`
`1990 VOLUME 25 ISSUE 6
`SISAC
`
`1111111
`:6-7
`0022-34
`Ref:S048203/ 02
`
`('1
`
`-o ... (r-
`(r-oo -0
`
`
`
`ISSN 0012-3<:x4
`
`nE,qR|:.H
`'_I
`JUURNQL DF ?EPIUDDNTHL RE’
`
`1990 UIJLIJHE 25 ISSUE ‘J
`
`gIilmIIIIIIIII1I|J|{
`
`0022-3aE:4<1?'?-3.i'25=°
`Fe.e+:sc:-432-:»3x"r.w2
`
`"
`
`D1S92?1b
`
`This material was copier:
`2: the NW and ray be
`Subject us Copyright Laws
`
`- ROY c. PAGE, SEATTLE
`. ASSOCIATE EDITORS:
`KI, FARMINGTON
`
`Exh. 1048
`
`
`
`Editor
`Roy C. Page. D.D.S.
`Director, Resea rch Center in
`O ra l Bio logy, SM-42
`Uni versity o f Wash ington
`Seattle, Washi ngton 98 195
`U.S.A.
`
`Associate Editors:
`Kla us N uki,
`Farmington, CT, U .S.A.
`Hiroshi Okacla,
`Osaka , Japa n
`Ubele van der Velden,
`Ams!"rdam , Tbc Netherla nds
`
`Editorial Board:
`Ju kka K. Ainamo ,
`Helsin ki, Finla nd
`Ca rl L. Ba nd !,
`Mi nneapolis, M N, U.S.A.
`P. Ma rk Ba rtold,
`Adclaide, Austra lia
`J-Ienni ng Birkeda i-H a nscn,
`Birmingham, AL, U .S.A.
`Jci'frey Ebcrsolc,
`San Ant onio , T X, U .S.A.
`Robcrl Frank,
`Strasbourg, Francc
`Philias R. Gara nt, Stony Brook.
`NY, U .S.A.
`Lo rne M. Golub,
`Sto ny Brook, NY, U.S.A .
`Ha ns G rönda hl ,
`Gothenburg, Sweden
`Bernard G uggenhcim,
`Zü rich, Switzcrla nd
`Kohji Ha ra,
`Niiga ta, Japa n
`Newell W. Jo hnson,
`Londo n, England
`Thorkild Ka rring,
`Ärhus, Denma rk
`Kenneth S. Kornma n,
`Sa n Antonio, TX , U .S.A.
`Max A. Listgarten,
`Philadelphi a, PA, U .S.A.
`Antho ny J-1 . Melchcr,
`Toronto, O nta rio, Canada
`A. Sampa th Na raya na n, .
`Seattle, WA, U .S.A .
`E. Newbrun ,
`Sa n Francisco, CA, U .S.A.
`Richa rd R. Ra nney,
`Birmingha m, A L, U.S.A.
`Pa ul B. Robcrtson,
`Va nco uver, B.C., Canada
`Huber! E. Schroedcr,
`Zürich, Switzcrla nd
`C hristophcr A. Squicr,
`lowa City, lA, U.S.A.
`An nc Ta nner,
`Boston, MA. U.S.A.
`Tho rnas E. Ya n Dykc,
`Atl a nta, GA, U.S.A .
`Joseph Zambon,
`Bu fTalo, NY, U.S.A.
`
`pcriodontal
`rcscarch
`
`Aim and Scope
`The Journal o/Periodonta/ Research is an in ternatio nal research peri od-
`ical the purpose ofwhich is to publish original investigatio ns concerned
`with every aspect of periodontology and rela ted sciences. Review a rti-
`cles, reports o f scientific meetings in peri odontology and o ther short
`commu nicatio ns may also be accepted. O nly articles written in English
`are accep ted and onl y if Lhey have no t been and will not be published
`elsewhere.
`
`Short papers ( 1- 3 pages) are accepted and Special efforts are made in
`order to ensure ra pid publicatio n of these.
`
`Ma nuscripts ancl ecli torial correspondence shoulcl be sent to the ecli tor.
`Acldress above. The Journal ol Periodantat Research has no objecti ons
`to the reprocl uction of sho rt passages ancl illustratio ns fro m this journal
`without further formality than acknowleclgement of the source. All
`business communica tio ns should be sent to the publisher.
`
`Subscription
`Published bi monthly. Subscription price 1990 per volume of six issues:
`DKK 11 86.00 including postage (GBP 11 6.00, DEM 324.00). USA, Canada
`and Japan: USD2 10.00 including postage and air freight, payable in advance.
`Prices are subject to exchange-rate nuctuations.
`Orders should be sent to:
`M unksgaa rcl , International Publishers Ltcl.
`35 NötTe Sögade, Postbox 2 148,
`DK -1 0 16 Copenhagen K , Denma rk
`
`rg 1990 M un ksgaard Interna tiona l Publishers Ltd . A ulhorizat io n to pho toeopy items
`fo r inlern al o r persona l use, or the internal or personal use of specifi c clien ts, is gra nted
`by M un ksgaa rd Intern ati ona l Publishers, Lld. Cor libraries and other users rcgisterccl
`with the Copy ri ght Clea ra nce Center (CCC) Tra nsacti ona l Repo rting Service, proviclecl
`tha t the base fcc o f $ 2.50 pcr copy is pa id directl y to CCC, 27 Congrcss Street, Salem,
`MA 0 1970. 0022-3484/90/$02.50+ 0.00. All othcr rights, incl ucli ng micro film , rescrvcd.
`
`Journal o f Periodontal Research (ISSN 0022-3484)
`is p ublished bi monthl y by
`M un ksgaa rd Interna ti onal Pu blishcrs, 35, N01Te Segadc, P.O. Box 2 148, DK- 10 16
`Copcnhagcn, K, Dcnmark. USA Sub ·eripti on priee is USO 2 10.00 includ ing a irspced
`deli vcry. Seeond class postage paicl at Jama ica, NY 1143 1. USA POSTMASTER for
`N.A. subscri bers: Send add ress ehanges to Pu blications Expcd iti ng lnc., 200 Meacham
`Avenue, Elmon t, NY II 003. Ai r frcigh! and mai ling in the USA by Pu blica ti ons
`Expcdit ing. Printcd in Denma rk by P..l . Schmidt , Yojens.
`
`f his matNia l w ascopie d
`atth e NLM a nd may !)e
`Subje.ct USCo.pyright l aws
`
`Exh. 1048
`
`
`
`Low-dose doxycycline therapy:
`Effect on gingival and crevicular
`fluid collagenase activity in
`humans
`
`Golub LM, Ciancio S, Ramamurthy NS, Leung M, McNamara TF: Low-dose
`doxycydine therapy: Eff"ect an gingiva/ and crevicular fluid collagenase activity
`in humans. J Periodont Res 1990; 25: 321- 330.
`
`Tetracyclines are now recognized to have non-antimicrobial properlies with
`therapeutic potential - for example, these agents can inhibit pathologic collage-
`nolysis by blocking mammalian collagenases and other matrix-degrading metal-
`loproteinases. In the current study, adult human subjects with moderate chronic
`periodontitis were administered specially formulated capsules of doxycycline,
`.;ontaining lower-than-usual amounts of this semi-synthetic tetracycline, on a
`daily basis for 2 weeks prior to a full-thickness flap procedure; control subjects
`were administered placebo capsules. The gingiva excised during this surgica l
`procedure were extracted, the extracts partially purified and analyzed for Col-
`lagenase activity using [3H-methyl] collagen as substrate and the tech niques
`of SDS-PAGE/fl:.IOrography or liquid scintillation spectrometry. In the absence
`of any drug pre-treatment, or after a 2-wk regimen of placebo capsules, the
`gingival extracts exhibited pathologically-excessive mammalian collagenase ac-
`tivity. The 2-wk regimen of low-dose doxycycline capsules reduced this activity
`by approximately 60- 80% (p < 0.05 and < 0.0 I, respectively); in vitro exposure
`of the gingival extract to doxycycline also inhibited its collagenase activity.
`Collagenase activity in the crevicula r fluid of periodontal pockets of an ad-
`ditional group of subjects was also significa ntly reduced, as was the severity
`of inflammation at the same gingival sites. The results suggest that a regimen
`of low-d0se doxycycline capsules may provide a safe (other studies indicate
`that this regimen may not induce tetracycline resistance in the subgingival
`plaque) and effective adjunct to instrumentation therapy in the management of
`J thologic collagenolysis in the periodontal patient. However, further studies
`are necessary to confirm this hypothesis.
`
`L. M. Golub, S. Ciancio*,
`N. S. Ramamurthy, M. Leungt and
`T. F. McNamara
`Oept. of Oral Biology and Pathology, School
`of Dental Medicine, S.U.N.Y. at Stony Brook,
`Stony Brook, N.Y. , Oept. of Periodontics·,
`S.U.N.Y. at Buffalo, Buffalo, N.Y. and Dept. of
`Chemistryt, S.U.N.Y. at Old Westbury, Old
`Westbury, N.Y., U.S.A.
`
`Accepted for publication March 13, 1990
`
`lntroduction
`Collagen breakdown is an essential pathway in
`the pathogenesis of periodontal and other diseases
`such as rheumatoid arthritis, corneal ulcers, the
`skin blistering disorder epidermolysis bullosa, and
`excessive bone resorption. The activity of collagen-
`ase, one of a series of matrix-degrading proteinases
`generated by the host tissues, is a rate-limiting step
`in pathologic collagenolysis (1 - 6). Recent thera-
`peutic concepts are beginning to address the poten-
`tial value of inhibitors of mammalian collagenases
`(see below) and collagenases from periodontopa-
`thic microorganisms such as Bacteroides gingivalis,
`Actinobacillus actinomycetemcomitans, and species
`of Spirochetes and Bacillus (7, 8). Examples of
`
`inhibitors of mammalian collagenases (activity or
`production) include retinoic acid (9), phenytoin (5),
`eriochrome black T (10) and synthetic compounds
`such as Cl-1
`( 11 ) and phosphonamidates (8).
`[These are in addition to the endogenaus physiol-
`ogic inhibitors, a2-macroglobulin, small cationic
`proteins and tissue inhibitor ofmetallo proteinases,
`or TIMPs (12)]. Chia ranil (quinone derivative), as
`weil as phosphonamidate and phosphoramidate,
`is being investigated as an inhibitor of bacterial
`collagenases (8, 13).
`Consistent with this potential therapeutic ap-
`proach, Golub and colleagues reported, using dif-
`ferent experimental systems and ln humans, that
`tetracycline antibiotics (long advocated as useful
`adjuncts in periodontal therapy) can inhibit mam-
`
`L
`
`This matE rial w as copoiEil
`a tth ~ NLM and may !)~
`Sui)jEtt US Copyright l aws
`
`Exh. 1048
`
`
`
`322
`
`Golub et al.
`
`m::dian collagenases and collagen breakdown by a
`mechanism independent of the antimicrobial efft-
`cacy of these drugs ( 14-25); this effect was recently
`confirmed in other laboratories (26, 27). Tetracy-
`clines may also inhibit other metalloproteinases
`(gelatinase, type IV /V collagenase, macrophage
`elastase) involved in connective tissue and base-
`ment membrane degradation (24, 25, 28, 29) in-
`cluding a collagenase produced by the periodonto-
`pathogen, Bacteroides gingivalis (27). In several
`studies on humans, routinely prescribed, antimi-
`crobially-effective doses of tetracyclines (tetracy-
`cline, minocycline, doxycycline) were found to re-
`duce the collagenase activity in the fluid of the
`periodontal pocket ( 14-17) which originales from
`the adjacent host tissues (16, 30- 32). The current
`study was carried out to determine whether a new-
`er, semi-synthetic tetracycline, doxycycline, could
`be administered to humans in a low-dose regimen
`(20 or 30 mg capsules, rather than the 50 or I 00 mg
`capsules which are commercially available) which
`would effectively inhibit collagenase activity in the
`gingival tissue as weil as in crevicular fluid. Doxycy-
`cline was chosen, in part, because a recent study
`by Burns et al. (26), which compared the relative
`potencies of different collagenase inhibitors includ-
`ing tetracyclines, reported that doxycycline (with
`an IC50 = 15 J1M) was a more potent collagenase
`inhibitor than two other tetracyclines, minocycline
`(lC50 = 190 J1M) and tetracycline (IC50 =350 J1M).
`
`Material and Methods
`Selection and management of human subjects
`Two studies (one on gingival tissue, the other on
`gingival crevicular fluid or GCF; see below) were
`conducted in which adult human subjects, between
`the ages of 35 and 56 years, with moderate adult
`periodontitis were prescribed a 2-wk regimen of a
`lower-than-usual dose of doxycycline, a newer
`semi-synthetic tetracycline. The subjects were de-
`termined by history to have had no dental treat-
`ment, including antibiotic therapy, for at least 4
`months prior to starting the experimental trial and
`to have no contraindications (e.g. allergy to tetra-
`cyclines) to the doxycycline regimen. All subjects
`gave written informed consent using a document
`approved by the University's Human Use Commit-
`tee. Special capsules of doxycycline were formu-
`lated, and dispensed for the studies by a licensed
`pharmacist, each containing 20 or 30 mg of Vibra-
`mycin''Y (Pfizer Pharmaceuticals, Groton, CT) with
`microcrystalline cellulose (Avicei'!Y) used as the in-
`active filler in each capsule. The subjects were in-
`structed to take one capsule in the morning and
`one al night, but not at mealtimes. Note that the
`recommended dose of doxycycline, when the drug
`
`is prescribed to combat infection, is 100 mg or
`50 mg two times per day; the former regimen is
`prescribed for the first 24 h as a loading dose and
`the latter is administered daily thereafter as the
`maintenance dose.
`
`Collection and analysis of gingival tissue (study no. 1)
`The first study included 8 human subjects, 4 of
`whom received the test regimen (30 mg doxycycline
`b.i.d . for 2 wk), the other 4 received the Placebo
`(AviceJ!w caps b.i.d. for 2 wk), according to the
`following experimental protocol: Subjects were se-
`lected who exhibited similar moderate severity of
`adult periodontitis bilaterally in their maxillary
`right and left posterior sextants, and who were
`judged to require full-thickness flap surgery as part
`of their therapy in both areas. Each subject re-
`ceived a standardized regimen of presurgical ther-
`apy in the test quadrants consisting of oral hygiene
`instruction and scaling and root planing. Immedi-
`ately prior to the surgical procedure in the right
`maxillary arch (and, 2 wk later, in the left maxillary
`arch), the gingival index (33) and packet depth
`were measured on the mesial and distal surfaces of
`each tooth tobe included in the surgical procedure.
`No prescription was given to the subjects prior to
`the first surgery (right side). The gingival tissues
`were removed during a full-thickness flap pro-
`cedure by an inverse bevel incision beginning ap-
`proximately 1- 1/2 mm apical to the free margin of
`the gingiva and ending slightly buccal to the al-
`veolar crest (care was taken to include the "col"
`area in the gingival tissue removed for analysis).
`Lidocaine 2% with I /100 000 epinephrine was used
`as the local anesthetic for all surgical procedures
`in the study, with care being taken to inject the
`solution only in the periapical tissues away from
`the gingiva to be analyzed. The tissues, once re-
`moved, were immediately and brietly washed to
`remove blood and debris, blotted, placed in coded
`glass scintillation bottles and frozen at - 80°C until
`analyzed. After completing the surgical procedure
`on the right side, each subject was given a vial
`containing a 2-wk supply (28 capsules) of low-dose
`doxycycline (30 mg caps; n = 4 subjects) or the
`placebo (Avicelf!P caps; n =4 subjects) with instruc-
`tions to take the capsules twice per day until the
`next surgical procedure. After the 2-wk regimen,
`the subjects were treated with the same type of
`surgical procedure in the left maxillary arch and
`the gingival tissue from this area was collected,
`washed and stored frozen in the coded bottles.
`The right and left side tissues were analyzed for
`collagenase activity, as described below, under
`blinded conditions.
`The procedure to extract, partially-purify, and
`
`f h.is mat erial w as <•Opie-d
`atthe NLM and may b<e
`Su b<j e ct US Copyright Laws
`
`Exh. 1048
`
`
`
`measure collagenase activity in the gingival speci-
`mens was essentially the same as that described
`previously (16, 34). In brief, the gingival tissues
`from each sextant (one right side- and one left
`side-sample per subject) were thawed, weighed and
`minced with scissors (all procedures at 0- 4°C). A
`known amount of minced tissue was then homog-
`enized, extracted ( 100 mg wet weight gingival
`tissue/5 ml buffer) first with I 0 mM Tris-HCI buf-
`fer (pH 7.5) containing 0.4 M sucrose and 5 mM
`CaCI2, and then with 50 mM Tris-HCI containing
`0.2 M NaCI, 5 mM CaCI2 and 5 M urea. After
`exha ustive dialysis
`(and after centrifugation)
`(NH4) 2 S04 was added to the supernatant to pro-
`duce 60% saturation and the precipitate was col-
`lected by centrifugation and dissolved in the Tris-
`NaCI-CaCI2 buffer. The enzyme preparation was
`exhaustively dialyzed against the same buffer and
`analyzed immediately for collagenase activity. As
`described by us previously ( 14, 16), 70 pl of par-
`tially purified gingival extract was incubated with
`lO pl of [3H-methyl] collagen ( 113 000 DPM; 10 {tg
`collagen) as substrate [the nativity of the collagen
`was ensured by its resistance to trypsin digestion
`and by circular dichroism measurements confirm-
`ing its helical conformation (14, 16, 49)] and 10 pl
`of p-aminophenylmercuric acetate (APMA) was
`added in a final concentration of 1.2 mM. The
`incubation was carried out for 24 h at 22°C and
`the collagen components and degradation products
`were examined, after thermal denaturation, by a
`combination of SDS-polyacrylamide gel electro-
`phoresis (SOS-PAGE) and tluorography (14). The
`co nversion of the collagen a components to the aA
`collagenase digestion products was calcula ted (31)
`after the tluorographs were a nalyzed densitome-
`trically using a Beckman DU8 spectrophotometer
`equipped with a film tracing attachment.
`Add itional aliquots of the gingival extract from
`the subjects treated either with no medication, pla-
`cebo, or low-dose doxycycline ca ps were incubated
`wi th the [3H -methyl] collagen a t 27°C for 18 h, the
`reaction was stopped and the undigested collagen
`was precipitated by the addition ofphenanthroline,
`dioxane and non-radioactive methylated collagen
`as carrier. The radiolabeled degradation products
`were counted in a liquid scintillation spectrometer
`(see Golub et al. (14, 16) for details).
`Following the analyses of the gingival tissues
`for each subject, the remaining gingival extracts,
`obtained only from the subjects during the first
`surgical procedure (prior to therapy with either the
`doxycycline or placebo capsules), were pooled and
`aliquots were incubated with the 3H-collagen at
`22''C, 18 h. The purpose was to determine whether
`different tetracyclines, including two which were
`chemically-modified
`to eliminate
`their antimi-
`
`Low-dose doxycycline and gingiva/ collagenase
`
`323
`
`crobial efficacy (35, 36), could directly inhibit hu-
`man gingival tissue collagenase. Minocycline or
`either of two chemically-modified tetracyclines, 4-
`de-dimethylaminotetracycline (prepa red and char-
`acterized in our laboratory; 23, 25), o r 6-deoxy 6-
`(pro-
`demethyl 4-de-dimethylaminotetracycli ne
`vided by Johnson & Johnson Dental Ca re, Inc. ,
`New Brunswick, N.J.) were added to the incu-
`bation mixture in a final concentration of 20 pg/
`ml and collagenase activity in the presence or ab-
`sence of these tetracyclines was analyzed using the
`SDS-PAGE/fluorography technique.
`
`Collection and analysis of GCF (study no. 2)
`In the second study, 4 adult humans with moder-
`ately severe periodontal disease at the sites mo ni-
`tared (see Fig. 4 for the pretreatment or time = 0
`mean values for pocket depth, gingival index and
`GCF tlow) were admin istered a 2-wk regimen of
`Iow-dose doxycycline (20 mg capsules, b.i.d. ; this
`formulation was selected to determine whether the
`"low-dose regimen" described in study no. l could
`be reduced even further and still be able Lo inhi bit
`collagenase activity in the periodontal pocket). T he
`subjects were monitored, as described below, im-
`their prescripti o n
`mediately befo re
`receiving
`(Time=O) and at
`l wk (Time= l) and 2 wk
`(Time=2) after initiating medication. No dental
`treatmentwas administered during the experimen-
`tal protocol and the subjects stated that they had
`not received dental treatment including antibiotics
`for at least 4 months prior to Time = 0.
`The techniques for collecting and ana lyzing the
`GCF samples have been described previously ( 16,
`17). In brief, for each subject, GCF was collected
`on sterile filter paper strips (Periopaper"''; ID E ln-
`terstate, Amityvi lle, N.Y.) inserted in to 7- 8 in ter-
`proximal pockets in the maxillary arch for a 30-
`second time period . The fluid vo lume was immedi-
`ately determined on a calibrated Periotron 6000
`(ID E Interstate) and the fi lter paper strips (each
`placed into a coded microfuge tube) were then
`frozen on dry ice. After l h, the collagenase activity
`on each stripwas ana lyzed using [3H-methyl] col-
`lagen as substrate ( 16); note, the GCF samples
`were not activated prior to o r during the incu-
`bation (18 h, 27°C) with the radiolabeled collagen .
`The gingival index (33) and packet depth werc
`measured, at the sites donating the GCF, immed i-
`ately after fluid collection.
`In additio n, 2 GCF samples were collected from
`one of lhese subjects at T = 0 (befo re drug ad minis-
`tration), the samples were activated by a brief ex-
`posure to trypsin (the trypsin was then inactivated
`by addition of a 5-fold molar excess of soybea n
`trypsin inhibitor; see Golub et al. ( 16) for detai ls),
`
`Th is m ateri a I w as co ~ i ed
`atthe· NLM and m ay ~e
`Su~j e.ct USCo~yrigh t Laws
`
`Exh. 1048
`
`
`
`324
`
`Golub et al.
`
`Table J. The effecl of low-dose doxycycline therapy on collagenase activity in human gingiva*
`Treatment with
`No Drug
`Doxycycline
`(right side)
`(left side)
`0.89±0.05 (48)
`0.85± 0.06 (40)
`4.7± 1.5 (48)
`4.6 ± 1.6 ( 40)
`
`Statistical
`Significance
`N.S.
`N.S.
`
`12.8 ± 12.8 (4)
`
`p < O.OI
`
`Gingival tissue obtained
`after no drug or doxy-
`cycline therapy
`Gingival Index
`Pockel Depth (mm)
`Gingival Collagenase
`69.0 ±3 .7(4)
`Activity (% Lysis)**
`* Each va lue represents the mean±S.E.M. ; (number of determinations).
`** Estimated by den sitometric analysis of fluorograms:
`4/3(aA + aA)
`4/3(a'1' + a]') + (a 1 + a 2)
`
`2
`
`'
`
`x I 00 ( see re f. 3 I)
`
`pooled and extracted in 450 pl Tris-NaCI-CaCI2
`buffer (I h, 22"C). Seventy-microliter aliquots of
`extract were incubated with [3H-methyl] collagen
`(see above for details) ± EDTA or doxycycline
`added in a final concentration of 25 mM and 20 {tg/
`ml, respectively, and the collagen reaction products
`assessed by SDS-PAGE/fluorography.
`
`Results
`The 8 human subjects in study no. I were selected
`tbey exhibited moderate adult perio-
`because
`dontitis and no differences in the severity of gin-
`gival inflammation and pocket depth in their right
`and left maxillary posterior arches. At the time
`of surgery, after a standardized protocol of pre-
`surgical therapy, the administration of a 2-wk regi-
`men of low-dose doxycycline or placebo bad no
`significant effect on the gingival index or pocket
`deptb (compare left side vs. rigbt side; Tables I and
`2). All of tbe gingival specimens from the subjects
`who were not administered either type of capsule
`(rigbt side) showed electrophoretic evidence of
`tissue collagenase activity (Figs. I and 2) as did the
`gingival tissues (left side) from the subjects wbo
`were administered the placebo capsules (Fig. 2) .
`Incubating these gingival extracts witb radio-
`labeled collagen resulted in the partial loss of a
`collagen components and the formation of 3/4
`cleavage products of a, called aA, wbich are cbarac-
`
`teristically produced by mammalian, including hu-
`man, collagenases (Figs. I and 2). In contrast, tbe
`gingival extracts (left side) from the subjects treated
`with low-dose doxycycline showed either no detect-
`able (Fig. L) or a reduction (separate fluorogram
`not shown) in collagenase activity. Densitometric
`analysis of the fluorograms shown in Figs. L and
`2 indicated that placebo administration produced
`no effect on Collagenase activity in the gingival
`tissues (Table 2), whereas low-dose doxycycline
`therapy reduced gingival tissue collagenase activi ty
`by about 8 L% (p < 0.0 I; Table I). When aliquots
`of the same gingival extracts were incubated with
`the PH-methyl] collagen (but at 27°C rather than
`22°C) and the collagen breakdown products were
`measured in a liquid scintillation spectrometer, the
`in vivo administration of low-dose doxycycline also
`produced a reduction (about 63%; p < 0.05) in col-
`lagenolytic activity and again the placebo therapy
`bad no effect (Table 3). Incubating the gingival
`extracts (from subjects prior to drug administra-
`tion) in vitro with three different tetracyclines (min-
`ocycline, a semi-synthetic tetracycline, which, like
`doxycycline, was previously shown to inhibit mam-
`malian collagenases (16), and two different chemi-
`cally-modified analogs) produced a similar inhi-
`bition of gingival Collagenase activity (Fig. 3) as in
`vivo low-dose doxycycline therapy.
`In the second study, 2 wk after low-dose doxycy-
`cline therapy (T = 2), the GCF collagenolytic activ-
`
`Table 2. The effect of placebo therapy on collagenasc activity in human gingiva*
`Treatment with
`G ingival tissue obtaincd
`No drug
`aftcr no drug or Placebo
`Placebo
`(right side)
`thcrapy
`(left side)
`1.04 ± 0.19 (28)
`G ingival I ndcx
`1.15 ± 0.37 (26)
`4.9 ± 1.0 (28)
`4. 7± 1.0 (26)
`Pockct Dcpth (mm)
`G ingiva l Co llagenase
`67.3±5.2 (4)
`Activity (% Lysis)**
`* Each va lue reprcsents the mea n ±S .E.M.; (no. of determin ations).
`** Estimated by dcnsitometric ana lysis:
`4/3(rl~1 + a2')
`x 1 OO
`4/3(a~ + a;' ) + (a 1 -1- a 2)
`
`70.2+ 7.4 (4)
`
`Sta tistical
`Significance
`N.S.
`N.S.
`
`N.S.
`
`f hi.s material w as co poied
`atthe NLM anl:l may b<e
`Su bj e.ct US Copoyright Laws
`
`Exh. 1048
`
`
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`Low-dose doxycyc/ine and gingival collagenase
`
`325
`
`6
`
`7 8 9
`
`10
`
`2
`
`3
`
`4
`
`5
`
`-
`
`--
`
`a,-
`
`Fig. 1. Fluorograph of eH-meth yl] collagen and its degradation
`products after SDS-polyacrylamide gel electrophoresis (PAGE).
`The 3H-collagen was incubated at 22"C for 24 h with partially-
`purified extracts of huma n gingiva . The gingiva l tissues were
`removed during periodanta t surgery first from the right side
`and , 2 wk later, from the left side of the maxillary arch of 3
`adu lt humans. Lane I = 31-1-collagen with no gingival ex tract;
`lanes 2, 4 & 6 = 31-1-collagen after incubatio n with gingiva from
`the right side o f 3 human subjects wilhout drug pre-treatmenl;
`lanes 3, 5 & 7 = 3H-collagen a fter incubation with gingiva from
`lhe lefl side of the maxillary arch of lhe same 3 subjects a fter
`a 2-wk regimen of low-dose doxycycline capsules.
`
`ity, GCF flow and gingival index were significantly
`reduced by 54% , 39% and 31%, respectively, com-
`pa red to baseline (T = 0) values (p < 0.0 I for all 3
`measurements); pocket depth showed 110 reduction
`at the 1-wk or 2-wk time periods (Fig. 4). As shown
`in Fig. 5, an extract of GCF obtained at T = 0
`(before low-dose doxycycline therapy), incubated
`wi th radiolabeled collagen in vitro, showed evi-
`dence of tissue collagenase activity (track 2) and
`ad ding 25 mM EDTA (Track 3) or 20 .ug/ml doxy-
`cycline (track 4) to the reaction mixture inhibited
`this activity.
`
`Discussion
`Tetracyclines are often advocated as useful ad-
`juncts in periodontal therapy based 011 their effec-
`tiveness against periodontopathogens (37- 41); an
`
`Fig. 2. Fl uorograph of the SOS-PAGE pattern of 31-1-collagen
`and its degradation producls after incubation with extract of
`gingiva l tissue, removed from thc right a nd left sidcs of the
`maxillary a rch o f 4 addi tional human subjccts. The experimen-
`tal protocol is the same as in Fig. I exccpt tha t thcse subjects
`were ad mini stered a 2-wk regimen o f placebo capsules prior to
`the surgery on the left side. Lanc I & 10= 31-1-collagen alone;
`lancs 2, 4, 6 & 8= 3H -collagen after incuba tion with gingiva
`from the righ t maxillary arch of the 4 subjects without drug
`pre-treatment; lanes 3, 5, 7 & 9= 3H-collagcn after incubation
`with gingiva from the left side of the same 4 subjects after a 2-
`wk regimen of placebo capsulcs.
`
`additional advantage is their u11ique ability, among
`antibiotics, to be highly concentrated within the
`fluid of the periodontal pocket (40, 42). Recently,
`novel properties of these drugs have been discover-
`ed which help explain their clinical effectiveness
`and may also expand their future applications be-
`yond their current use as antimicrobials. As one
`example, tetracycli11es now appear to possess anti-
`inflammatory properties when administered to pa-
`tients with certain skin diseases, diseases such as
`rosacea, dermatitis herpetiformis and pyoderma
`gangrenosum which are not believed to have a
`microbial etiology (43, 44, 58, 59). Several non-
`antimicrobial mechanisms may be involved, includ-
`ing tetracycline's ability to inhibil prostaglandin
`production (at least in high concentrations of the
`drug), to suppress leukocyte aclivity, a nd lo ful1c-
`tion as scavengers of superoxicle radical (45- 47).
`In additio11, the object of the current study - an
`ability of tetracyclines to inhibit marnmalian col-
`lagenase and other metalloprolei nase aclivities (see
`
`l
`
`Th is m at e ri a I w as CQpi e-d
`atth e NLM and m ay be·
`Su bj ect US Copy right Laws
`
`Exh. 1048
`
`
`
`326
`
`Golub et al.
`
`Table 3. Thc effect of low-dosc doxycyclinc or placebo administration on collagenolytic activity in human gingiva*
`G ingiva l Collagenolytic
`G ingiva l tissue ob tained
`Activity: % 11-1-collagen
`after no d rug, doxycycline
`or Placebo therapy (Rx)
`lysed at 27"C
`No drug (right siele)
`16.0 ± 3.4
`5.9 ± 1.7
`Doxycyclinc Rx (left siele)
`No drug (right siele)
`13.2±0.4
`N.S.
`Placebo Rx (left siele)
`14.4±0.7
`* Each val ue represcnts the mea n ±S .E.M. for gingival specimens from 4 human subjects (4 subjects per control and 4 subjects
`pe r experimental gro up).
`
`Statistical
`Significance
`
`p < 0.05
`
`[ntroduction) - may also contribute to the anti-
`inflammatory efficacy of these drugs. In this re-
`ga rd, Chang et al. (48) reported that collagen
`breakdown products are chemotactic for polymor-
`phonuclear leukocytes (PMNs). Thus, a reduction
`in collagen degradation due to the anticollagenase
`properties of tetracyclines could contribute to the
`
`drug's ability to decrease the severity of inflam-
`mation. Of supreme interest, patients with dystro-
`phic epidermolysis bullosa, a serious and some-
`times life-threatening skin-blistering disorder as-
`sociated with excessive collagenase production,
`appear to respond to the anti-collagenase prop-
`erties of minocycline (58, 59) and tetracycline ad-
`ministration was recently reported to significantly
`
`2
`
`3
`
`4
`
`5
`
`200
`~
`0 u
`U)
`0
`0
`0 c.o
`c: ::
`~ Q)
`
`Cl..
`
`lll
`0
`lL
`LL..
`(..)
`(.!)
`
`..c:
`0..
`Q) a
`
`Q)
`
`"""" u
`0
`Cl..
`
`*p< O.OI,vs T=O
`•• p<0.05,vs T=O
`
`-
`
`Q)
`
`c: -Q)
`
`20
`
`~
`>..."T.:I
`·;:: ""' ::
`ü
`0>
`<t Q)
`a
`. ~
`ö
`0>
`c: .2
`Q) 0
`0> (..)
`~
`,.--,
`ö
`-;:.,
`(..)
`..c:
`Q;
`LL..
`Ii=
`(..)
`(.!)
`::z:::
`!'(')
`~
`
`0
`
`2
`
`><
`
`Q) ""' c:
`
`Fi!J. 3. f'l uorograph of thc SOS-PAGE pattern of 11-1 -co llagen
`incubated (22''C, 24 h) wi th poolcd cx tract o f huma n gi ngival
`tissuc. On ly tissues rcmovcd from the subjects witho ut any drug
`prc-trca tm cnl (i.c. from thc righl maxi llary a rch) werc used in
`this cxpcrimcnt. Lanc I = 31-1-collagen alonc; lanes 2, 3, 4 a nd
`5 = 11-1-collagcn after incuba tio n wilh cx lracl o r gingiva in the
`prcscncc of no adclcd drug (lan e 2), o r wi th minocyclinc (lane
`3), 4-dc-cl imcth ylami no lctracycli nc (lanc 4), or 6-deoxy 6-deme-
`thyl 4-clc-dimcthylam inotctracyclinc (ianc 5) acldecl to thc incu-
`ba tion mi xture in a final concentratio n of 20 pg/ml.
`
`Time, wks.
`Fig. 4. Thc efTect of a 2-wk regimcn of low-dose doxycycli ne
`on GCF fl ow and collagenolytic activity and on gingiva l inflam-
`mation (G. l. ) and pocket clepth in 4 adult human subjects. Each
`va lue reprcsents the mean ±S .E.M. a t timc = O (bcfore drug
`treatment) and I a nd 2 wk after initiating drug trea tment.
`
`f h is m at e ri a I w as copi e.d
`atthe· NLM and m ay l>e
`Sul>j ec:t USCopyright Laws
`
`Exh. 1048
`
`
`
`2
`
`3
`
`4
`
`Fig. 5. F luorograph of the SOS-PAGE pallern of -'I-I-collagen
`after incubation (22°C, 18 h) wilh a trypsin -activaled ex tract
`of GCF; the GCF was collected on filter paper strips from a
`human subject before doxycycline administration . Lane I = 31-1-
`collagen alone; lancs 2, 3 & 4 = 31-l-collagen after incuba tion
`with GCF alone (lane 2) or with GCF plus EDTA (lane 3) or
`doxycycline (lane 4) added in a final concentration of 25 mM
`and 20 Jtg/ml, rcspectively.
`
`reduce clinical signs of rheumatoid arthritis in a 4-
`month study on humans (63).
`In the current study, a low-dose regimen of doxy-
`cycline appeared to substantially reduce the col-
`lagenolytic activity in the fluid of the periodontal
`packet (GCF) and in the gingiva of humans with
`adult periodontitis. This is the first study to demon-
`strate an anti-proteinase effect of a tetracycline on
`human gingival tissues and the reduced collageno-
`lytic activity in this tissue (and in GCF) reflects an
`inhibiting of a specific matrix-degrading metallop-
`roteinase derived from the host, namely mam-
`malian collagenase. The evidence includes the abil-
`ity of the drug, bothin vivo and in vitro, to inhibit
`the breakdown of undenatured coll~tgen by an
`EDTA-inhibitable neutral proteinase, thus prevent-
`ing
`the production of aA cleavage fragments
`characteristically produced by host collagenases.
`Previous studies on humans demonstrated that reg-
`ul a r doses of different tetracyclines, as weil as low
`
`Low-dose doxycycline and gingival collagenase
`
`327
`
`doses of minocycline, produced similar reductions
`in GCF Collagenase activity (14- 17, 51). A direct
`inhibition of this metal-dependent proteinase by a
`portion of the tetracycline molecule (one of its
`metal-binding sites) not involved in the drug's anti-
`microbial activity was the mechanism proposed
`(23). In fact, experiments are now being carried
`out to localiz