Effect of in vitro passage of healthy human gingival fibroblasts on cellular morphology and cytokine expression

Archs oral Biol. Vol. 41, No. 3, pp. 263 270, 1996 Copyright © 1996 Elsevier Science Ltd. All rights reserved

Pergamon

Printed in Great Britain 0003-9969/96 $15.00 + 0.00

0003-9969(95)00127-1

EFFECt[" OF IN VITRO PASSAGE OF H E A L T H Y H U M A N G I N G I V A L FIBROBLASTS ON C E L L U L A R M O R P H O L O G Y A N D C Y T O K I N E EXPRESSION L E I G H W. K E N T , RENI~E A. D Y K E N , F I R O Z A N T H O N Y C. A L L I S O N * and S U Z A N N E M. Departments of Microbiology and Oral Biology, School of Dentistry, Birmirtgham, 845 South 19th Street BBRB 258, Birmingham,

RAHEMTULLA, MICHALEK? The University of Alabama at AL 35294-2170, U.S.A.

(Accepted 24 October 1995)

Summary--Cytokines have been implicated in the regulation of antibody and inflammatory responses, but their role in periodontal diseases has not been elucidated. In the present study, cytokine production by human gingiw.1 fibroblasts (HGF) following in vitro passage was assessed in order to determine the basal levels of cytokine message and protein and to determine if the cellular morphology and the profile of cytokines procLuced differed with passage. The HGF cell line F-CL was established by explantation from non-inflamed gingival tissue, and cells from passages 1-10 were studied. The number of cells was determined in ,zonfluent cultures and cell morphology was examined by light microscopy. Fibroblasts from confluent cultnres were examined for cytokine mRNA by reverse transcription-polymerase chain reaction and culture sapernatants were assessed for cytokines by enzyme-linked immunosorbent assay. The morphology of F-CL fibroblasts in passages 1 4 was normal, while fibroblasts in passages 5-10 were larger. In general, the number of cells decreased from early to late passage. Fibroblasts from passages 1 I0 contained message for interleukin-lfl, -6 and -8, but not for interleukin-lc¢ or tumour necrosis factor-c~. Interleukin-6 was detected in culture supernatants of F-CL fibroblasts by the enzyme immunoassay and its level decreased with increasing passage. Copyright © 1996 Elsevier Science Ltd. Key words: cytokines, human gingival fibroblasts, cell passage, morphology.

liferation, and the induction of IL-6 and IL-8 (Oppenheim et at., 1991). IL-I and T N F - ~ stimulate bone resorption (Bertolini et al., 1986; Dewhirst et al., 1985; Gowen et al., 1983; Stashenko et al., 1987a) and inhibit bone formation in vivo (Konig, Muhlbauer and Fleisch, 1988; Nguyen et al., 1991; Sabatini et al., 1988) and in vitro (Bertolini et al., 1986; Stashenko et al., 1987b). There is now considerable interest in determining the role of cytokines in the remodelling of diseased periodontal tissues, but little information is currently available. IL-I~, IL-1/3 and T N F - ~ has been demonstrated in frozen sections of gingival tissue, in tissue extracts (Stashenko et al., 1991) and in crevicular samples obtained from inflamed periodontal sites (Masada et al., 1990; Rossomando, Kennedy and Hadjimichael, 1990). Mononuclear cells from inflamed human gingiva expressed IL-I, -5, -6 and -8, and T N F - ~ message (Fujihashi et al., 1993; Matsuki, Y a m a m o t o and Hara, 1992). Gingival fibroblasts, which are resident cells of the periodontal tissue, are responsible for synthesizing and secreting components of the extracellular matrix and play an important part in the remodelling of connective tissues during periodontal inflammation. These cells may also play an important part in the progression of periodontal

INTRODUCTION

Cytokines are protein hormones that mediate and regulate immune, as well as inflammatory, responses. They are often made by, and act on, many cell types in an autocrine or paracrine fashion. These molecules are believed to influence the events involved in wound healing and tissue repair. IL-1, IL-6, and T N F - ~ are involved in the acute inflammatory response (Akira et al., 1990; Dinarello, 1991). They are associated with endogenous pyrogen fever, synthesis of acutephase proteins by tlae liver, increased vascular permeability, and T- and B-cell activation. IL-6 is also involved in platelet production and in the terminal differentiation of B cells into immunoglobulin-secreting plasma cells, whereas IL-1 and T N F - ~ are involved in the increased expression of adhesion molecules on vascular endothelium, fibroblast pro-

*Current address: 2513 Hastings Drive, BelMont, CA 94002, U.S.A. tTo whom all correspondence should be addressed. Abbreviations: BSA, b,~vine serum albumin; DMEM, Dulbeceo's modified Eagle's medium; ELISA, enzymelinked immunosorbent assay; FCS, fetal calf serum; IL, interleukin; PBS, phosphate-buffered saline; RT-PCR, reverse transcription-polymerase chain reaction; TNF, tumour necrosis factor. AOB 41~3 B

263

264

Leigh W. Kent et al.

disease. Inflammatory mediators, especially IL-1 and TNF-~, stimulate gingival fibroblasts to synthesize and secrete collagenase and other metalloproteinases, which leads to the destruction of the extracellular matrix of the periodontal tissues (Birkedal-Hansen, 1993). Transcription of metalloproteinase genes is activated by IL-1, TNF-~, platelet-derived growth factor, and epidermal growth factor, and activation is suppressed by prostaglandin E~ and transforming growth factor-//(Page, 1991). Fibroblasts have also been shown to secrete cytokines in response to stimulation by microbial components, such as lipopolysaccharide, or by other cytokines (Kumar, Millis and Baglioni, 1992; Takashiba et al., 1992; Yamazaki et al., 1992). When stimulated with lipopolysaccharide from the periodontopathogen Baeteroides, cultures of human gingival fibroblasts secrete IL-I/~, IL-6, and cellassociated IL-I~ (Takada et al., 1991). IL-8 is also produced by fibroblasts when stimulated by IL-1/~, but not by Escherichia coli lipopolysaccharide (Odake et al., 1993; Takashiba et al., 1992). Most studies on cytokine production by gingival tissue have used cells obtained from several different sources and the steady-state profile of the mRNA and secreted protein has not yet been defined for normal, healthy tissue. Our objective now was to determine the effect of in vitro passaging on cellular morphology and on cytokine production by a human gingival fibroblast line cultured from healthy gingiva. Here we report on cytokine production at the levels of mRNA and protein, and the effect of in vitro cell passage on changes in the expression of cytokines by this cell line.

MATERIALS

AND

~< <~

~ < ~

[..

e~

=

8

METHODS

Tissue culture

A human gingival fibroblast line was established by explantation from the wedge of tissue distal to a non-periodontally involved mandibular molar of a donor undergoing surgery. Clinically, the tissue appeared firm and was not erythematous, oedematous, or bleeding. The method used for the culture of gingival fibroblasts was reported by Larjava, H/ikkinen and Rahemtulla (1992). In brief, tissue biopsies were cut into small pieces using a sterile technique under a laminar-flow hood. The pieces were allowed to attach to the walls of plastic culture flasks (Falcon, Oxnard, CA) containing DMEM (Gibco Chemical Co., Grand Island, NY) supplemented with sodium pyruvate (0.1g/I), L-glutamine (1.16g/I), streptomycin sulphate (100mg/l), penicillin (100,000U/l), and 10% FCS. The medium was tested for endotoxin with the limulus amoebocyte lysate as described by the manufacturer (Bio,Whittaker, Walkersville, MD), except that half the volume of assay reagents was used.

~<<<<<

~.~ 0

Cytokines produced by gingival fibroblasts Fibroblast subcultures were prepared from the primary cultures by removing the spent medium, washing the tissue fragments several times with sterile PBS, and dissociating the fibroblasts from the outgrowth by treatment with 0.25% trypsin, 0.1 mM EDTA in CA2+/MgZ+-free Hanks' solution (Gibco) at 37°C for 3-5 min. The cells were dislodged from the flasks by gentle tapping and the enzyme activity was stopped by adding D M E M containing 10% FCS. The dissociated cells were collected by gently pipetting the culture supernatant and cells were harvested by centrifugation. For each passage, the cells were resuspended in D M E M medium containing 10% FCS and seeded into plastic dishes at a ratio of 1:4. When the subcultures reached confluence, serial passaging was done by trypsinization as described above. Cells from 10 consecutive passages (1-10) were used in the present study. Cell count and photography

Cell cultures from passages 1-9 were photographed after fixing the cells in 3.7% formaldehyde and staining with 0.1% crystal violet in boric acid (Kueng, Silber and Eppenberger, 1989). The cells from passages 1-10 were counted in a haemocytometer after trypsinization, and their viability was assessed by trypan blue exclusion (Phillips, 1973).

265

Table 2. Effect of in vitro cell passage on the viability and number of fibroblasts in culture Passage no. 1 2 3 4 5 6 7 8 9 10

% viable cells

Cell count x 104/ml

99.2 94.2 97.2 97.5 95.0 97.0 97.4 86.0 92.6 87.2

138 120 108 118 81 104 41 29 27 24

thimerosal in PBS) added. Plates were then incubated for 1 h at room temperature and washed four times with ELISA wash buffer. The amount of colour that developed after adding substrate (100/~l/well) consisting of I mg/ml o-phenylenediamine, 0.03% H202, and 0.2M citrate buffer (pH 4.9) was recorded in a Vmax microplate reader (Molecular Devices Corp., Menlo Park, CA) at dual wavelength: 450nm (sample filter)/630nm (reference filter). The amount of cytokine in each sample, run simultaneously with the standard, was determined by interpolation from the standard curve using a four-parameter logistic algorithm (Softmax, Molecular Devices Corp.).

Enzyme-linked imm~nosorbent assay

The ELISA used for the quantification of IL-I~, -lfl, -6 and TNF-~ involved the use of two monoclonal antibodies wilh specificity to different epitopes on each cytokine molecule of interest, which were developed at Syntex Laboratory (Palo Alto, CA). The immunoassays are specific (Kenney et al., 1987; Kenney, Masada and Allison, 1990). The assay standards were recombinant human IL-lc~, -lfl, -6, and TNF-~t and were purchased from R&D Systems, Inc. (Minneapolis, MN). Flexible, polyvinylchloride, 96-well microtitre plates (Dynatech Laboratories, Inc., Chantilly, VA) were coated with the appropriate first antibody diluted in PBS (pH 7.2; 100 #l/well). Plates were incubated overnight at 4°C, and then were washed twice with ELISA wash buffer (0.1% BSA and 0.05% thimerosal in PBS). This wash buffer was used after each step. Non-specific binding sites were blocked by incubating the plates with 5% non-fat dry milk and 0.05% thimerosal in PBS (200 pl/well) for 1.'; h at room temperature. Plates were washed three times with ELISA wash buffer and then twofold serial dilutions of the standard or sample were added to the wells (100#1) in duplicate. Plates were incubated for 2 h at room temperature, washed, and the appropriate biotinylated monoclonal antibody then added (50#1). After 2-h incubation at room temperature, plates were washed and 100/~1 of peroxidase-streptavidin (ZYMED Laboratories, Inc., San Francisco, CA) diluted 1:3000 in BSA buffer (1% BSA and 0.05%

Reverse transcription and polymerase chain reaction

Confluent fibroblast cultures from passages 1-10 were used for the detection of cytokine mRNA. Total R N A was extracted from the cultures by the guanidinium isothiocyanate-phenol-chloroform method (Chomczynski and Sacchi, 1987). In brief, 1 ml of 4 M guanidinium isothiocyanate in 25 mM sodium citrate buffer, pH 7.0, containing 0.5% sacrosyl and 0.1 M fl-mercaptoethanol was added to the cells and the fibroblast monolayer was scrapped using a rubber 'policeman'. The solution was divided into 0.4-ml portions and passed through a 21-gauge needle five times to shear the DNA. To the 0.4-ml portions, 40/tl of 2 M sodium acetate (pH4.0), 80/~1 chloroform:isoamyl alcohol (24:1), and 400#1 of watersaturated phenol were added sequentially, the mixture vortexed, and then incubated for 10 min on ice. Samples were then centrifuged at 7000g for 20 min at 4°C. The aqueous phase containing the R N A was transferred to a new tube, taking care not to disturb the phenolic interface containing proteins and DNA. To the aqueous phase containing the RNA, 0.4 ml of isopropanol was added and the samples were kept at - 7 0 ° C for approx. 2 h to precipitate RNA, which was then harvested by centrifugation. The precipitate was washed three times with 80% ethanol and centrifuged to further purify the RNA. Finally, ethanol was removed from the samples by vacuum drying in a Speed Vac and the pellet was dissolved in 22/~1 of diethyl pyrocarbonate-treated distilled water (0.1% v/v)

266

LeighW. Kent et al. PASSAGE t

PASSAGE 5

PASSAGE 7

PASSAGE 8

Fig. 1. Human gingival fibroblast cultures stained with 0.1% crystal violet in boric acid taken by light microscopy at matching magnifications (x 85).

c o n t a i n i n g 0.3% R N A a s e inhibitor. R N A extracted in a similar m a n n e r from stimulated peripheral blood m o n o n u c l e a r cells was used as a positive control. Peripheral blood m o n o n u c l e a r cells were prepared from heparinized blood diluted with Dulbecco's PBS (1:2) a n d layered (10 ml) over 3 ml of Ficoll H y p a q u e no. 1077 in 15-ml conical tubes. After centrifugation (450 g ) for 20 min at r o o m temperature, the interface was collected a n d washed with 45 ml of PBS. The cells were pelleted by centrifugation (5 min, 2 5 0 g ) , the red cells lysed with lysing buffer ( 1 5 0 m M amm o n i u m chloride, 1 0 m M potassium bicarbonate, 0.13 m M E D T A ) , a n d the remaining m o n o n u c l e a r cells suspended in complete medium. Cells (106/ml) were cultured in the complete R P M I 1640 m e d i u m (Gibco) containing FCS, penicillin-streptomycin,

Table 3. Levels of IL-1, TNF-e and IL-6 in culture supernatants of F-CL fibroblasts after in vitro passage as determined by ELISA Cytokine level (pg/ml) Passage no. 1 2 3 4 5 6 7 9 l0

IL-I~

IL-1B

TNF-~

IL-6

0 156 0 96 180 0 0 0 0

0 0 0 0 414 0 0 0 0

269 323 238 0 0 172 0 272 327

5860 3185 ND 2056 2569 1784 1850 561 602

ND = not determined. 0, values were less than the limits of sensitivity.

Cytokines produced by gingival fibroblasts r-glutamine and gentamicin, and incubated in the presence of concanavalin A (Boehringer Mannheim, Mannheim, Germany) and phytohaemagglutinin (Wellcome Diagnostics, Dartford, U.K.) at 37°C in an atmosphere of 5% CO2 in air for 24 h. The cells

267

were harvested and washed, and then R N A extracted as described above. The method described by Brenner et al. (1989) was used to reverse transcribe the cellular m R N A . In brief, 1 - 2 / l g of R N A (10/~1) was first incubated at

IL-I(z (420 bp)

IL-I~ (802 bp)

I

If 1.6 Kb

1.6 Kb

1 Kb

1Kb

Q

500 bp

500 bp

1 2 3 4 5 6 7 8 910111213

1 2 3 4 5 6 7 8 910111213

IL-6 (628 bp)

IL-8 (298 bp)

1.6 Kb

1.6 Kb

1Kb

1Kb i i o e o m o O a

~O

500 bp

500 bp

1 2 : 3 4 5 6 7 8 9 1011 12 13

1 2 3 4 5 6 7 8 910111213 ~actin (1126 bp)

TNF-~t (695 bp)

1.6 Kb

1.6 Kb

1 Kb

1Kb 500 bp

500 bp

1 2 3 4 5 6 7 8 9 10 11 12 13

1 2

3 4

5 6 7 8 9 10 11 12 13

Fig. 2. Analysis of cytokine message in F-CL fibroblasts following in vitro culture and passage. Lane I, l-Kb DNA itadder; lanes 2-11; cytokine message in passages 1-10 of F-CL gingival fibroblasts, respectively; lane 12, cytokine message in stimulated peripheral blood mononuclear cells (positive control) and lane 13, sterile water (negative control).

268

Leigh W. Kent et al.

65°C for 2 min and then stored on ice. R N A was transcribed in a final volume of 100/~1 using RT master mixture consisting of 5 mM MgC12, 1 x PCR buffer II, 1 mM each of dATP, dGTP, dCTP, dTTP, 2.5 # M oligo [d(T)16], 1 U//.tl RNAsin, and 10 U//~I Moloney murine leukaemia virus reverse transcriptase. A negative control was prepared using 10 #1 of distilled water and 90/~1 of RT master mixture. Samples were incubated in a Thermal Cycler (Perkin Elmer 9600, Norwalk, CT) in the sequence: 42°C for 15 min, 37°C for 7 min, and 99°C for 5 min. Ten #1 of RT reaction product was added to 40 #1 of the PCR master mixture for a final concentration of 2 m M MgCI2, 1 × PCR buffer, 1.25 U/#I Ampli Taq polymerase, and 15 mM 3' and 5' primers. The specific primers used are listed in Table 1 and were prepared at the University of Alabama Oligonucle0tide Facility (Lagoo-Deenadayalan et al., 1993). All reagents used for RT reaction were obtained from Perkin Elmer Cetus unless otherwise stated. The samples were amplified by PCR in a D N A Thermal Cycler in the sequence: 95°C for 2 min; 35 cycles of a sequence consisting of 94°C for 45 s, 60°C for 2 min, and 72°C for 3 min; and finally 72°C for 7 min. Fifteen microlitres of each resulting reaction mixture was electrophoretically separated in 1.5% agarose gels using TBE buffer (0.089 M tris, 0.089 M boric acid, 0.0025 M EDTA, pH 8.0), stained with ethidium bromide, visualized with ultraviolet light, and photographed using a Polaroid land film. 8Actin was used as a control and transcribed first to rule out failure of the reaction. RESULTS

cells from all passages were morphologically fibroblasts (Fig. 1), their size drastically increased after passage 3. The number of cells decreased after passage 4 (Table 2), and a 2.5-fold reduction in cell number was observed when the counts from passages 6-7 were compared. All cell preparations were healthy, as shown by their granular cytoplasm and the absence of vacuoles, especially in the earlier passages. Some cells in passages 9 and 10 contained vacuoles. At higher magnification, the nuclei were round and regular, with no visible sign of senescence such as irregular shape and/or bimicronucleation.

Cytokine levels by ELISA The amount of cytokines obtained from cell passages 1-10 of the F-CL fibroblast line were determined by ELISA. The sensitivity of the ELISA used to measure each cytokine was shown to be 55.6-4500pg/ml for IL-l~t, 274.1-7400pg/ml for IL-lfl, 400-20,000pg/ml for IL-6, and 100400 pg/ml for TNF-~. Culture supernatant collected from each passage showed low or no detectable IL- I ct or IL-lfl (Table 3). IL-6 was detected in all passaged cells, and the amount of IL-6 produced decreased with passage.

Cytokine expression at the level of mRNA In order to confirm and extend the results obtained by ELISA, the total R N A was extracted from passages 1-10 and analysed by RT-PCR (Fig. 2). Normal, unstimulated, human gingival fibroblasts contained m R N A for IL-lfl, IL-6, and IL-8 in cells from all passages (Fig. 2). However, no IL-I~ or TNF-~ m R N A was detected, fl-Actin m R N A was demonstrated in all preparations.

Culture of gingival fibroblasts Within 3-5 days of incubation, outgrowth of fibroblasts was observed from the gingival explants and elongated cells were seen at the leading edge of the outgrowth. Within 20-21 days the cultures from the explants were 60-70% confluent and ready for subculture. Subcultured gingival fibroblasts divided rapidly and seeding at a 1:4 ratio resulted in confluency in the T-75 tissue-culture flasks within 10-15 days.

Cell counting and cell morphology Confluent, trypsinized F-CL gingival fibroblasts were harvested, mixed with trypan blue and counted in a haemocytometer. Viability ranged from 86 to 99%: cells in passages 8-10 had the lowest viability; the viability of cells in passages 1-7 was in excess of 90% (Table 2). The number of cells recovered declined from passages 1-10 (Table 2). There was a gradual decrease in cell number from 138 x 104 to 24 x l04. The reduction in cell number from passages 1-10 was approx. 80%. The reduction in cell number with passage corresponded with a change in morphology. Although

DISCUSSION

We have examined morphological differences and cytokine expression in a healthy gingival fibroblast cell line after passage in vitro. It is generally accepted that cells of 'fibroblastic' origin are bipolar or multipolar, with a length twice their width, but variations in the cell density, substrate and composition of the medium can affect morphology. We established the present fibroblast (F-CL) line from clinically healthy gingival tissue. In the F-CL line, cells from passages 1-4 were not significantly different from each other morphologically or in size, but both size and shape changed from passages 5-10. Cell size increased drastically; the number of cells decreased sharply. Pendergrass, Angello and Norwood (1989) observed that one of the most striking morphological changes with advancing levels of passage was an increase in cell volume. Their systematic measurement of cell volume in cultures over the entire lifespan showed that the mean volume increased exponentially with time, from an original volume of approx. 2000/~m 3 to a mean volume of 5000 # m 3 in senescent cells. They

269

Cytokines produced by gingival fibroblasts also observed a loss of proliferative activity, which is considered a sign of ageing. Their observations were made on fibroblast cultures that had gone through approx. 70 population doublings. We observed a significant increase irL cell size from passages 4 to 10, but found no significant loss of proliferative activity, indicating that the q~ultures were not senescent or aged. Several studies have shown that periodontal fibroblasts can produce cytokines when stimulated with either bacterial lipopolysaccharide or various cytokines (Bartold and Haynes, 1991; Shimzu et al., 1992; Yamazaki et al., 1992). We show that the normal human gingival fibroblast line (F-CL) expresses IL-6, IL-8, some IL-1/~, bat no IL-lct or TNF-~ mRNA, after its growth frem gingival explants, and that culture supernatants contain IL-6 and low but detectable IL-1 and TNF-~ (their amount was at the lower limits of detection by the ELISA). That TNF-ct and especially IL-1 were not detected in culture supernatants from all passages, and that TNF-ct or IL-lct message were not seen, suggest that these cytokines were induced early in the culturing. The presence of IL-6 me,;sage and protein in all passages suggests that there was constitutive production of this cytokine by these cells. Bartold and Haynes (1991) demonstrated that the culture supernatant of human gingival fibroblasts contained a significant amount of IL-6 activity; the presence of IL-6 in the fibroblasts was also demonstrated immunohistochemically. The F-CL cell line produced a high basal activity of IL-6, which gradually decreased in the culture supernatant after passage in vitro. The decreasing amount of IL-6 detected might be due to the lower number of cells present in the culture or to a decrease in IL-6 mRNA or IL-6 protein synthesis. These findings indicate that the reported levels of cytokines in supernatants should be normalized as a function of cell number or the quantity of the enzyme hexosaminidase. Recently, Goodman and Stein (1994), using human diploid fibroblasts, have reported a decrease in both basal and induced levels oJ? IL-6 mRNA and in the amount of IL-6 protein prodaced during cellular ageing. They suggest that the mechanism involved in the loss of IL-6 gene expression may involve the retinoblastoma protein. There are recent reports on factors that increase cytokine production by human gingival fibroblasts. IL-6 activity was stimulated above basal levels by FCS and IL-lfl (Bartold and Haynes, 1991); TNF
by stimulants such as IL-lfl. However, it is unlikely that stimulation by lipopolysaccharide occurred, as the culture medium used was negative for endotoxin activity by the limulus amoebocyte lysate assay. This study has provided information that gingival fibroblasts vary in size and number during subculture and that number decreases while cell size increases in later passages. Passaging may be another variable to consider when studying fibroblasts. This study has also demonstrated that cultures of normal human gingival fibroblasts express IL-lfl, -6, and -8, but not IL-lc~ or TNF-~, when cultured under standard conditions. These findings provide evidence for a biological role of these cytokines as contributing factors in the regulation of periodontal tissue in health and disease. Acknowledgements--We thank Frank Roberts for technical assistance and Vickie Barron for editorial assistance. This work was done by Leigh W. Kent in partial fulfilment of requirements for a Ph.D. from The University of Alabama at Birmingham and was supported by USPHS grants DE 00279, DE 08466, DE 08228, DE 09081, and DE 08182.

REFERENCES

Akira S., Hirano T., Taga T. and Kishimoto T. (1990) Biology of multifunctional cytokines: IL-6 and related molecules (IL-1 and TNF). F A S E B 4, 2860-2867. Bartold P. M. and Haynes D. R. (1991) Interleukin-6 production by human gingival fibroblasts. J. perio. Res. 26, 339-345. Bertolini D., Nedwin G., Bringman T., Smith D. and Mundy G. (1986) Stimulation of bone resorption and inhibition of bone formation in vitro by human tumor necrosis factors. Nature 319, 516-518. Birkedal-Hansen H. (1993) Role ofcytokines and inflammatory mediators in tissue destruction. J. perio. Res. 28, I 11. Brenner C. A., Tam A. W., Nelson P. A., Engleman E. G., Suzuki N., Fry K. E. and Larrick J. W. (1989) Message amplification phenotyping (MAPPing): a technique to simultaneously measure multiple mRNAs from small numbers of cells. BioTechniques 7, 1096-1103. Chomczynski P. and Sacchi N. (1987) Single-stepmethod of RNA isolation by acid guanidinium thiocyanate-phenolchloroform extraction. Analyt. Biochem. 162, 156-159. Dewhirst F., Stashenko P., Mole J. and Tsurumachi T. (1985) Purification and partial sequence of human osteoblast-activating factor: identity with interleukin 1B. J. Immunol. 135, 2562-2568. Dinarello C. (1991) Interleukin-1 and interleukin-1 antagonism. Blood 77, 1627-1652. Fujihashi K., Kono Y., Beagley K., Yamamoto M., McGhee J. R., MesteckyJ. and Kiyono H. (1993) Cytokines and periodontal disease: immunopathological role of interleukins for B-cellresponses in chronic inflamed gingival tissues. J. Periodont. 64, 400 406. Goodman L. and Stein G. (1994) Basal and induced amounts of interleukin-6 mRNA decline progressively with age in human fibroblasts. J. biol. Chem. 269, 19,250-19,255. Gowen M., Wood D. D., Ehrie E., McGuire M. and Russell R. (1983) An interleukin I-like factor stimulates bone resorption in vitro. Nature 3116, 378-380. Kenney J., Masada M., Eugui E., Delustro B., Mulkins M. and Allison A. (1987) Monoclonal antibodies to human

270

Leigh W. Kent et al.

recombinant interleukin 1 (IL 1)fi: quantitation of IL lfl and inhibition of biological activity. J. Immunol. 138, 4236-4242. Kenney J. S., Masada M. and Allison A. (1990) Laboratory Methods in Immunology (Ed. Zola H.), pp. 231-240. CRC Press, Boca Raton. Konig A., Muhlbauer R. and Fleisch H. (1988) Tumor necrosis factor ~ and interleukin 1 stimulate bone resorption in vivo as measured by urinary [3H]tetracycline excretion from prelabeled mice. J. Bone Min. Res. 6, 621 627. Kueng W., Silber E. and Eppenberger U. (1989) Quantification of cells cultured on 96-well plates. Analyt. Biochem. 182, 16-19. Kumar S., Millis A. and Baglioni C. (1992) Expression of interleukin l-inducible genes and production of interleukin 1 by aging fibroblasts. Proc. natn. Acad. Sci. U.S.A. 89, 4683 4687. Lagoo-Deenadayalan S., Lagoo A. S., Barber W. H. and Hardy K. J. (1993) A standardized approach to PCRbased semiquantitation of multiple cytokine gene transcripts from small cell samples. Lymphokine Cytokine Res. 12, 59 67. Larjava H., Hfikkinen L. and Rahemtulla F. (1992) A biochemical analysis of human periodontal tissue proteoglycans. Biochem. J. 284, 267-274. Masada M. P., Persson R., Kenney J. S., Lee S. W., Page R. C. and Allison A. C. (1990) Measurement of interleukin-lc~ and -lfl in gingival crevicular fluid: implications for the pathogenesis of periodontal disease. J. perio. Res. 25, 156 163. Matsuki Y., Yamamoto T. and Hara K. (1992) Detection of inflammatory cytokine messenger RNA (mRNA)expressing cells in human inflamed gingiva by combined in situ hybridization and immunohistochemistry. Immunology 76, 42-47. Nguyen L., Dewhirst F., Hauschka P. and Stashenko P. (1991) Interleukin lfl stimulates bone resorption and inhibits bone formation in vivo. Lymphokine Cytokine Res. 10, 15-21. Odake H., Koizumi F., Hatakeyama S., Furata I. and Nakagawa H. (1993) Production of cytokines belonging to the interleukin-8 family by human gingival fibroblasts stimulated with interleukin-1/~ in culture. Exp. molec. Pathol. 58, 14-24. Oppenheim J., Zachariae C., Mukaida M. and Matsushima K. (1991) Properties of the novel proinflammatory supergene 'intercrine' cytokine family. Ann. Rev. lmmunol. 9, 617 648.

Page R. C. (1991) The role of inflammatory mediators in the pathogenesis of periodontal disease. J. perio. Res. 26, 230 242. Pendergrass W., Angello J. and Norwood T. H. (1989) The relationship between cell size, the activity of DNA polymerase ~ and proliferative activity in human diploid fibroblast-like cell cultures. Exp. Gerontol. 24, 383 393. Phillips H. J. (1973) Dye Exclusion Test for Cell Viability: in Tissue Culture. Academic Press, New York. Rossomando E. F., Kennedy J. E. and Hadjimichael J. (1990) Tumor necrosis factor alpha in gingival crevicular fluid as a possible indicator of periodontal disease in humans. J. oral Biol. 35, 431 434. Sabatini M., Boyce B., Aufdemorte T., Bonewald L. and Mundy G. (1988) Infusions of recombinant interleukins 17 and lfl cause hypercalcemia in normal mice. Proc. natn. Acad. Sci. U.S.A. 85, 5235-5239. Shimizu N., Ogura N., Yamaguchi M., Goseki T., Shibata Y., Abiko Y., Iwasawa T. and Takiguchi H. (1992) Stimulation by interleukin-1 of interleukin-6 production by human periodontal ligament cells, Archs oral Biol. 37, 743 748. Stashenko P., Dewhirst F., Peros W., Kent R. and Ago J. (1987a) Synergistic interactions between interleukin 1, tumor necrosis factor, and lymphotoxin in bone resorption. J. lmmunol. 138, 1464-1468. Stashenko P., Dewhirst F., Rooney M., DesJardins L. and Heeley J. (1987b) Interleukin 1/~ is a potent inhibitor of bone resorption in vitro. J. Bone Min. Res. 2, 559-565. Stashenko P., Jandinski J. J., Fujiyoshi P., Rynar J. and Socransky S. S. (1991) Tissue levels of bone resorptive cytokines in periodontal disease. J. PeriodontoL 62, 504-509. Takeda H., Mihara J., Morisaki I. and Hamada S. (1991) Induction of interleukin-I and -6 in human gingival fibroblast cultures stimulated with Bacteroides lipopolysaccharides. Infect. Immun. 59, 295 301. Takashiba S., Takigawa M., Takahashi K. and Murayama Y. (1992) Interleukin-8 is a major neutrophil chemotactic factor derived from cultured human gingival fibroblasts stimulated with interleukin-lfl or tumor necrosis factor alpha. Infect. Immun. 60, 5253-5258. Yamazaki K., Ikarashi F., Aoyagi T., Takahashi K., Nakajima T., Hara K. and Seymour G. J. (1992) Direct and indirect effects of Porphyromonas gingivalis lipopolysaccharide on interleukin-6 production by human gingival fibroblasts. Oral Microbiol. Immunol. 7, 218 224.