Toxicology, 75 (1992) 145-157 Elsevier Scientific Publishers Ireland Ltd.

145

Effect of cigarette smoking on salivary epidermal growth factor (EGF) and EGF receptor in human buccal mucosa S.-L. Wang a, M. Milles b, C.-Y. Wu-Wang a, G. Mardirossian b, C. Leung c, A. Slomiany a and B.L. Slomiany a aResearch Center, bDepartment of Oral and Maxillofacial Surgery, New Jersey Dental School and CDepartment of Anatomy, New Jersey Medical School, University of Medicine and Dentistry of New Jersey, Newark, NJ 07103-2400 (USA) (Received March 5th, 1992; accepted June 22nd, 1992)

Summary The mouth acts as a primary target for ciga?ette smoke which is associated with several oral diseases and cancer. The present study investigated the effect of cigarette smoking on salivary EGF and the buccal EGF receptor. Samples of whole saliva and buccal biopsy were obtained from 15 healthy volunteers (10 smokers and 5 non-smokers). The smokers smoked 20 or more cigarettes/day for more than 5 years. Salivary cotinine (a major metabolite of nicotine) was determined by radioimmunoassay (RIA). The salivary cotinine level was consistent with the self-reported smoking status (smokers, 106-530 ng/ml saliva; non-smokers, < 2 ng/ml saliva). As compared to the non-smokers, the salivary EGF concentration (determined by RIA) was 32% lower in those smokers whose salivary cotinine level was 250 ng/ml or higher (non-smokers, 2.21 4. 0.16; smokers, 1.57 4- 0.09 ng/ml saliva; mean 4- S.E.M., P < 0.01). There was no significant difference in 125I-labeled EGF binding to the buccal receptor between the two groups. However, EGF stimulated the autophosphorylation of a 170-kDa protein band in the sample of nonsmokers, but not in the smokers. The immunoblot analysis using anti-EGF receptor antibody indicated that the smoking-related deficiency in EGF receptor autophosphorylation was due to the functional alteration of the receptor proteins. In conclusion, cigarette smoking reduces the salivary EGF level and impairs the function of buccal EGF receptor, which may be associated with the pathology of smokingrelated oral disease.

Key words: Epidermal growth factor; Receptor; Buccal mucosa; Smoking; Human

Introduction I t is well d o c u m e n t e d t h a t c i g a r e t t e s m o k i n g h a s a d v e r s e effect o n h u m a n h e a l t h [ 1 - 3 ] . T h e m o u t h a c t s as a p r i m a r y t a r g e t f o r t o b a c c o s m o k e . V a r i o u s s t u d i e s

Correspondence to." S.-L. Wang, UMDNJ-NJ Dental School, Research Center, University Heights, 110 Bergen Street, Newark, NJ 07103-2400, USA. 0300-483X/92/$05.00 © 1992 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland

146

reported that smoking is associated with periodontal disease [4,5], specific lip lesion, hyperkeratinisation, leukodema, leukoplakia and cancer of the oral mucosa [6-8]. The mechanism by which cigarette smoking causes these untoward effects in the oral cavity remains poorly understood. Epidermal growth factor (EGF) is a mitogenic polypeptide secreted primarily by the submandibular glands [9,10]. EGF provides cytoprotective function throughout the alimentary tract [11]. Although the oral cavity is the first segment of the alimentary tract and is exposed to a variety of physical, chemical and microbial insults, little is known about the function of EGF receptor in soft oral tissues. Nevertheless, it has been suggested that EGF may play a significant role in the protection of oral mucosal tissue [12-14]. The way by which EGF interacts with the target cells is via its binding to a specific high affinity receptor located on the cellular membrane [10]. EGF receptors are present in various tissues of the oral cavity [15-17]. Our previous studies characterized the biochemical properties of specific EGF receptors in both rat buccal mucosal cells [121 and human buccal membrane [13]. The biochemical features of EGF receptor are altered by various environmental chemicals [18-21]. Our recent studies [22] provided evidence that ethanol decreases EGF receptor binding, impairs EGF receptor autophosphorylation and reduces EGF-stimulated cellular function in rat buccal mucosa. The present study was conducted to investigate the effect of cigarette smoking on salivary EGF level and the EGF receptor in human buccal mucosa. Materials and methods

Materials Bovine serum albumin (BSA) fraction V, mouse submaxillary EGF, Nonidet P-40, phenylmethylsulfonylfluoride(PMSF), aprotinin, leupeptin, prestained protein standard marker and urethane were from Sigma Chemical Co., St. Louis, MO. EGF receptor antibody (Ab-2), raised from purified EGF receptor of A431 cells, was purchased from Oncogene Sci., Inc., Manhasset, NY. Chemicals for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), Western blot and the nitrocellulose paper were purchased from Bio-Rad, Hempstead, NY. [125I]-labeled EGF (specific activity of 100 Ci/tzg) and EGF radioimmunoassay (RIA) kit were obtained from Amersham Co., Arlington Heights, IL. Agarose-bound wheat germ agglutinine (WGA) was obtained from Vector Laboratories, Inc., Burlingame, CA. BCA protein assay kit was purchased from Pierce Biochemicals, Rockford, IL.

Human subjects The subjects for this study were recruited from the patient population of the oral surgery clinics of NJDS and University Hospital and the medical and dental students at UMDNJ. All subjects selected for the study were asked to sign an Informed Consent Form approved by the Institutional Review Board (IRB), NJDS-UMDNJ. All subjects were healthy volunteers who had good oral health. The volunteers included 5 non-smokers and 10 smokers. The smoker subjects were the ones who smoked an average of 20 cigarettes or more per day for the past 5 years or longer.

147

Saliva collection The procedure was conducted as previously described [23]. Subjects in the smokers' group were instructed not to smoke on the day of sampling. All subjects were instructed to collect saliva in their mouth for 5 min without swallowing and to spit into a clean plastic container. The first 5-min collection was discarded as an adaptation period. Each subject was asked to collect six sequential 5-min specimens, which were then pooled. The saliva sample was kept in ice during the collection and the pooled samples were placed in a freezer at -20°C within 2 hours. Collection of buccal mucosal tissue The oral mucosal biopsy specimens were obtained from the buccal surfaces of each subject as described in our previous study [13]. The biopsy technique consisted of making a 5-10 mm elliptical incision in the cheek mucosa after local anesthesia had been infiltrated into the area. The incision was carried down to, but not into the buccinator muscle. An Allis clamp was used to grasp the incised mucosa. The mucosal specimen was lifted up and freed from the muscle attachment at its base. Once the ellipse was removed, the mucosa was undermined around the periphery of the incision. This allowed for tension free primary closure. Bleeding was controlled with pressure or ties as necessary. The incision was primarily closed using 3 - 0 plain gut interrupted suture. One biopsy sample was obtained from each individual involved in the studies. The specimens, immediately after collection, were rinsed with cold saline, transferred from the clinic to the Dental Research Center laboratory and stored in liquid nitrogen until use. EGF level in saliva An RIA kit for E G F determination was purchased from Amersham Inc. The procedure was performed exactly as described in the Instruction sheet which comes with the RIA kit. Briefly, saliva sample was diluted in assay buffer (Amerlex-M) and mixed with 125I-labeled E G F tracer and E G F antiserum. The mixture was incubated for 3 h at 25°C. After incubation, the Amerlex-M containing second antibody was added to the reaction mixture and incubated for 10 min at 25°C followed with centrifugation (1500 × g, 10 min). The pellet was counted by the gamma counter. The E G F concentration in the sample was determined by using an RIAbinding program by which a linear regression of a logit B/B o versus log dose transformation from the standard curves was performed for the determination of unknown samples. Saliva cotinine analysis To adjust the variation of different brand and number of cigarettes consumed by the smokers, salivary cotinine (major metabolite of nicotine) level was determined and served as the index for the amount of cigarette used. Saliva samples were packed in a styrofoam container with dry ice and shipped to the America Health Foundation Clinical Biochemistry Laboratory, Valhalla, New York. Saliva cotinine was analyzed by RIA as described elsewhere [23,24]. The laboratory was not informed of the source of sample. The cut-off point for identifying smokers and non-smokers, based on previously reported work [24], ranges from 8-30 ng/ml saliva.

148

Membrane preparation Crude membrane of buccal tissue was prepared as previously described [13]. Tissue was mixed with membrane preparation buffer (0.25 M sucrose with 25 mM Tris-HCl buffer, pH 7.4 containing 1 mM PMSF, 100 kunits/ml aprotinin and 1 t~g/ml leupeptin) and homogenized in an ice-chilled beaker by a Teckmar Tissuemizer for 30 s. The homogenate was centrifuged at 600 × g for 25 min and the supernatant was separated, aliquoted and frozen at -70°C until use. Protein was determined by a BCA protein assay kit (Pierce Chem. Co., Rockford, IL. 125I-labeled EGF binding assay Standard 125I-labeled EGF binding assay was conducted by a previous method [13]. Tissue membranes were incubated in triplicate with 125I-labeled EGF and 5 mM Tris-HCl (pH 7.0), containing 125 mM sucrose, 75 mM NaC1, 0.5 mM Ca ++ and 0.5% BSA (w/v) for 30 min at 25°C. Membrane-bound 125I-labeled EGF was separated immediately by centrifugation and counted in a Packard Gamma counter. Specific binding was obtained by subtracting the radioactivity of non-specific binding from total binding and expressed as fmol 125I-labeled EGF bound/mg protein. EGF receptor protein tyrosine kinase To study the effect of cigarette smoking on EGF receptor protein tyrosine kinase activity, the membrane samples from each group were pooled to obtain enough protein. The pooled samples were solubilized and purified by a WGA agarose method as previously described [18]. EGF receptor phosphorylation was performed by a modification of the method of Lowe et al. [25] and our own [18]. Lectin-purified receptor preparations was preincubated in the absence or presence of 1.0 ~M EGF in a final volume of 80 ~1 50 mM HEPES buffer (pH 7.6) with 2 mM PMSF. After 30 min at 22°C, the reaction mixture was chilled to 0°C and phosphorylation was initiated by the addition of 20 txl reaction mixture to give final concentrations of 5 /xM [32p]ATP (spec. act., 20 t~Ci/nmol) 1 mM cytidine-5-triphosphate, 3 mM MnC12, 20 mM MgC12 and 1 mM sodium vanadate. After 5 min at 0°C, the reaction was terminated by the addition of an equal volume of sodium phosphate buffer (10 raM, pH 7.4) that contained NaC1 (0.15 M), Triton X-100 (10%, v/v), deoxycholic acid (0.55%, w/v), SDS (0.1%, w/v) and bovine serum albumin (5 ng/ml)). The samples were clarified by centrifugation at 800 × g for 20 min at 4°C. The 32p_ incorporated EGF-receptor was mixed with 100 tA of SDS-PAGE sample buffer (0.17 M Tris-HC1, 10% SDS and 100 mM dithiothreitol, pH 6.8) and heated at 100°C for 10 min. Aliquots (80 /~1) were analyzed by SDS-PAGE followed by autoradiography. The intensity of 32p-labeled protein bands shown on X-ray film was determined by a Biomed laser densitometer. Western blot of EGF receptor EGF receptor protein was examined by SDS-PAGE and immunostaining procedures. Lectin-purified receptor preparations were electrophoresed on 7.5% SDS-polyacrylamide gels and transferred to nitrocellulose paper, according to the methods of Laemmli [26] and Towbin et al. [27], respectively. The blotted nitrocellulose sheet was washed in 20 mM Tris, containing 500 mM NaCI, pH 7.5

149 (TBS), on a rocker platform for 10 min followed by 3% gelatin for 25 min. The nitrocellulose paper was further washed twice with TBS containing 0.05% Tween-20 (TTBS) and incubated with E G F receptor antiserum (Ab-2, Oncogene Sci., Inc.) for 2 h at room temperature. The blotted nitrocellulose paper was washed twice with TTBS and then incubated with second antibody-peroxidase conjugate for 2 h. Immunoreactive proteins were visualized by incubation for 15-45 min with 3-amino-Oethylcarbazole in the presence of 0.015% hydrogen peroxide. A parallel experiment was performed to monitor the quantity of protein used for the Western blot analysis. After SDS-PAGE separation, the protein bands were stained by the silver-stain kit (Bio-Rad Inc.) according to the company's protocol. Results

Figure 1 shows the levels of E G F and cotinine in salivary samples from nonsmokers and smokers. Although the human subjects involved in this study have been screened to be individuals who smoked more than 20 cigarettes per day for more than 5 years, salivary cotinine level was determined to serve as an index for the amount of cigarette consumed. Cotinine, a major metabolite of nicotine, is a useful and reliable indicator of nicotine intake [23,24]. Recent study indicates that the cotinine level in serum is closely related to salivary cotinine level [23,24]. Salivary cotinine level was consistent with the self-reported smoking status. Salivary cotinine concentration was significantly higher in the smokers compared to that of the nonsmokers, which was undetectable ( < 2 ng/ml saliva). The smokers'group was further

II r~ r~

***

P