BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS 229, 982–989 (1996)ARTICLE NO. 1911

Activation of Hepatic NF-kB by Phenobarbital in Rats

Yixin Li,* Lai K. Leung,†,‡ Brett T. Spear,§ and Howard P. Glauert*,†,‡,1

*Graduate Center for Toxicology, †Department of Nutrition and Food Science, ‡Nutritional Sciences Program, and§Departments of Microbiology and Immunology, and Pathology and Laboratory Medicine,

University of Kentucky, Lexington, Kentucky 40506

Received October 31, 1996

Nuclear factor kB (NF-kB) is an important stress-induced transcription factor in many cell types. In thisstudy we examined if the hepatic tumor promoting agent phenobarbital could activate NF-kB in the liver.Rats were fed 0.05% phenobarbital in the diet for 3, 6, or 10 days. Benzyloxyresorufin dealkylase activitywas greatly increased in rats receiving phenobarbital. The DNA binding activity of NF-kB was determinedusing electrophoretic mobility shift assays. After 3 days, NF-kB activity was increased two-fold in ratsreceiving phenobarbital. NF-kB activity continued to increase at 6 and 10 days in rats fed phenobarbital,with a four-fold increase seen at day 10. Cold competition and antibody supershift experiments showedthe DNA binding activity to be specific for NF-kB. These data show that phenobarbital activates hepaticNF-kB in rats. q 1996 Academic Press

Phenobarbital has been demonstrated to be one of the most efficacious tumor promotingagents in the liver (1). The promoting activity of phenobarbital is likely related to its effectson cell proliferation and apoptosis. After beginning administration, phenobarbital greatly in-creases cell proliferation in the liver (2). After several days, however, cell proliferation in theliver returns to control levels, and further mitogenic stimulation does not again increase cellproliferation (2). Phenobarbital also inhibits apoptosis in rat liver, both in normal and preneo-plastic hepatocytes (3). The biochemical and molecular mechanisms by which phenobarbitalbrings about these changes in cell proliferation and apoptosis, however, have not been defini-tively determined. While the promoting activity of phenobarbital is likely to be related to itseffects of cell proliferation or apoptosis, other mechanisms are possible.

One mechanism by which phenobarbital may exert its promoting activity is by increasingcellular oxidative damage. Phenobarbital increases specific forms of cytochrome P-450 in theliver. The promoting activity of several barbiturates has been correlated with their ability toinduce cytochrome P-450 in the liver (4). Active oxygen in the form of superoxide or hydrogenperoxide can be released as a by-product from cytochrome P-450 and thus could contributeto the tumor promoting activity of these chemicals, for example by causing lipid peroxidationor oxidative DNA damage (5).

It is also possible that active oxygen may influence carcinogenesis is by altering geneexpression. Active oxygen can directly modulate transcription through antioxidant responseelements (AREs) found in the promoters of certain genes (6). For example, phenobarbitalactivates the promoters of the glutathione-S-transferase Ya subunit and NAD(P)H: oxidoreduc-tase1 genes through AREs (7). Active oxygen can also induce transcription through membersof the nuclear factor-kB (NF-kB) family. While NF-kB was originally identified as an inducibletranscriptional regulator in B lymphocytes (8), additional studies have shown that NF-kB is

1 Please address all correspondence to: Dr. Howard P. Glauert, University of Kentucky, Department of Nutritionand Food Science, 204 Funkhouser Bldg., Lexington, KY 40506-0054. Fax: (606) 257-3707. E-Mail: nfshpg@ukcc.uky.edu.

0006-291X/96 $18.00Copyright q 1996 by Academic PressAll rights of reproduction in any form reserved.

982

AID BBRC 5865 / 6914$$$821 12-10-96 13:08:48 bbrca

Vol. 229, No. 3, 1996 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

an important activator in many different cell types, including hepatocytes (9). NF-kB consistsof two subunits (p50 and p65) bound to an inhibitory subunit (IkB) in the cytosol (10). Afteractivation, NF-kB is released from IkB and is translocated to the nucleus, where it binds toits cognate DNA sequences and increases the transcription of specific genes (9). NF-kB isactivated by bacterial LPS, phorbol esters, tumor necrosis factor, and hydrogen peroxide,whereas its activation is inhibited by antioxidants (11). These effects by hydrogen peroxideand antioxidants have led to the suggestion that NF-kB is an active oxygen-induced transcrip-tion factor.

Previous studies have demonstrated that NF-kB is activated in the liver after partial hepatec-tomy and during the acute phase response (9,12). In addition, we have shown that the peroxi-some proliferator ciprofibrate activates NF-kB in the liver and in rat hepatoma cells (13,14).These data indicate that NF-kB is an important stress-induced factor in the liver. We thereforeexamined if phenobarbital could activate NF-kB in the liver. Rats were fed phenobarbital inthe diet for 3, 6, or 10 days and the DNA binding activity of NF-kB in the liver was determinedusing electrophoretic mobility shift assays. We found that phenobarbital increased the activationof NF-kB, with the DNA binding activity of NF-kB increasing from 3 to 10 days.

METHODS

Chemicals. Phenobarbital was obtained from Sigma Chemical Company (St. Louis, MO).Experimental design. Male Sprague-Dawley rats, purchased from Harlan Sprague-Dawley (Indianapolis, IN), were

allowed to adjust for one week before starting the study. Rats were fed an unrefined diet (Purina Rodent LaboratoryChow No. 5001, Purina Mills, St. Louis, MO) containing 0.05% phenobarbital; control rats received the same dietwithout phenobarbital. Rats were killed after 3, 6 or 10 days by overexposure to carbon dioxide. The livers werequickly removed, frozen in liquid nitrogen, and stored at 0807C until further use.

Analysis of cytochrome P-450 2B enzyme activity. Liver tissue was homogenized in ice-cold KCl-EDTA solution(1.15% KCl, 0.1 mM EDTA, pH 7.4). The homogenate was centrifuged at 10,000 1 g for 10 minutes and thesupernatant was then centrifuged at 100,000 1 g for 1 hour. The microsome pellets were suspended in KCl-EDTAand stored at 0807C. The activity of the cytochrome P-450 2B family (benzyloxyresorufin dealkylase) was determinedin microsomes as described by Burke et al. (15).

Electrophoretic mobility shift assays. Rat liver nuclear extracts were prepared as described by Deryckere andGannon (16). To monitor the presence of NF-kB DNA binding activity in the nuclear extracts, we performed electropho-retic mobility shift assays (EMSA). A fragment containing two NF-kB binding sites (5* CAAGGGGACTTTCCATGG-ATCCAAGGGGACTTTCCATG) was flanked by GATC overhangs and cloned into the BglII site of a modified pUC9vector (Spear et al., 1995). The NF-kB 2-mer was excised by digestion of the plasmid with SalI and XhoI, dephosphory-lated with calf intestinal phosphatase, and purified by acrylamide gel electrophoresis. For use as a radioactive probe,the NF-kB 2-mer was end-labeled with [g-32P]ATP using polynucleotide kinase. A gel-purified, size-matched fragmentfrom the pUC9 polylinker region was used as a non-specific competitor. Binding reactions were performed byincubating rat liver nuclear extracts (10 mg protein/20 ml) in reaction buffer [250 mM KCl, 50 mM Tris (pH 7.4), 50mM HEPES, 6.25 mM DTT, 5.5 mM EDTA, 75% glycerol (v:v)] and 1 mg polydI:dC in the presence or absence ofnon-radioactive competitor DNA fragments for 5 min on ice, followed by a 25 min incubation at room temperaturewith roughly 1 ng (approximately 20,000 cpm) of the labeled NF-kB probe. For supershift experiments, 1 ml of rabbitanti-p50, rabbit anti-p65, or preimmune rabbit antiserum (Santa Cruz Biotechnology, Santa Cruz, CA) were incubatedwith nuclear extracts for 15 min on ice prior to the addition of radiolabeled probe. Reaction mixtures were electropho-resed on 6% polyacrylamide gels in 1 1 TGE buffer (0.25 M Tris, 2.5 M glycine and 2.5 mM EDTA) at 150 V for2 hours. Gels were dried and autoradiographed using Kodak X-omat film. To quantitate the radioactivity in a particularband, we also analyzed dried gels by using a radioanalytic imaging system (Ambis, San Diego, CA).

Statistics. Values for phenobarbital-treated rats were compared to those for control rats using students t-test (17).

RESULTS

Phenobarbital is an efficacious hepatic tumor promoter that induces hepatocyte proliferationand specific forms of cytochrome P-450 and that blocks apoptosis (3,18), but the mechanismsof these phenobarbital-induced effects on the liver are not clear (19). Since phenobarbitalinduces cytochrome P-450s, which produce reactive oxygen intermediates in in vitro systems(5), we sought to investigate whether phenobarbital would activate NF-kB in rat liver.

983

AID BBRC 5865 / 6914$$$821 12-10-96 13:08:48 bbrca

Vol. 229, No. 3, 1996 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 1. Benzyloxyresorufin dealkylase activity in the livers of rats in response to phenobarbital treatment. Ratswere fed a diet containing 0.05% phenobarbital; control rats were fed the same diet without phenobarbital. Animalswere sacrificed 3, 6, and 10 days after beginning treatment. Each column represents data from 8 rats. Benzyloxyresorufindealkylase activity was measured as described in the Materials and Methods. Statistical analysis was done by theStudent’s t-test. *Põ0.05

Rats were fed a diet containing 0.05% phenobarbital; control rats were given the same dietwithout phenobarbital. After 3, 6, and 10 days, rats were killed and the livers were removed.Phenobarbital induces the activity of several isoforms of cytochrome P-450, including the 2Bfamily. To measure the efficacy of phenobarbital treatment, we measured a cytochrome P-4502B-associated enzyme activity (benzyloxyresorufin dealkylase) (Fig. 1). Whereas low basalbenzyloxyresorufin dealkylase levels that did not change over time were observed in controlrats, a 20-fold increase in benzyloxyresorufin dealkylase activity was seen by day three andremained elevated in phenobarbital-treated rats. These results are consistent with previousstudies and indicate that the rats were responding to the phenobarbital.

Since the transcription factor NF-kB is normally bound to IkB in the cytosol, NF-kB activityis best measured by the increased presence of DNA binding activity in the nucleus. Nuclearextracts therefore were prepared and used in EMSAs with a radiolabeled probe containing twoNF-kB binding sites. Low but detectable levels of NF-kB remained unchanged over the 10day period in nuclear extracts from control rats (Figure 2). Three days after the initiation oftreatment, an increase in NF-kB DNA binding activity was observed in phenobarbital-treatedrats (Fig. 2A). NF-kB levels in phenobarbital-treated rats continued to increase six and tendays after treatment. Quantitative radioanalytic image analysis indicated that the level ofinduction was nearly two-fold by day 3 and about four-fold by day 10 (Fig. 2B).

To confirm that the phenobarbital-induced complex in liver extracts contained NF-kB, weperformed several controls. EMSAs were carried out in the presence of an excess of unlabeledself (fragment containing the two NF-kB binding sites) and non-self (size-matched unrelatedDNA) fragments to determine if the complex bound the consensus NF-kB site (Fig. 3). Theself fragment could effectively compete for DNA binding activity in a dose-dependent manner(lanes 4 and 5). A 25- or 50-fold molar excess of the unrelated fragment did not diminishbinding to the radiolabeled probe (compare lane 1 with lanes 2 and 3), demonstrating specificityfor the NF-kB site. In addition, EMSAs were performed with antibodies which react witheither the NF-kB p50 or p65 subunits (Fig. 4). The presence of pre-immune rabbit serum didnot alter the NF-kB complex (compare lanes 1 and 2). The addition of the anti-p50 antiserum(lane 3) resulted in the presence of two new, supershifted complexes (labeled * and #). Thesenovel complexes presumably contain the anti-p50 antibodies and p50, resulting in complexeswith new molecular weights. The anti-p65 antiserum also caused formation of a new complex

984

AID BBRC 5865 / 6914$$$821 12-10-96 13:08:48 bbrca

Vol. 229, No. 3, 1996 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 2. A. EMSA to detect NF-kB induction in response to phenobarbital treatment in rat liver. Liver nuclearextracts were prepared from rats which were sacrificed three (lanes 1, 2, 7, 8), six (lanes 3, 4, 9, 19), or ten (lanes5, 6, 11, 12) days after beginning administration of the diet. Lanes 7-12 (/ ciprofibrate) represent samples from ratsfed a diet containing 0.05% phenobarbital; lanes 1-6 (control) correspond to samples from rats receiving same dietwithout phenobarbital. Each lane represents a sample from a single rat. 10 mg of nuclear protein was used in eachlane. F.P., free probe; NF-kB, DNA/NF-kB complexes. B. Quantitation of the EMSA data shown in Fig. 2A using aradioanalytic imaging system (Ambis Corp. San Diego, CA). Dried acrylamide gels shown in A were used. Totalcpms in the different bands were determined; net cpms were obtained by subtracting background radioactivity. Eachdata point represents the average of the two rats in each sample group shown in A. Statistical analysis was done bythe Student’s t-test. *Põ0.05

(also labeled *) which likely contained anti-p65 antibodies and p65 (lane 4). These resultsindicate that p50 and p65 are components of the phenobarbital-induced complex.

DISCUSSION

The biological effects of phenobarbital on cell proliferation, apoptosis, and tumor promotionin the liver have been known for some time (1-3). The mechanisms by which phenobarbitalproduces these effects, however, are still not clear. Changes in gene expression, such as the

985

AID BBRC 5865 / 6914$$$821 12-10-96 13:08:48 bbrca

Vol. 229, No. 3, 1996 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 3. EMSA competition assays to confirm NF-kB binding activity in phenobarbital-treated rats. Liver nuclearextracts were used from rats fed a diet containing 0.05% phenobarbital for 10 days. This assay used non-radioactiveDNA fragments to show specificity for the NF-kB consensus binding site. Nuclear extracts were incubated in theabsence (lane 1) or presence of a 25-fold (lanes 2 and 4) or 50-fold (lanes 3 and 5) molar excess of non-specific(N.S., lanes 2 and 3) or specific (Self; lanes 4 and 5) unlabeled fragment before the addition of 1 ng of the radiolabeledfragment containing 2 copies of the NF-kB consensus binding site.

induction of specific cytochrome P-450s, may be important in mediating these cellular changes.In this study, we have demonstrated that phenobarbital activates the transcription factor NF-kB, which represents a potential new mechanism by which phenobarbital may influence tumorpromotion.

The mechanism by which phenobarbital activates NF-kB is likely related to the productionof active oxygen. Reactive oxygen intermediates are thought to be in a common pathway toactivate NF-kB; antioxidants have been shown to block NF-kB activation by all stimuli testedto date (11). Thus, phenobarbital could activate NF-kB by shifting the balance of oxidantsand antioxidants within cells. Since the cytochrome P-450 2B family is highly induced byphenobarbital (19), it is possible that cytochrome P-450 is involved in the activation of NF-kB by phenobarbital. Cytochrome P-450s have been found to produce active oxygen in theform of hydrogen peroxide or superoxide as by-products (20). It is also possible that phenobar-bital increases NF-kB by decreasing antioxidant levels. Previous studies suggest that this isnot likely, however. Phenobarbital has been found to increase hepatic vitamin C concentrationsand catalase activity, to have no effect on vitamin E, and to have variable effects on glutathioneconcentrations and glutathione peroxidase activity (21-27).

986

AID BBRC 5865 / 6914$$$821 12-10-96 13:08:48 bbrca

Vol. 229, No. 3, 1996 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 4. Antibody supershift assays demonstrate that p50 and p65 are in the NF-kB complexes found in liver-extracts from phenobarbital-treated rats. Liver nuclear extracts were used from rats that were fed a diet containing0.05% phenobarbital for 10 days. EMSAs were performed with no antibodies (lane 1), non-specific rabbit antiserum(lane 2), anti-p50 antibodies (lane 3) or anti-p65 antibodies (lane 4). F.P., free probe; NF-kB, DNA/NF-kB complexes.# and * indicate supershift bands.

Another piece of evidence that cytochrome P-450 or other enzymes induced by phenobarbitalmay be important in NF-kB activation is the timing of the increase. The rapid induction ofNF-kB activity is a hallmark of many experimental systems; for example, NF-kB activationis evident 30 minutes after partial hepatectomy (12). A modest increase in NF-kB activity wasseen at day 3, with a continued increase between days 3 and 10. If NF-kB were activateddirectly by phenobarbital, one would expect a large increase in NF-kB activation by day 3.But the moderate increase seen at day 3 in combination with the increase in DNA bindingbetween days 3 and 10 suggest that NF-kB activation is due to biochemical changes (such asactive oxygen release from cytochrome P-450) rather than a direct effect of phenobarbital.

We previously showed that NF-kB was also activated by the peroxisome proliferator ci-profibrate (13). From EMSA supershift studies, it appears that phenobarbital and ciprofibratemay activate different members of the Rel/NF-kB family, however. The addition of the anti-p50 antibody to EMSAs caused two supershift bands in nuclear extract from phenobarbital-treated rat liver, whereas only one supershift band was observed in nuclear extract fromciprofibrate-treated rat liver. After addition of the anti-p65 antibody, a supershift was seenwith phenobarbital, whereas with ciprofibrate the NF-kB complex was diminished with no

987

AID BBRC 5865 / 6914$$$821 12-10-96 13:08:48 bbrca

Vol. 229, No. 3, 1996 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

supershift. These data imply that the activation of the NF-kB family by these two chemicalsmay be more complex than simply increasing cellular hydrogen peroxide concentrations. Sincethe promoting properties of phenobarbital and peroxisome proliferators are different (28,29), theactivation of different members of the NF-kB family by these agents would not be surprising.

Studies suggest that NF-kB activation is important in the regulation of both cell proliferationand apoptosis in the liver. NF-kB is activated very early after partial hepatectomy in rodentsand likely plays an essential role during liver regeneration (12). NF-kB is not activated beforeor during hepatocyte proliferation in vivo immediately after administration of the peroxisomeproliferator BR931, but is activated after administration of lead nitrate (30). It is currentlyunknown whether NF-kB is important in the induction of cell proliferation by other xenobiotics.Studies with gene-knockout animals have provided information about the roles of NF-kBsubunits in cell proliferation and apoptosis. B lymphocytes from mice lacking p50 do notundergo normal cell proliferation following mitogenic stimulation, although effects on hepato-cyte proliferation were not tested (31). The livers of mice lacking p65/RelA, however, undergomassive apoptosis at 15-16 days of gestation (32). This study suggests that the p65 subunit ofNF-kB is a major regulator of programmed cell death in hepatocytes.

Further studies will be required to investigate the consequences of increased NF-kB levelsin rat liver after phenobarbital administration. NF-kB may contribute to changes in the tran-scription of genes following phenobarbital treatment. It will also be important to determinewhich cell types in the liver are producing the increased NF-kB DNA binding activity inresponse to phenobarbital. While hepatocytes are the major cell type in the liver, numerousother cell types exist. The determination of whether NF-kB activation is essential for thepromoting activity of phenobarbital is also an important question. Regardless of the answersto these questions, the ability of phenobarbital to activate NF-kB in the liver provides furtherevidence that NF-kB is an important transcriptional regulator in the liver.

ACKNOWLEDGMENTSThese studies were supported by NIH Grants ES-05815 and CA-01688, by American Cancer Society Institutional

Grant 163, and by the Kentucky Agricultural Experiment Station.

REFERENCES1. Pitot, H. C., Goldsworthy, T. L., Moran, S., Kennan, W., Glauert, H. P., Maronpot, R. R., and Campbell, H. A.

(1987) Carcinogenesis 8, 1491–1499.2. Jirtle, R. L., and Meyer, S. A. (1991) Dig. Dis. Sci. 36, 659–668.3. Bursch, W., Lauer, B., Timmermann-Trosiener, I., Barthel, G., Schuppler, J., and Schulte-Hermann, R. (1984)

Carcinogenesis 5, 453–458.4. Lubet, R. A., Nims, R. W., Ward, J. M., Rice, J. M., and Diwan, B. A. (1989) J. Am. Coll. Toxicol. 8, 259–268.5. Bondy, S. C., and Naderi, S. (1994) Biochem. Pharmacol. 48, 155–159.6. Jaiswal, A. K. (1994) Biochem. Pharmacol. 48, 439–444.7. Pinkus, R., Bergelson, S., and Daniel, V. (1993) Biochem. J. 290, 637–640.8. Sen, R., and Baltimore, D. (1986) Cell 46, 705–716.9. Siebenlist, U., Franzoso, G., and Brown, K. (1994) Annu. Rev. Cell. Biol. 10, 405–455.

10. Verma, I. M., Stevenson, J. K., Schwarz, E. M., Van Antwerp, D., and Miyamoto, S. (1995) Genes Develop. 9,2723–2735.

11. Pahl, H. L., and Baeuerle, P. A. (1994) Bioessays 16, 497–502.12. Cressman, D. E., Greenbaum, L. E., Haber, B. A., and Taub, R. (1994) J. Biol. Chem. 269, 30429–30435.13. Li, Y., Leung, L. K., Glauert, H. P., and Spear, B. T. (1996) Carcinogenesis, in press.14. Li, Y., Glauert, H. P., and Spear, B. T. (1995) Proc. Am. Assoc. Cancer Res. 36, 173.15. Burke, M. D., Thompson, S., Elcombe, C. L., Halpert, J., Haaparanta, T., and Mayer, R. T. (1985) Biochem.

Pharmacol. 34, 3337–3345.16. Deryckere, F., and Gannon, F. (1994) Biotechniques 16, 405.17. Gill, J. L. (1978) Design and Analysis of Experiments in the Animal and Medical Sciences, Volume 1, The Iowa

State University Press, Ames, Iowa.18. Andersen, M. E., Mills, J. J., Jirtle, R. L., and Greenlee, W. F. (1995) Toxicology 102, 223–237.

988

AID BBRC 5865 / 6914$$$821 12-10-96 13:08:48 bbrca

Vol. 229, No. 3, 1996 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

19. Waxman, D. J., and Azaroff, L. (1992) Biochem. J. 281, 577–592.20. Goeptar, A. R., Scheerens, H., and Vermeulen, N. P. E. (1995) Crit. Rev. Toxicol. 25, 25–65.21. Lee, K. W., Wang, H. J., Murphy, P. A., and Hendrich, S. (1995) Nutr. Cancer 24, 267–278.22. Schramm, H., Robertson, L. W., and Oesch, F. (1985) Biochem. Pharmacol. 34, 3735–3739.23. Hahn, H. K. J., Barak, A. J., Tuma, D. J., and Sorrell, M. F. (1978) Pharmacology 17, 341–344.24. Liotti, F. S., Menghini, A. R., Guerrieri, P., Mariucci, G., and Armellini, R. (1993) Bull. Cancer 80, 62–69.25. Ahotupa, M., Bereziat, J. C., Mantyla, E., and Bartsch, H. (1993) Carcinogenesis 14, 1225–1228.26. Hendrich, S., Duitsman, P., Krueger, S. K., Jackson, A., and Myers, R. K. (1991) Nutr. Cancer 15, 53–62.27. Sewerynek, E., Abe, M., Reiter, R. J., Barlow-Walden, L. R., Chen, L., McCabe, T. J., Roman, L. J., and Diaz-

Lopez, B. (1995) J. Cell. Biochem. 58, 436–444.28. Cattley, R. C., and Popp, J. A. (1989) Cancer Res. 49, 3246–3251.29. Glauert, H. P., Beer, D., Rao, M. S., Schwarz, M., Xu, Y. D., Goldsworthy, T. L., Coloma, J., and Pitot, H. C.

(1986) Cancer Res. 46, 4601–4606.30. Ohmura, T., Leddacolumbano, G. M., Piga, R., Columbano, A., Glemba, J., Katyal, S. L., Locker, J., and Shino-

zuka, H. (1996) Am. J. Pathol. 148, 815–824.31. Sha, W. C., Liou, H. C., Tuomanen, E. I., and Baltimore, D. (1995) Cell 80, 321–330.32. Beg, A. A., Sha, W. C., Bronson, R. T., Ghosh, S., and Baltimore, D. (1995) Nature 376, 167–170.

989

AID BBRC 5865 / 6914$$$821 12-10-96 13:08:48 bbrca