Effect of addition of antimicrobial drugs to human collagen membrane

CIinicalMutrriuO 12 (1993) 169-179

Effect of Addition of Antimicrobial Collagen Membrane F. Al Arrayed, Department (Received

S. Thomas

Drugs to Human

& J. Moran*

of Periodontology,

University

2 May 1992; accepted

8 June 1992)

of Wales College of Medicine,

Heath Park, Cardiff,

CF4 4XY, UK

Abstract: Antimicrobial agents included in graft material for use in guided tissue regeneration of periodontally diseased tissue may be of value in combating infection, but may also alter the properties of the membrane material and exert an effect upon the host immune response. Metronidazole, niridazole and tinidazole were added to a cross-linked freeze-dried human type I collagen membrane in various doses and the following measured: (i) daily drug release into an aqueous solution, (ii) minimum inhibitory concentration (MIC) of the drugs against periodontopathogens, (iii) the effect of the drugs on mechanical properties of the membrane, and (iv) degradation by bacterial collagenase. In addition, the effects of the drugs on in-vitro cellular response was assessed by measuring blastogenesis of mononuclear cells obtained from patients suffering from periodontal disease and age/sex matched controls following incubation with the periodontopathogen Actinobacillus actinomycetemcomitans (AaY4). It was found that the collagen membranes released high levels of the drugs, at concentrations well above the MIC values. The mechanical properties of the membranes were not affected by the addition of the drugs, although resistance to the collagenases were. The cellular immune response was likewise suppressed in both patient and controls at drug doses comparable with the in-vitro drug release patterns. It is concluded that incorporation of antimicrobial drugs in a collagen barrier membrane may be of value when used in guided tissue regeneration.

Millipore filters,2 and Gortex Teflon barriers.315 Resorbable materials include polylactic acid membrane6 and collagen.7’8 With resor’bable materials such as collagen any exposed portion would presumably be subjected to enzymic degradation. However, there is a risk also of contamination with potentially pathogenic plaque flora. Et was the ultimate intention of this study to choose one of three antimicrobial agents to add to human collagen membrane for use in guided tissue regeneration. The three antimicrobial agents examined were niridazole, metronidazole and tinidazole. Niridazole (Ciba Geigy, Basel, Switzerland) (l(5-Nitrothiazole-2-)imidazolidine-2-one) is an antischistosomal agent, with known serious side-effects. It has good antimicrobial properties against periodontopathic bacteria’ and prolongs graft survival in man and experimental animals. lo Metronidazole

INTRODUCTION Chronic periodontitis is an inflammatory condition which leads tcs slow destruction of the tissues supporting the teeth. The ultimate goal of periodontal therapy is the predictable regeneration of lost periodontium. This may be facilitated by allowing repopulation with periodontal ligament cells and preventing the colonisation of the exposed root surface with gingival and bone cells.’ Recently, the technique of guided tissue regeneration has been used in surgical therapy of advanced periodontal disease with this aim in mind.2-4 A variety of materials have been used to date: these have included both non-resorbable and resorbable materials. A non-resorbable technique has been used based on the placement of materials such as * To

whom correspondence

should be addressed.

169 Clinrcal Materials

0267-6605/93/$6.00tf]1993

Elsevier Science Publishers

Ltd, England

E. Al Arrayed, S. Thomas, J. Moran!

170

(May and Baker Ltd, UK) (1,2,hydroxy ethyl-l?methyl-5-nitroimidazole) is an antimicrobial agent against periodontopathic with good activity bacteria.““2 It is also capable of suppressing schistosoma mansonia granuloma formation in miceI and haemagglutination of sheep red blood cells following injection of SRB antigen in rats administered orally with the drug.‘4 Tinidazole (Pfizer, Sandwich, Kent, UK) (I-(2-(ethylsolphonyl) also possesses ethyl)-2-methyl-5-nitroimidazole) activity against periodontopathic bacteria and has been shown to affect the cellular immune 15 response. The amount of each drug to be added to the collagen was determined by measuring the minimum inhibitory concentrations (MIC) of each compound against three selected microorganisms thought to be of an aetiological importance in periodontal disease. The pattern of release into water of different admixtures of the drugs to collagen was also determined: from these two sets of results, the optimum ratio of drug to collagen was chosen and used in the subsequent studies. Some information as to the physical properties of the drug--collagen membranes was also determined. Finally, the drugs were also tested for their effect upon the in-vitro lymphocyte responses to Actinobacillus actinomycetemcomitans (Aa) strain a measure of the effect of the cellular immune defence of the grafted host site against a typical periodontopathogenic organism. Since metronidazole is most widely used in the treatment and study of periodontal disease, the mechanism of action of any specific immunosuppression due to metronidazole was also studied through modification of the experimental transformation assay.

MATERIAL

previously. I6 added to the collagen, by saturating t at ratios of collagen 1 : 0.25 w/w and left to dry for 24 h, followed by u-radiation Minimum i~~ibit~~y ~~nce~tratio ~~~~~~~r~~~~~~~~~i~~iv~~~,~~ ( ~~~~~~~Q~~~~.~) ~~~~~~~e~~~.~~ and tevium tested for ~~~~~e~t~~ WCX-& s~sce~t~bi~~t~ to the a~~~~icr~b~~l agents as follows Ten samples of drug-collagen (I 0 x 5 x 2 mm) were placed separately in test tubes containing 3 ml of distilled water. After 24l-1: 2 md of each pernatant was collected an ncorporated mto ilkins Chalgren agar (Ii8 ml) fco Lid) In daub hng dilutions. Colonaes of the test ~r~~~l~s~s~ prewn for 48 h in fastidious anaerobrc en-rented with bloo , were collected and buBTered sahne ( ), The lx was adjusted

to

hkX-

~5 turbidity

standard. taining pla.tes were inocufated with the suspension of each organism usin a multipoint inoculator (Denley) and m.~bate for 96h in a.83 anaerobic cabinet ( I. plate containing no antimicro was also prepared. The IC of each strain was ntration a~~~~~~,~ no taken as the lowest co grow.th, one colony or a Pine yisi

Sets of treated

collagen sponges (5 x !O x 2 mm, laced, separately into 5 ml test of d,~st~~~edwater, sealed, and

AND METHODS

Preparation of antimicrobial solutions Stock solutions (1%) of metronidazole, niridazole and tinidazole, were prepared by dissolving each drug in distilled water. Niridazole was firstly dissolved in a small amount of N, N-dimethylformamide and the solution made up to 1% with distilled water. Preparation of drug-collagen

membranes

Freeze-dried cross-linked type 1 collagen pared from human placenta as described

was prein detail

recess was repeate water were examin

of drug-corsages

untreated collagen graft examined as described previously.17 Ten stri collagen sponge (40 x IO mm), previously i led water, were examined Bn an

es

Antimicrobial immunosuppression Tester model 1185 at lO%/min strain rate at 22°C. The gauge length was 20mm and crosshead speed 2 mm/min. Stress-strain curves obtained from this analysis were used to calculate Young’s modulus in both low and high strain regions. This ratio of stress (force/cross-sectional area) to strain (final length - initial length/initial length) was obtained from the slope of the stress-strain curve. Collagenase degradation assay The enzymic biodegradability of the collagen graft, following addition of the drugs was assessed by immersion in bacterial collagenase solution (Sigma Chemical Company, St. Louis, MO, USA). Samples of collagen membrane (1 mg) were added to 2 ml of stock buffer solution (1OmM Tris, and 25m~ CaCl,, pH 7.4) containing 200 units/ml of collagenase and incubated at 37 “C. The samples were checked for degradation every 15 min and the times recorded when the collagen buffer solution became clear. Effect upon in-vitro lymphoblastogenesis Subjects were selected from patients and staff of the Dental Hospital, Cardiff, UK. Fifteen patients, mean age 38.9 f 9.5 (5 male, 10 female) suffering from moderate to severe periodontal disease (PI > 1) were chosen, as were 15 age and sex matched control subjects (age mean 40 f 12 years). Participants in neither group were medically compromised. Separation qfmononuclear cells (MNCs) A 40ml sample of heparinised blood was obtained from each subject and diluted 2: 1 with PBS, layered over FicolllHypaque separation medium (Flow Laboratories, UK) and centrifuged at 400g for 30min. The interface cell fractions were collected, washed three times in PBS and resuspended in 5 ml of minimum essential medium (MEM, Flow Laboratories, Irvine, UK) containing 10% foetal calf serum (FCS, Flow Laboratories) at a final concentration of 1O6 cells/ml Preparation of antimicrobial suspensions From the stock solutions, all the antimicrobial agents were diluted to 12.5, 25, and 50pg/ml with distilled water. Preparation of AaY4 AaY4 (NCTC 305, Central

Public Health

Labora-

171

tory, Colindale, London) was grown in fastidious anaerobic broth, washed three times in PBS and centrifuged at 400g for 3Omin, ultrasonicated in distilled water at 4 “C for 5 min and washed again. The protein concentration was estimated according to the Lowry procedure using a protein assay kit (Sigma Chemical Company). The cell-free supernatant, at concentrations of @5pg and 5 pg protein/ml was added to the lymphocyte suspensions as described below. Lymphoblastogenic assay Suspensions (100 ~1) of MNCs ( lo4 cells/well) were cultured in triplicate in round-bottomed 96 well culture plates (Sterilin, Teddington, UK) for 5 days at 37 “C with 50~1 of the Aa supernatant, containing either O-5 or 5pg/protein, alone or together with metronidazole, tinidazole or niridazole (12.5, 25, 50pg/ml). Control wells included Aa alone, metronidazole, tinidazole, or niridazole alone: 0.4 and 4 pg/ml phytohaemagglutinin (PHA) (Sigma Chemical Company) was added as a positive control. Tritiated thymidine (Amersham International, Amersham, UK) 0.5 &i in 50~1 MEM was added to each well 8 h before termination of the culture. The cells were harvested on glass-fibre filter discs in a multiple automated harvester (Dynatech, Sussex, UK), washed three times with PBS, and 5% trichloracetic acid and finally washed in ethanol. The discs were dried and placed in scintillation fluid and counted in an Inter-technique Beta Counter. Counts were corrected for quenching and the stimulation index (SI) calculated as the ratio of counts per minute with additive to counts per minute with medium. Mechanism of action of metronidazole Addition of Aa( Y4) for spec$c periods to MNCs MNCs were cultured as describ’ed above and AaY4 (5 pg protein/ml) added to separate triplicate cultures for 6 h, 24 h and 5 days. At 6 and 24 h supernatants were removed and replaced with MEM plus 10% FCS containing 50 pg/ml metronidazole and reincubated for 5 days. Blastogenie responses were assessed after 5 days as described above. Addition of metronidazole suspension ,for specljic periods to MNCs MNCs were cultured with metronidazole for 6 and 24 h in triplicate. Similarly, MNCs were

172

F. Al

Armyed, 5 Thomas,

cultured with metronidazole, both with and without Aa for 5 days. In the case of the 6 and 24h cultures, supernatants were removed and replaced with MEM containing 10% FCS and 5 pg protein/ml Aa(Y4). At the same time, the supernatants of MNCs which were cultured with 50pg/ml metronidazole and 5 pg protein/ml Aa were removed at 6 and 24h and added to fresh cultures of MNCs with MEM containing 10% FCS. All cultures were incubated at 37 “C fsr 5 days. Blastogenic responses were assessed as described above.

MICS The MICs for the thr isms are shown !n $a niridazole, tinidazole,

inhibitory

concentrations

Strain (NCTC)

rugs against the test organ1I The order of efhcacy was

MBkigen

Release 0

Depletion of adherent cells MNCs were obtained by separation as described above, and adjusted to a concentration of 4 x IO6 cells/ml in MEM containing 20% FCS; 4 ml of this cell suspension was added to polystyrene tissue culture plates and incubated for 1 h at 37 “C. After the non-adherent cells the incubation period, were transferred to another plate and reincubated for a further hour. The total non-adherent cells were collected and the plates washed with MEM three times, the cells pooled, washed, resuspended in medium and divided into two equal aliquots each containing 1 x lo6 cells/ml. After removal of non-adherent cells, 4ml of cold MEM was added to each plate and incubated for an additional 30min at 4 “C. The adherent cells were gently harvested, washed and resuspended in MEM at 0.5 x lo6 cells/ml and added to aliquots of nonadherent cells in the ratio 1 : 8 of adherent cells to non-adherent cells. Blastogenic responses to Aa, Aa plus metronidazole or metronidazole alone were assessed with, without and with added back adherent cells, as described above. Table 1. Minimum

J. Moran

I.

oral pattern of ~~~~~~~~ into e basis of the amount released m the largest quantity, over the longest period of time the ranking order was niridazole, ~~t~~~~~~~~~~, and tinidazole. A ratio of met~~~idazo~e to collagen of 1 : 1 was perties and co!chosen for use in the mechanica entration which lagenase studies as the highest might not prove to be det~~rne~~a~ to the collagen sponge. s ertres of the WEthe addition of

and tinidazole treated the elastic ~~d~~~s of

bane

for metronidazole,

Metvonidazole

Actinobacillus actinomycetemcomitans (305)

0.25

16.0

were the samc as

Table 3 shows the effect of the addition of the three drugs upon the resistance to enzymic degradation by bacterial collagenase. tinidazole,

and niridazoie

TinidaZol4~

NidllZ&

_____~_-____.___~__ Bacteroides intermedius (9336)

es

0.25;

043

24

0.5

Fusobucterium nucLatum (10562)

0.03

0.03

04W

Bacteroides gingivulis (33277)

WI08

0.008

< 0~0005

(&ni)

Antimicrobial

immunosuppression

173

Conc.vg/ml 25Q 1

1:l

-

+

0.5:l

x

0.25:1

200

50

0 0

1

2

3 Time

(a)

4

5

6

7

5

6

7

(days)

400

0

1

2

3 Time

@I

4 (days)

300

-

I

I:1

+-

0.5:1

*

0.25:l

250 -

200 -

150-

IOQ-

50 -

0 0

1

2

(c) Fig. 1. The pattern of release of (a) niridazole,

3 Time

4 (days)

(b) metronidazole,

*

*

5

6

and (c) tinidazole

into water during

x 7

the 7-day study.

F. Al Arrayed,

174 Table 2. Determination

of mechanical Untreated collagen

Samples

Length (mm) Width (mm) Thickness (mm) Extension (mm) Maximum load (N) Stress (N) Strain (mm2) Elastic modulus (g/mm2)

S. Thomas, J. Moran properties

of a~t~~~crobia~-~r~ate~ -__________.__

Metronidaz&treated collagen

20 10 0.702 444 041 0.058 0.22 27.07

collagen -.__

Nvidazoietreated collagen

20 IO 0.58 5.2 0.41 0.07 1 0.26 27.74

n?idazoie~ Ireafeti coilngen

20 io 0.59 242 0.38 OG4 3.21 3 i “08

20 20 0.59 5.2 0.4 I !HK9 526

2: 27 .~I--___.-----

---.___Table 3. Degradation Time

15min 6h 8h 24h 48h 72h

of antimicrobial-treated

Collagen alone

collagen by collagenase ____~~

Collagen

Unchanged Small fragments Turbid Turbid Clear Clear

treated with

Metvonidazole

Tzrlidm~ie ______-.

Unchanged Unchanged Unchanged Small fragments Turbid Turbid

Unchanged Unchanged Unchanged Turbid Turbid Turbid

-___

Niridazate

Unchanged Unchanged Unchanged Turbid Turbid Turbid ._ ___--_

to produce the greatest to degradation.

prolongation

of resistance

hen the a~~~rn~cr~b~a~drugs were adde sion of response

sion was always pro-

Effect on lymphocyte blastogenesis Figures 2(a)-(c) show the mean iSI3 of the peak lymphoblastic responses in the 15 patients and 15 controls to AaY4 (0.5 or 5 pg protein/ml), AaY4 + metronidazole, AaY4 + tinidazole, or AaY4 + niridazole, all at concentrations of 12.5, 25 and 50pg/ml. Both patients and controls responded to AaY4 at levels which were higher than those generally accepted as representing stimulation.‘8 There was no significant different (P > 0.5) in the responses between the two groups. The peak response was achieved with the higher concentration of AaY4 (5pg/ml) in all but two patients who responded maximally to the 0.5 pg/ml concentration of Aa. In the control group the maximum response was always with the 5pg protein/ml concentration of Aa. When the antimicrobial agents were added to the Aa-stimulated MNCs the responses were significantly depressed (P < 0.001) at all the -doses of each of the three drugs examined. There appeared to be a dose-response relationship with the doses examined for each of the drugs in which larger doses produced greater depression of response.

with

Bon of Aa and the drugs. Additiovm of A 62,jbr The lym~b~b~astic

occurred

ec$c ~~~~~~S to response to AaY4

in all cases. less marked

When metr~~~d~zo~e

when

was added for 24h or for 5

hen metron~~az~~e was added hen supernatants from et~o~~d~z~~~ for 6

natants,

but not wit

175

Antimicrobial immunosuppression Peak

S.I.

6

Ria -

5

Aa(O.Sug Aa(5ug

protein/ml) prUiein/ml)

Aa(5ug/ml)*N(12.5ug) Aa(Sug/ml)

+ N(25Vg)

Aa(Sug/ml)

+ NW&Jug)

N (12.5ug) N (25ug) N (5gug)

Patients

NIRIDAZOLE

Controls

NIRIDAZOLE

(4 Peak S.I. 61

(b) Fig. 2a. Lymphoblastic

responses

of (a) patients

and (b) controls to various concentrations materials alone (mean +CSD).

Eflect of adherent cells on lymphoblastogenesis Lymphoblastogenic response to Aa, Aa + metronidazole and Aa + metronidazole with, without and with added back adherent cells in culture are shown in Fig. 4. The lymphoblastic response to Aa was enhanced where adherent cells had not been removed or where they had been added back to the MNCs. This response was significantly reduced (P < 0.05) when the adherent cells were separated from the MNCs. The mean values of the SIs observed with Aa in the presence and absence of, and with added back adherent cells were 4.9 f 0.57, 1.28 f O-15, and 4.24 f 0.32, respectively. 14

of Aa and niridazole

and the two

DISCUSSZON Collagen has been successfully used on previous occasions for the delivery of drugs, maintaining an effective level of delivery over a considerable period agent of time’9m2’ and inclusion of an antimicrobial in the collagen membrane would, therefore, seem to offer a solution to the problem of potential infection. It seemed important, however, to consider the effect of the addition of such an antimicrobial agent upon the properties of the collagen membrane itself and upon the local defence mechanism. The three drugs studied for addition to the collagen membrane for use in guided tissue ECM12

176 Peak

.%I.

6

5

Patient (a)

Fig. 2b. Lymphoblastic

responses

of (a) patients and (b) controls tcs various concentrations alone (mean i SD).

regeneration are all effective against anaerobic infecorganisms 9,22 that may cause post-operative tion. These drugs were also capable of down regulating the immune response to a range of antigens.t5 In this study, the release of the antimicrobial agents from the collagen membrane led to levels of drug which were inhibitory for the test organisms for 24448 h. In addition, all three drugs reduced the cellular immune response to the preparation of the test microorganism Actinobacillus actinomycetemcomitans (Aa). Since the mode of action of this immunosuppressive effect of the antimicrobial agents is unknown, it is not possible to extrapolate the findings to the healing surgical

wound.

owever,

ofAa

and tmidazoie and the two materials

it would

se

likeljj Shar cym-

for producing secondary immune cells. ~~~tia~

bacterial collagenase, drugs was an unexpe aration

of collagen

Antimicrobial immunosuppression

177

Peak S.I.

"1 Aa(OSug Aa(Sug

protein/ml) protein/ml)

Aa(Sug/ml)+M(lZ.Sug~ Aa(Svg/ml)

+ M(25vg)

Aa(Spg/ml)

??

M(50ug)

M t12.5ug) M G?5vg) M (5Oug)

Patients

METRONIDAZCXE

Peak S.I. 6 54-

3-

Controls

Fig. 2c. Lymphoblastic

responses

of (a) patients

METRONIDAZOLE

and (b) controls to various concentrations materials alone (mean f SD).

reduce the aldehyde fixation, a process which has potentially harmful effects.25126 The resistance to enzymic degradation was examined in this study with bacterial collagenase which degrades collagen in a different manner from mammalian collagenase.27 Although resistance to bacterial collagenase was found to be increased following addition of the three antimicrobial agents, there is only indirect evidence that resistance to collagenolysis, in vivo, by mammalian collagenase may be enhanced. Preliminary animal studies which examined the rate of resorption of collagen membrane with added metronidazole show an unusually long extension of the pattern of resorption of the metronidazole incorporated collagen compared to collagen a1one.28

of Aa and metronidazole

and the two

A similar effect is found with tetracycline acting either as a chelating agent or as an inhibitor of superoxide anion release by polymorphonuclear Rnhibition of colleucocytes.2g’30 The in-vitro lagenolysis exerted by the antimicrobial agents examined in this study may be enhanced in vivo by the apparent immunosuppressive effect of the drugs. Since metronidazole was to be used in the clinical trial, it was of interest to examine further the mode of action of the immunosuppression observed with the response to Aa. The exposure of the 5 day culture of the Aa-stimulated peripheral d MNCs to metronidazole for 6 h, 24 h an ys revealed that the major component of the immunosuppressive

F. Al Arrayed,

MNCs= Mononuclear cells WI= Metronidazole Aa= Actinobacillus Actinomycetemcomitans MEM= Minimum Essential Medium

(a)

(b)

S

A

Fig. 3. Lymphoblastic responses to (a) Aa 5pg protein/ml, and (b) metronidazole and conditioned medium of cultured MNCs + Aa + metronidazole added for specific times (6, 24 h and 5 days). effect

occurred

after

6 h and before

24 h (Fig. 3(b)) ~

In addition, transfer of the conditioned medium from these 6 and 24 h exposed cultures revealed that the major immunosuppressive activity had arisen in the medium after 6 h and before 24 h. After administration of metronidazole in the human, three major metabolites appear in the tissue, glycoconjugates, the hydroxmetabolite and the acetic acid metabolite.31 In-vitro studies with isolated hepatocytes32 suggest that the formation of the two oxidative metabolites is catalysed by dif-

Fig. 4. Lymphoblastic responses to Aa, Aa+metronidazole and metronidazole with, without and with added back adherent (A) cells.

1. Nyman, S., Gottlow, J., arrmg, T. & L&he, J., T regenerative potential of the periodontal ligament. A experimental study in monkeys. J. Cl%_ ~~~~~~Q~~~~., (1982) 257-65. 2. Nyman, S., Lindhe, J., Kar attacb~e~t following surgic ontal disease. J. cl&~. ~e~iod~~~o~~,9 (1982) 290~-6. 3. Gottlow, J., Nyman, S., Karring, T. Lindhe, J., New attachment formation as the result of controlled tissue regeneration. /. C/in. ~~~iQ~Q~~~~., I 4. Gottlow, J., Nyman, S., Lindhe, J., Marring, T. & acneStrom, J., New attachment formation in the human periodontium by guided tissue regeneration. J. C/in. Periodontoi., 13 (1986) 604m 16. 5. Magnusson, 1.; Nyman, S., Karring, T. 6% Egelberg, J.; Connective tissue attachment formation following cxcl~sion of gingival connective tissue and c~~~b~l~~rn during healing. J. Periodonl. Res., 20 (1985) 201-8. 6. Magnusson, I., Batich, C. & Collins, ment formation following controlled tissue regeneration using biodegradable membranes. J. ~~~~~~Q~~~~., (1988) l-6. 7. 1 Azar-Avidan, 0. c?L nts the apical migraperiodontal wound healing. J.

E., The use of Irradiated-cross-

8.

ne in guided tissue regeneration

J. Clin. Periodonl08.,

Antimicrobial 9. Wade, W. & hddy, M., Comparison of in vitro activity of niridazole, metronidazole, and tetracycline against subgingival bacteria in chronic periodontitis. J. Appl. BacterioL., 63 (1987) 455-7. IO. Salaman, J. R. & Miller, J. R., Non-specific chemical immunosuppression. Transplantation Proceedings, 11 (1979) 845550. 11. Loesche, W. J., Syed, S. A., Morrison, E. C., Laughon, B. & Grossman, N. S., Treatment of periodontal infections due to anaerobic bacteria with short term treatment with metronidazole. J. Clin. Periodontof., 8 (1981) 29-44. 12. Loesche, W. J., Syed, S. A., Morrison, E. C., Kerry, G. A., Higgens, T. & Stoll, J., Metronidazole in periodontitis. I. Clinical and bacteriological results after 15 to 30 weeks. J. Periodontal., 55 (1984) 325-35. 13. Grove, D. I., Mahmoud, A. A. F. & Warren, K. S., Suppression of cell mediated immunity by metronidazole. ht. Arch. Allergy Appl. Immunol., 54 (1977) 422-7. 14. Saxena, A., Chugh, S. & Vinayak, V. K., Modulation of Indian J. Med. host immune responses by metronidazole. Res., 81 (1985) 387-90. 15. Gonzalez, L., Frajman, M., Saenz, E., Boza, R. & Bolanos, H. B., Effect of tinidazole on the cellular immune response. /. Antimicroh. Chemother., 18 (1986) 499-502. 16. Quteish, D., Singh, G. & Dolby, A. E., Development and testing of a human collagen graft material. J. Biomed. Mater. Res., 24 (1990) 749-60. 17. Weadock, K., Olson, R. M. & Silver, F. H., Evaluation of collagen crosslinking techniques. Biometer. Med. Dev. Art. Org., ll(4) (1984) 293-318. 18. Lang, N. P. & Smith, F. N., Lymphocyte response to T-cell mitogen during experimental gingivitis in humans. Infect. Immurz., 15 (1976) 108813. 19. Minabe, M. et al. Application of local drug delivery system to periodontal therapy. 1. Development of the collagen preparations with immobilized tetracycline. J. Periodonto/., 60 (1989) 509 -66. 20. Minabe, M., Takeuchi, K., Tomomatsu, E., Hori, T. & Umemoto, T., Clinical effects of local application of collagen film immobilized tetracycline. J. Clin. Periodontol., 16 (1989) 291.-4. 21. Minabe, M., Takeuchi, K., Tamura, T., Hori, T. & Umemoto, T., Subgingival administration of tetracycline on a collagen film. b. Periodontol., 60 (1989) 55226. 22. Wust, J., Susceptibility of anaerobic bacteria to metronid-

immunosuppression

23.

24.

25.

26.

27. 28.

29.

30.

31.

32.

33.

34.

179

azole, ornidazole, and tinidazole and routine susceptibility Antimicrob. Agents testing by standardized methods. Chemother., 11 (1977) 631-7. Al-Arrayed, F., Thomas, S. & Dolby, A. E., Metronidazole inhibition of collagen implant induced mediators. jr. Dent. Res., 70 (1991) Abs 54, 675 McPherson, J. M., Ledger, P. W., Swamura, S., Conti, A., Wade, S., Reihanian, H. & Wallace, D. G., The preparation and physicochemical characterisation of injectible form of reconstituted glutaraldehyde crosslinked bovine corium collagen. J. Biomed. Mater. Res., 20 (1986) 79-92. Speer, D. P., Chvapil, M., Eskelson, C. D. & Ureich, J., Biological effects of residual glutaraldehyde in glutaraldehyde-tanned collagen biomaterials. J. Biomed. Mater. Res., 14 (1986) 753-64. Chvapil, M., Speer, D., Mora, W. & Eskelson, C., Effect of tanning agent on tissue reaction to tissue implant collagen sponge. J. Sung. Res., 35 (1983) 402--9. Perez-Tomayo, R., Pathology of collagen degradation. Am. J. Pathol., 92 (1978) 509-566. Al-Arrayed, F., Thomas, S. & Moran, J., Drug induced modification of host response to graft. Bit. Sot. Immunol. Meeting (1991), p. 42. Golub, L. M. et al. Further evidence that tetracycline inhibit collagenase activity in human cr’evicular fluid and from other mammalian sources. J’. Periodont. Res., 20 (1985) 12.-23. Golub, L. M., McNamara, T. F., Angelo, G. D., Greenwald, R. A. & Ramamurthy, N. S., A non-antibacterial chemically modified tetracycline inhibits mammalian collagenase activity. J. Dent. Res., 66 (1987) 1310~14. Stambaugh, J. E., Leo, F. G. & Manthe, R. W., The isolation and identification of the urinary oxidative metabolite of metronidazole in man. J. Pharm. Exper. Ther., 161 (1968) 373~-81. Loft, S. & Poulsen, H. E., Metabolism of metronidazole and antipyrime in isolated rate hepatocytes. Biochem. Pharmacol., 38 (1989) 1125S36. Stashenko, P., Regulatory effect of monocytes on T cell proliferation responses to oral microbial antigens. Infect. Zmmun., 38 (1982) 938-47. Amer, A., Singh, G., Darke, C. Bc Dolby, A. E., Cellular immunity to collagen in periodontal disease: role of T, B lymphocytes and adherent cells. J. Clip. Lab. Immunol., 24 (1987) 183.-7.