 provided further sumably released during resorption of alveolar bone at evidence of the therapeutic potential of SDD in subjects the same pocket site (based on the bone-specific pyr- with periodontitis characterized by alveolar bone loss idinoline content of the collagen telopeptide cross-link and, perhaps, in subjects with systemic bone loss.30 fragments) in these PM women with local and systemic Regarding the latter condition, serum biomarkers of bone loss. In fact, this positive correlation was main- bone remodeling (note that SDD produced a significant tained for placebo- and SDD-treated subjects over reduction in biomarkers of systemic bone resorption in the 2-year protocol. The beneficial effects of SDD on serum, at least in subgroups of these PM subjects34) as these biomarkers of connective tissue and bone de- well as systemic inflammation are being analyzed for the struction during this long-term clinical trial likely con- current clinical trial and, recently, markers of systemic tributed to improved clinical and radiologic measures inflammation, such as C-reactive protein, were found of periodontal disease severity (at least in subgroups to reflect susceptibility to skeletal bone–deficiency dis- of these subjects) described by us previously.16,17 ease (PM osteoporosis).35 This acute-phase protein in Moreover, abnormally elevated levels of these GCF blood samples also was reduced by SDD administration biomarkers might signal an increased susceptibility in subjects with severe cardiovascular disease.36 The of these subjects to more severe periodontal (including data on biomarkers of systemic inflammation (e.g., alveolar bone) breakdown if their commitment to reg- C-reactive protein) in serum samples of these PM osteo- ular periodontal maintenance therapy (provided to penic women with periodontitis are being analyzed and these subjects at no cost to them during the 2-year pro- will be reported elsewhere.
tocol to enhance compliance) should ever falter.
As described previously,16 and of clinical impor- Perhaps the most dramatic effect of SDD therapy tance particularly considering that all subjects received was the strong, long-term reduction in collagenase ac- the study medications daily over a prolonged period tivity in the periodontal pockets of PM women, a finding (2 years), AEs (such as gastrointestinal upset, infec- supported by an equally dramatic reduction in protein tion, and aches/pains) were similar for the placebo- levels of the most predominant type of collagenase, and SDD-treated groups. However, significantly fewer MMP-8, in the GCF (MMP-8 accounted for ;80% of SDD subjects experienced a dermatologic AE, such the total collagenase protein, with much smaller rela- as rash, acne, rosacea, and hives, during the clinical tive amounts of MMP-13 [0% to 18%] and MMP-1 [0% trial (2% for the SDD group versus 17% for the placebo to 9%], which is very similar to the pattern described group; P = 0.002). These data are consistent with J Periodontol • August 2008 Golub, Lee, Stoner, et al.
previous studies showing evidence of the safety and 2. Golub LM, Ramamurthy NS, McNamara TF, Greenwald efficacy of SDD in adults with the inflammatory skin RA, Rifkin BR. Tetracyclines inhibit connective tissue diseases acne and rosacea37-39 and in subjects with breakdown: New therapeutic implications for an oldfamily of drugs. Crit Rev Oral Biol Med 1991;2:297-321.
the inflammatory joint disease, rheumatoid arthritis.40 3. Golub LM, Lee HM, Ryan ME, Giannobile WV, Payne J, Sorsa T. Tetracyclines inhibit connective tissue break- down by multiple non-antimicrobial mechanisms. Adv The authors acknowledge the following individuals Dent Res 1998;12:12-26.
for their dedication to this clinical trial: E. Boilesen, 4. Sorsa T, Tjaderhane L, Konttinen YT, et al. Matrix programmer/analyst II, College of Public Health Of- metalloproteinases: Contribution to pathogenesis, diag-nosis and treatment of periodontal inflammation. Ann fice of the Dean, A. Lahners, research coordinator, Department of Biostatistics, College of Public Health, 5. Ryan ME, Usman A, Ramamurthy NS, Golub LM, University of Nebraska Medical Center, M. Morris, re- Greenwald RA. Excessive matrix metalloproteinase search nurse coordinator, Department of Biostatis- activity in diabetes: Inhibition by tetracycline ana- logues with zinc reactivity. Curr Med Chem 2001;8:305-316.
Nebraska Medical Center, J. Layton, research coordi- 6. Gomes BC, Golub LM, Ramamurthy NS. Tetracyclines nator, Department of Surgical Specialties, College of inhibit parathyroid hormone-induced bone resorption Dentistry, University of Nebraska Medical Center, T.
in organ culture. Experientia 1984;40:1273-1275.
Meinberg, dental hygenist, Department of Surgical 7. Golub LM, Ramamurthy N, McNamara TF, et al.
Specialties, University of Nebraska Medical Center, Tetracyclines inhibit tissue collagenase activity. A newmechanism in the treatment of periodontal disease. J T. Powell, office associate 1, Dental administration, Periodontal Res 1984;19:651-655.
College of Dentistry, University of Nebraska Medical 8. Golub LM, Ramamurthy NS, Llavaneras A, et al. A Center, M. Schmid, research technologist II, Dental chemically modified nonantimicrobial tetracycline Administration, College of Dentistry, University of (CMT-8) inhibits gingival matrix metalloproteinases, peri- Nebraska Medical Center, and Ruth Tenzler, research odontal breakdown, and extra-oral bone loss in ovariec-tomized rats. Ann N Y Acad Sci 1999;878:290-310.
nurse, Stony Brook University School of Dental Med- 9. Golub LM, Ramamurthy NS, Kaneko H, Sasaki T, icine. We thank the Nebraska Periodontitis Referral Rifkin B, McNamara TF. Tetracycline administration Network and several Long Island clinicians and Stony prevents diabetes-induced osteopenia in the rat: Initial Brook University faculty for referring subjects to this observations. Res Commun Chem Pathol Pharmacol clinical trial. The authors thank CollaGenex Pharma- 10. Sasaki T, Ramamurthy NS, Golub LM. Tetracycline ad- ceuticals, Newtown, Pennsylvania, for providing SDD ministration increases collagen synthesis in osteoblasts of and matched placebo tablets. We also thank Rene streptozotocin-induced diabetic rats: A quantitative auto- Martin, Stony Brook University School of Dental Med- radiographic study. Calcif Tissue Int 1992;50:411-419.
icine, for typing assistance. The project was sup- 11. Rifkin BR, Vernillo AT, Golub LM, Ramamurthy NS.
ported by grant R01DE012872 from the National Modulation of bone resorption by tetracyclines. Ann NY Acad Sci 1994;732:165-180.
Institute of Dental and Craniofacial Research (NIDCR) 12. Craig RG, Yu Z, Xu L, et al. A chemically modified (JBP, principal investigator [PI], and LMG, co-PI). The tetracycline inhibits streptozotocin-induced diabetic content is solely the responsibility of the authors and depression of skin collagen synthesis and steady-state does not necessarily represent the official views of the type I procollagen mRNA. Biochim Biophys Acta 1998; NIDCR or the National Institutes of Health. Additional 13. Tezal M, Wactawski-Wende J, Grossi SG, Dmochowski support was provided by a grant to TS from the Acad- J, Genco RJ. Periodontal disease and the incidence of emy of Finland, Helsinki, Finland, and Helsinki Uni- tooth loss in postmenopausal women. J Periodontol 2005; versity Central Hospital research funds. Dr. Golub is listed as an inventor on several patents for the drug 14. Payne JB, Zachs NR, Reinhardt RA, Nummikoski PV, mentioned in this publication, and these patents have Patil K. The association between estrogen status andalveolar bone density changes in postmenopausal been fully assigned to his institution, State University of women with a history of periodontitis. J Periodontol New York at Stony Brook. Drs. Golub and Ryan were consultants for CollaGenex Pharmaceuticals. Dr. Sorsa 15. Payne JB, Reinhardt RA, Nummikoski PV, Patil KD.
is listed as an inventor on four oral fluid biomarker/ Longitudinal alveolar bone loss in postmenopausal diagnostic patents. Drs. Lee, Stoner, Reinhardt, Wolff, osteoporotic/osteopenic women. Osteoporos Int 1999;10:34-40.
Nummikoski, and Payne report no conflicts of interest 16. Payne JB, Stoner JA, Nummikoski PV, et al. Subantimi- related to this study.
crobial dose doxycycline effects on alveolar bone loss inpost-menopausal women. J Clin Periodontol 2007;34: 1. Golub LM, Lee HM, Lehrer G, et al. Minocycline reduces 17. Reinhardt RA, Stoner JA, Golub LM, et al. Efficacy of gingival collagenolytic activity during diabetes. Prelim- subantimicrobial dose doxycycline in post-menopausal inary observations and a proposed new mechanism of women: Clinical outcomes. J Clin Periodontol 2007;34: action. J Periodontal Res 1983;18:516-526.
SDD and GCF Biomarkers in Postmenopausal Women Volume 79 • Number 8 18. Walker C, Puumala S, Golub LM, et al. Subantimicrobial 31. Payne JB, Reinhardt RA, Nummikoski PV, Golub LM.
dose doxycycline effects on osteopenic bone loss: Abstract title. Doxycycline effects on oral bone loss in Microbiologic results. J Periodontol 2007;78:1590-1601.
postmenopausal women. J Dent Res 2001;80:55.
19. Golub LM, McNamara TF, Ryan ME, et al. Adjunctive 32. Looker AC, Bauer DC, Chesnut CH 3rd, et al. Clinical treatment with subantimicrobial doses of doxycycline: use of biochemical markers of bone remodeling: Current Effects on gingival fluid collagenase activity and attach- status and future directions. Osteoporos Int 2000;11: ment loss in adult periodontitis. J Clin Periodontol 2001; 33. Sorsa T, Golub LM. Is the excessive inhibition of matrix 20. Golub LM, Lee HM, Greenwald RA, et al. A matrix metalloproteinases (MMPs) by potent synthetic MMP metalloproteinase inhibitor reduces bone-type colla- inhibitors (MMPIs) desirable in periodontitis and other gen degradation fragments and specific collagenases inflammatory diseases? That is: ‘Leaky' MMPIs vs in gingival crevicular fluid during adult periodontitis.
excessively efficient drugs. Oral Dis 2005;11:408- Inflamm Res 1997;46:310-319.
21. Uematsu S, Mogi M, Deguchi T. Interleukin (IL)- 34. Golub LM, Lee HM, Stoner J, et al. Bone turn- 1beta, IL-6, tumor necrosis factor-alpha, epidermal over markers in postmenopausal-osteopenic women growth factor, and beta 2-microglobulin levels are with periodontitis (POWP): Subantimicrobial-dose- elevated in gingival crevicular fluid during human doxycycline (SDD). J Dent Res 2008;87: Spec. Issue orthodontic tooth movement. J Dent Res 1996;75: B, Abstract No. 3491.
35. Kim BJ, Yu YM, Kim EN, Chung YE, Koh JM, Kim GS.
22. Kiili M, Cox SW, Chen HY, et al. Collagenase-2 (MMP-8) Relationship between serum hsCRP concentration and and collagenase-3 (MMP-13) in adult periodontitis: biochemical bone turnover markers in healthy pre- Molecular forms and levels in gingival crevicular fluid and postmenopausal women. Clin Endocrinol (Oxf) and immunolocalisation in gingival tissue. J Clin Peri- 36. Brown DL, Desai KK, Vakili BA, Nouneh C, Lee HM, 23. Liang KY, Zeger SL. Longitudinal data analysis using Golub LM. Clinical and biochemical results of the generalized linear models. Biometrika 1986;73:13-22.
inhibition with subantimicrobial 24. Herr AE, Hatch AV, Giannobile WV, et al. Integrated doses of doxycycline to prevent acute coronary syn- microfluidic platform for oral diagnostics. Ann N Y dromes (MIDAS) pilot trial. Arterioscler Thromb Vasc Acad Sci 2007;1098:362-374.
25. Polson AM, Bouwsma OJ, McNamara TF, Golub LM.
37. Sapadin AN, Fleishchmajer R. Tetracyclines: Nonan- Enhancement of alveolar bone formation by tetracy- tibiotic properties and their clinical implications. J Am cline administration in squirrel monkeys. J Appl Res Acad Dermatol 2006;54:258-265.
Clin Dentist 2005;2:32-42.
38. Del Rosso JQ, Webster GF, Jackson M, et al. Two 26. Williams S, Barnes J, Wakisaka A, Ogasa H, Liang CT.
randomized phase III clinical trials evaluating anti- Treatment of osteoporosis with MMP inhibitors. Ann N inflammatory dose doxycycline administered once Y Acad Sci 1999;878:191-200.
daily for treatment of rosacea. J Am Acad Dermatol 27. Williams S, Wakisaka A, Zeng QQ, et al. Minocycline prevents the decrease in bone mineral density and 39. Skidmore R, Kovach R, Walker C, et al. Effects of trabecular bone in ovariectomized aged rats. Bone subantimicrobial-dose doxycycline in the treatment of moderate acne. Arch Dermatol 2003;139:459-464.
28. Aoyagi M, Sasaki T, Ramamurthy NS, Golub LM.
40. O'Dell JR, Elliott JR, Mallek JA, et al. Treatment of Tetracycline/flurbiprofen combination therapy modu- early seropositive rheumatoid arthritis: Doxycycline lates bone remodeling in ovariectomized rats: Prelim- plus methotrexate versus methotrexate alone. Arthri- inary observations. Bone 1996;19:629-635.
tis Rheum 2006;54:621-627.
29. Zernicke RF, Wohl GR, Greenwald RA, Moak SA, Leng W, Golub LM. Administration of systemic matrix Correspondence: Dr. Lorne M. Golub, Department of Oral metalloproteinase inhibitors maintains bone mechan- Biology and Pathology, School of Dental Medicine, Stony ical integrity in adjuvant arthritis. J Rheumatol 1997; Brook University, Stony Brook, NY 11794. Fax: 631/632- 9705; e-mail: [email protected].
30. Payne JB, Reinhardt RA. Potential application of low-dose doxycycline to treat periodontitis in post- Submitted December 3, 2007; accepted for publication menopausal women. Adv Dent Res 1998;12:166-169.
February 19, 2008.

Source: http://www.sidp.it/progetti/www.periomedicine.it/newserfile/53/78aaff/GolubLMetal_JPerio2008.pdf