Adenylate Cyclase Activation by Cholera Toxin in Pig Epidermis: An Obligatory Role of the GTP-Regulatory Protein

Transcription

1 22-22X/83/ $2./ TH JOURNAL OF INVSTIGATIV DRMATOLOGY, 81: , 1983 Copyright 1983 by The Williams & Wilkins Co. Vol. 81, No. 2 Pn:nted in U.S.A. Adenylate Cylase Ativation by Cholera Toxin in Pig pidermis: An Obligatory Role of the GTP-Regulatory Protein JuNJI TAKDA, M.D., KNJI ADACHI, M.D., PH.D., KNNTH M. HALPRIN, M.D., VIToR LVIN, B.S., AND CLYD WooDYARD, B.S. Dermatology Servie, Veterans Administration Medial Center, and Department of Dermatology, University of Mia.mi Shool of Mediine, Mia.mi, Florida., U.S.A. Cholera toxin (CT) stimulates the epidermal adenylate ylase system in vitro. This stimulation was demonstrated in the skin (slie) floating system and the homogenate (membrane) assay system. With the floating system, the addition of CT to the inubation medium aused a marked aumulation of AMP intraellularly, whih was both dose- and time-dependent. A 1-h lag time was present before ativation started. Pretreatment of the skin with CT hanged the nature of the stimulatory effet aused by epinephrine and histamine, i.e., the transient aumulation of AMP (a peak at 5 min and subsequent derease) was no longer observed but the stimulation beame persistent. With the membrane assay system in whih the reeptor omponents had been unoupled, adenylate ylase ativities were markedly stimulated by CT (with guanosine-5'-triphosphate, GTP), guanylyl-,b,y-imidodiphosphate (GTP-analog, Gpp[NH]p), or sodium fluoride. The stimulation was both dose- and time-dependent without an initial time lag. ither CT or Gpp[NH]p ould fully ativate adenylate ylase, and the simultaneous addition of both did not ause further additive stimulation. These data are onsistent with the view that the GTPregulatory protein plays a key role in the ativation of adenylate ylase, and that CT both ativates the atalyti unit and modifies the response to reeptor hormones through its ation on this protein. The adenylate ylase system in the ell membrane is omposed of at least 3 omponents: a reeptor site (R), a nuleotide regulatory protein (N), and a atalyti unit (C). Reent studies have shown that N has a pivotal role in the adenylate ylase system [1-6] and the ativity of N is regulated by the guanine nuleotide; i.e., N is ative when guanosine-5'-triphosphate (GTP) is present and N is inative when guanosine-5' -diphosphate (GDP) is attahed. In the basal state GDP oupies N [7]. Hormones that ativate adenylate ylase at by enhaning replaement of the bound GDP by GTP [7], and subsequent hydrolysis of the bound GTP terminates the ativation of adenylate ylase [8]. Replaement of GTP by the hydrolysis-resistant analogue guanylyl-,b,y-imidodiphosphate (Gpp[NH]p) produes a persistently ative state [9]. Cholera toxin (CT) an ativate adenylate ylase of intat ells and broken ells [1] by atalyzing the transfer of adenosine-5'-diphosphate (ADP)-ribose from Manusript reeived August 23, 1982; aepted for publiation Marh l, Supported in part by grants AM 171(9 from the National Institutes of Health, the Dermatology Foundation of Miami, and the A.. Wellman Fund. Reprint requests to: Dr. K. Adahi, Veterans Administration Medial Center, 121 Northwest 16th Street, Miami, Florida Abbreviations: ADP: adenosine-5' -diphosphate ATP: adenosine-5' -triphosphate C: atalyti unit (adenylate ylase) CT: holera toxin niotinamide adenine dinuleotide (NAD) to a 42, M,. polypeptide whih is part of N [11]. CT ativation has been explained by either inhibiting GTP hydrolysis at N [12] or enhaning attahment of GTP instead of GDP in a manner similar to hormones [13,14], or both. To date, we have reported epidermal adenylate ylase systems oupled with various reeptors, suh as ateholamine [15], prostaglandin [16], histamine [17], and adenosine [18]. These reeptors individually stimulate their own adenylate ylase systems in the epidermis in a speifi manner [19], while CT is supposed to modify N in a nondisriminatory way to ativate all of the reeptor systems at the same time. In an in vitro system, then an be oupled with a foreign R (from other speies) to mediate a hormone reation [2]. The effets of CT on growth and differentiation of keratinoytes as well as those on the yli AMP levels in ultured epidermal ells have been reported [21-23]. In skin, CT presumably ats on the N omponent of adenylate ylase as it does in other tissues, however its obligatory role has not been doumented. In our attempt to disset eah omponent of the epidermal adenylate ylase systems, we now demonstrate that CT stimulates the N omponent of epidermal adenylate ylase both in an in vitro floating system and in an homogenate assay system. M a.teria.ls MATRIALS AND MTHODS [2,8-3 H]AMP (sp at 36.4 Ci/mmol) was purhased from New ngland Nulear (Boston, Massahusetts), [a- 32 P] adenosine-5'-triphosphate (ATP) (sp at 3-5 Ci/nmol) from Amersham, U.K., a AMP radioimmunoassay kit from Collaborative Reseru h In. (Waltham, Massahusetts), holera toxin (whole, A and B unit) from Calbiohem (LaJolla, California), and alumina (Neutral Type WN-3) and Dowex 5W (5 X 4-4) from Sigma (St. Louis, Missouri). All other hemials used were of the highest grade available. Floating xperiment Skin slies (approximately 8% epidermis) from domesti pigs were taken by a keratome with the utting blade adjusted to.2 mm or.3 mm. The skin slies were ut 5 X 5 mm squru e at 4 C and the skin squares were floated keratin layer up on Hanks' balaned salt solution. After preinubation at 37 C, the squru es were inubated in various hemials and drugs. In a ase of long iilubation, the squares were plaed in a humidity- and temperature-ontrolled inubator having a 95% air/5% C 2 supply. After inubation they were frozen by dry ie to stop the reation; the AMP ontents were measured by radioimmunoassay [24] with miromodifiation [25]. Protein was measured by the method of Lowry et a! with human albumin as a standard [26]. DT: dithioerythreitol GT A: ethyleneglyol-bis-(,8-aminoethyl ether)-n,n' -tetraaeti aid GDP: guanosine-5' -diphosphate Gpp[NH]p: guanylyl-,8, y-imidodiphosphate GTP: guanosine-5' -triphosphate Hepes: N-2-hydroxyethylpiperazine-N' -2-ethanesulfoni aid IBMX: isobutylmethylxanthine N: nuleotide regulatory protein NAD: niotinamide adenine dinuleotide NaF: sodium fluoride R: reeptor site 131

2 132 TAKDA T AL Membrane Assay System Pig epidermal ell membrane was prepared as follows. Fresh pig skin taken by a keratome with.3-mm thikness was finely mined with sissors and then homogenized by a glass-teflon homogenizer (Potter lvehjem) with 15-2 strokes in 5 mm Hepes buffer (ph 8.) ontaining.1 mm DTA and 1 mm MgC!,. The homogenate was filtered through double-layered gauze. The dermis, whih ould not be destroyed by this homogenization, remained on the gauze [27]. The filtrate was entrifuged at 2, g for 2 min and the resulting preipitate (the membrane) was resuspended in the above buffer (4-5 mg protein/ rnl). These proedures were repeated 3 times for washing. The membrane preparations were quikly frozen in aetone-dry ie and stored at -7 C until assayed. Adenylate Cylase Assay Assays were performed in a fmal volume of 2 Jll. The reation mixture ontained 25 mm Hepes (ph 8.), 5 mm MgC!,,.4 mm GTA,.2 mm isobutylmethylxanthine (IBMX),.2 mm ATP, f.'ci/assay [a- 32 P]ATP,.2 mm AMP, 1 mm phosphoenolpyruvate, and 2 JLg/rnl pyruvate kinase. When CT was added for the ativation, NAD (1.25 mm), dithioerythreitol (DT) (1 mm), and GTP (2 JLM) were also added to the mixture. After inubation at 37 C, reation was stopped by the addition of 2 f.'l of 2% sodium dodeyl sulfate and the mixture was immediately boiled for 3 min. Subsequently, 5 f.'l of CHJAMP (3, pm) and 5 JL! of 1 mm imidazole buffer (ph 6.8) were added and the entire mixture was passed through Dowex 5 and alumina olumns onseutively to isolate [a- 32 P]AMP [28]. [ 32 P]AMP reoveries from the onseutive olumns were 6-8%. ah experiment was repeated to test reproduibility. However, due to onsiderable variations in adenylate ylase ativities in eah membrane preparation, a representative figure rather than an average is shown. Variations within speifi experiments are minimal as indiated in Tables I and II. RSULTS Results with an In Vitro Floating System The inubation of CT with the epidermal squares aused stimulation of pig epidermal adenylate ylase in both a timeand a dose-dependent manner (Fig 1). However, the CT stimulation required prolonged inubation (over 1 h) and the effet persisted. Usually a 3-min inubation with CT (.1-2 JLg/rnl) was not effetive and a dose at.1 JLg/rnl or less resulted in little or no AMP aumulation through a 6-h inubation. Beause the AMP level in the epidermis in this floating system is due to a balane of AMP prodution (by adenylate ylase) and degradation (by phosphodiesterase), the effet of a phosphodiesterase inhibitor (IBMX) addition was also examined. Fig 2 shows that the addition of IBMX (1 mm) markedly potentiated the effet of CT, even at the CT onentration level of.1 JLg/rnl, whih showed only little stimulation by itself after 4 h (Fig 1). These data suggest that a low dose of CT an ativate adenylate ylase in the epidermis and its effet is not through the inhibition of phosphodiesterase. Sine inhibitors of RNA and protein synthesis potentiated the effet of CT in ultured ells [29], we have also added yloheximide (1 JLg/rnl) to the CT stimulation medium. However no potentiation ourred (data not shown). Fig 3 shows that a 2-h pretreatment of CT (1 JLg/rnl) hanged the responses to epinephrine (5 X 1-5 M) and histamine (5 X 1- M) stimulation. This pretreatment itself did not affet the AMP level in the epidermis (Fig 1) and therefore any differene in hormone stimulations with and without CT pretreatment an be ompared. After the CT pretreatment, the epinephrine or histamine stimulation aused faster and higher AMP aumulation than did ontrol (no treatment) and the AMP levels did not derease for at least 3 min (Fig 3A,B). The same results were found when the slies were inubated another 2 h in Hanks' balaned salt solution alone at 37 C following pretreatment with CT and washing (data not shown). This persistent effet is onsistent with the fat [1] that adenylate ylase in the epidermis is irreversibly hanged (or ativated) by the CT pretreatment. Vol. 81, No.2 TABL I. Combinations of various stimulators on the adenylate ylase (N -C omplex of membrane preparation) Addition None NaF (1 mm) NaF + Gpp[NH]p (1JLM) NaF + CT (15 JLg/rnl) NaF + GTP (1 JLM) Gpp[NH]p Gpp[NH]p + CT Gpp[NH]p + GTP CT GTP Adenylate ylase ativity (pmol/mg protein/ 3 min) 13 ± ± ± ± ± ±.7 13 ± ±.2 9 ± ±.8 Details of assay ondition are shown in Materials and Methods. CT stimulus ontained GTP (1 JLM), NAD (1.25 mm), DT (1 mm). Values shown represent the means± S of tripliate determinations. TABL II. ffets of ofator of CT stimulation on NaF-stimulated adenylate ylase ativity Addition NaF (1 mm) NaF + DT (1 mm) NaF + NAD (1.25 mm) NaF + DT + NAD NaF + DT + NAD + GTP (1 JLM) NaF + DT + NAD + CT (1 JLg/ rnl) NaF + DT + NAD + GTP + CT Adenylate ylase ativity (pmol/mg protein/3 min) 89 ± ± ± ± ± ± ± 1.3 Details of assay ondition are shown in Materials and Methods. a; Q. O"l - FIG 1. Time ourse of the effet of CT on the AMP levels in the skin slie floating system. Details of experimental onditions are desribed in Materials and Methods. The onentrations of CT were: 2 Jlg/ ml, Jlg/rnl, - 1 Jlg/ml, -.1 Jlg/ml, X-X ontrol (no addition). Data are averages of 2 separate experiments in dupliate assay eah. Results with a Membrane (Homogenate) Assay System Fig 4 shows the time ourse of AMP aumulation with the homogenate assay system. Both CT and sodium fluoride (NaF) markedly stimulated epidermal adenylate ylase. Although CT potentiated adenylate ylase in both the floating system and the homogenate assay system, NaF did not ativate the floating system (data not shown). The stimulatory effets of CT and NaF were also tested with varying amounts of epidermal membrane (Fig 5). Thus we hose inubation times of 3 min or 45 min with 2-3 JLg of membrane for subsequent exper-

3 Aug.1983 CHOLRA TOXIN STIMULATION OF PIDRMAL ADNYLAT CYCLAS 133 'Ci)... I- 3 2 ') D == 'I 5 hours FIG 2. ffets of IBMX (1 mm) addition to the in vitro floating system: A-A ontrol,!::.-!::. ontrol + IBMX, g/ ml CT, g/ml CT + IBMX, -1!-'g/ ml CT, D-O 1!-'g/ ml CT + IBMX. Data are averages of 3 different skin slies in dupliate assay eah. 2 / 'Q)... lo....! )... <( u 3 / I 1.,..,.---6 IS,. -x min FIG 4. Time ourses of the effets of NaF, CT, and ontrol ativity on the N-C omplex of the adenylate ylase system with epidermal membrane preparation. Details of assay onditions are desribed in Materials and Methods. The onentration of stimulators are - NaF 1 mm,!::.-!::. CT 1 / ml, X-X ontrol (no addition). In eah assay 3 1-'g membrane was used. Values shown represent dupliate determinations. 15 / X 1 : a; Ol - 2 CL 1 ;)/ /o M -... a.. ), /' 2 4 g FIG 5. ffets of CT, GTP analog, and NaF with varying amounts of epidermal membrane preparation. The onentration of stimulators are - NaF 1 mm, - Gpp(NH]p 1 /LM,!::.-!::. CT 1 1-1g/ ml. Data are expressed as pmol AMP formed/3 min. Values shown represent dupliate determinations. 3 min FIG 3. ffet of CT pretreatment. Skin slies were preinubated in Hanks' media for 2 hat 37"C with CT (1!-'g/ ml) (-) or without CT (._e). Subsequently, (A) epinephrine (5 x 1-s M) was diretly added to the media; (B) histamine (5 X 1-4 M) was diretly added to the media. iments. Fig 5 also shows marked stimulation by an analog of GTP (Gpp[NH]p). The CT ativation of adenylate ylase requires GTP in a dose-dependent manner (Fig 6A). Whole CT and "A" subunit (ative unit of CT) had the same dependeny on GTP onen-

4 134 TAKDA T AL 21 Q)... e Q. C) -Q)... A 1 B I J I FIG 6. ffet of GTP and its analog on the membrane preparation. A, The effet of GTP on CT-stimulated adenylate ylase. The onentrations of CT used were whole toxin 15 JLg/ml (e-e) and "A" subunit 5 JLg/ml (-). The basal ativities without CT or GTP, and those with CT but without GTP ranged from pmol/mg protein/3 min. Absissa shows different GTP onentrations in a log sale. DT (1 m M) was omitted when the "A" subunit of CT was used. B, The effet of Gpp[NH]p or GTP alone without CT on the membrane adenylate ylase preparation. Absissa indiates onentrations of GTP ( - ) or the analog Gpp[NH]p (e----e) in a log sale. No CT was added. The basal ontrol value without guanine nuleotides was 24.5 pmol/mg protein/3 min. In both A and B, the ordinate indiates adenylate ylase ativity as desribed in Materials and Methods. 3 6 (V)... 'Q) 1 ; +-'..... )... Q)... ;;:.=flp- gl Jg/ml FIG 7. ffet of CT on the membrane preparation. The CT used was whole toxin (e-e) or "A" subunit ( - ). Other additions to the reation mixture were GTP (1 JLM), NAD (1.25 mm), and DT (1 mm). In the ase of holera toxin "A" subunit ativation, DT was omitted. The average adenylate ylase ativity without both CT and exogenous GTP and that without CT but with GTP were about 2 and 33 pmol/mg protein/3 min, respetively. The inset shows gross (maro) onentration effets of CT whih was a separate experiment. ah value represents an average of dupliate determinations. M Q,) o 'o +-' :rr..... en... a.. Vol. 81, No. 2 I I I I I I M FIG 8. ffet of NAD on the adenylate ylase stimulators. The absissa indiates the different onentrations of N AD (M) in a log sale and the ordinate indiates adenylate ylase ativity. The onentration of stimulators was NaF ( ) 1 mm and Gpp[NH]p (-) 1 JLM, and adenylate ylase ativities at NAD onentration were 58 and 48 pmol/mg protein/3 min, respetively. Basal ontrol ativities (without CT, NaF, or Gpp[NH]p but with different onentrations of NAD as indiated) are shown as -. ah value represents an average of tripliate determinations. tration. The stimulatory effet by GTP alone (without CT) was minimal, while that by the analog alone was learly dosedependent (Fig 6B). The CT stimulation of adenylate ylase (with 2 p.m GTP) was also dose-dependent (Fig 7), but 1 p.g/ ml CT was not effetive in this system. The CT stimulation with the homogenate assay system appears to be less sensitive than that with the in vitro floating system (f. Fig 2). Table I shows the result of ombinations of various stimulators suh as NaF (1 mm), Gpp[NH]p (1 p.m), CT (15 p.g/ ml), and GTP (1 p.m) on adenylate ylase with the membrane assay system. The onentrations of stimulators were used at maximum stimulatory doses to examine possible additive effets. With this homogenate assay system, one annot demonstrate a hormone effet (reeptor site), and all stimulators in Table I at through the nuleotide binding protein(s) (N) and not through t he reeptor site (R). Table I shows that the ombinations ofnaf and guanine nuleotide (GTP or Gpp[NH] p) were not additive: the fat suggests that both at on N with a ommon adenylate ylase (C). The effet of Gpp[NH]p was inhibited by GTP, suggesting that Gpp[NH]p and GTP bind to the same site on N. Table I shows that the ombination of N af and CT gives an additive stimulation. For the CT stimulation with the homogenate assay system, we always added 3 ofators suh as GTP,

5 Aug.1983 CHOLRA TOXIN STIMULATION OF PIDRMAL ADNYLAT CYCLAS 135 NAD, and DT. Therefore, we further examined whether or not one of the ofators might have enhaned NaF stimulation (Table II). It is obvious that NAD (1.25 rnm) strongly enhaned the NaF stimulation. As shown in Fig 8, only the NaF stimulation is enhaned by NAD at rather high onentrations. The basal adenylate ylase- and Gpp[NH]p-stimulated ativities were not affeted at all. DISCUSSION Although the N protein was suggested as a omponent of the adenylate ylase system 1 years ago [3], its signifiane was unknown until reently [1,2,5,6,1]. One typial example that exemplifies the obligatory role of N is the fmding that in mutants of S49lymphoma ells laking this N, hormonal stimulation did not ativate adenylate ylase and the addition ofn (from another strain) restored the hormonal effet, resulting in AMP aumulation [6]. In the present study we have demonstrated unequivoally the presene of N in epidermal adenylate ylase systems. Our experiments with the membrane fration showed diret ations of GTP analog and of CT (plus GTP) with the N-C omplex. With the epidermal slie system we found that modifiations of N via CT altered the hormonal responses of epidermal adenylate ylase. With the floating (intat ell) system, the CT stimulation of adenylate ylase required exposure for more than 1 h, while the stimulation was immediate with the membrane (broken ell) system. CT is omposed of "A" and "B" subunits; the "A" subunit penetrates further into the ell membrane [31] while the "B" subunits remain on the ell surfae and are bound to the putative CT reeptor. The initial time lag with our floating system is probably due to the time required for "A" subunit separation, penetration, and subsequent ativation of N. Thus far we an ount only a few epidermal surfae reeptors suh as those related to adenylate ylase systems [15-18) and epidermal growth fator reeptors [32]. Perhaps now we an add this CT reeptor to the above list as one of the ell surfae reeptors of skin. A peuliar finding in the CT ation with pig epidermis is the lesser sensitivity of the membrane system as ompared with that of the floating system. Generally, in other tissues the membrane system is so sensitive that only nanogram quantities of CT are required for ativation. This insensitivity may be due to a defiieny of another ofator alled "ytosol fator" for CT ativation [1,33]. Beause ofthis possibility, we have added skin extrat, whih was separated by Sephadex G 15, to the assay mixture. The addition did not inrease the sensitivity of the membrane system, so an explanation for the insensitivity is still laking. The ativation of purified GTP-regulatory protein by fluoride ion requires the presene of magnesium ion and ATP [34]. Sternweis and Gilman [35] just reently reported that an aluminum ontamination of ATP but not ATP itself is the atual fator for the ativation. Signifiant amounts of ::Jurninum ion were also found in other ommerial nuleotides suh as ADP and GTP. Our unexpeted finding that the ombination ofnaf and NAD ativates maximally adenylate ylase ativities may be explained by this newly disovered aluminum effet. The data that the ativation was evident only at high NAD onentration (Fig 8) suggest that the effetive omponent is a ontaminant suh as aluminum rather than NAD itself. Aluminum ion does not appear to ativate then protein in the absene of NaF [35]. This is also onsistent with qur observation that the high onentration of NAD itself did not enhane the stimulation by the GTP analog (Fig 8). The membrane preparation in our study was best prepared using a glass-teflon homogenizer for a short period. Under this ondition, the N-C omplex is well preserved. Use of a glassglass homogenizer, even under mild onditions, led to a loss of the omplex. Our rude membrane preparation also ontains the R unit sine we an measure the.8-adrenergi reeptor with a radiolabeled antagonist (unpublished observation). Despite the fat that the membrane preparation ontains all 3 basi omponents for the hormone ation (R, N, and C), epinephrine no longer auses AMP aumulation one the epidermal ell has been mehanially destroyed. The data suggest that for the oupling of the entire adenylate ylase system, ertain strutural (mehanial) supports and/or other ofator(s) may be needed. In order to keep the hormone responsiveness, we are now developing a new assay system with epidermal ells that are prepared by trypsinization followed by osmoti shok. Studies with suh "leaky ells," in whih all 3 omponents are still oupled and yet phosphorylated ofators an be inorporated, are urrently under way. RFRNCS l. Rodbell M: The role of hormone reeptors and GTP-regulatory proteins in membrane transdution. Nature 284:17-21, Pfeuffer T: GTP-binding proteins in membranes and the ontrol of adenylate ylase ativity. J Bioi Chern 252: , Limbird L, Gill DM, Lefkowitz RJ: Agonist-promoted oupling of the,8-adrenergi reeptor with the guanine nuleotide regulatory protein of the adenylate ylase system. Pro Nat! Aad Si USA 77: , Stadel JM, Shorr RGL, Limbird L, Lefkowitz RJ: videne that a,8-adrenergi reeptor-assoiated guanine nuleotide regulatory protein onveys guanosine 5'--(3-thiotriphosphate) dependen t adenylate ylase ativity. J Bioi Chern 256: , Citri U, Shramm M : Resolution, reonstitution and kinetis of the primary ation of a hormone reeptor. Nature 287:297-3, Howlett AC, Sternweis PC, Maik BA, Van Arsdale PM, Gilman AG: Reonstitution of ateholamine-sensitive adenylate ylase. J Bioi Chem 254: , Cassel D, Selinger Z: Mehanism of adenylate ylase ativation through the,8-adrenergi reeptor: ateholamine indued displaement of bound GDP by GTP. Pro Nat! Aad Si USA 75: , Cassel D, Selinger Z: Cateholamine-stimulated GTPase ativity in turkey erythroyte membranes. Biohim Biophys Ata 452: , Shramm M, Rodbell M: A persistent ative state of the adenylate ylase system produed by the ombined ations of isoproterenol and guanylyl irnidodiphosphate in frog erythroyte membrane. J Bioi Chem 25: , Gill DM: Mehanism of ation of holeratoxin. Adv Cyli Nuleotide Res 8:85-118, Cassel D, Pfeuffer T: Mehanism of holeratoxin ation: ovalent modifiation of the guanyl nuleotide-binding protein of the adenylate ylase system. Pro Nat! Aad Si USA 75: , Cassel D, Selinger A: Mehanism of adenylate ylase ativation by holeratoxin. Inhibition of GTP hydrolysis at the regulatory site. Pro Nat! Aad Si USA 74: , Lad PM, Nielsen TB, Preston MS, Rodbell M: The role of the guanine nuleotide exhange reation in the regulation of the adrenergi reeptor and in the ations of aeteholamine and holeratox.in on adenylate ylase in turkey erythroyte membranes. J Biol Chern 255: , Burns DL, Moss J, Vaughan M: Choleragen-stimulated release of guanyl nuleotide from turkey erythroyte membranes. J Bioi Chern 257:32-34, Yoshikawa K, Adahi K, Halprin KM, Levine V: The effet of ateholamine and related ompounds on the adenylate ylase system in the epidermis. Br J Dermatol 93:29-36, Adahi K, Yoshikawa K, Halprin KM, Levine V: Prostaglandins and AMP in epidermis. Br J Dermatol92: , Iizuka H, Adahi K, Halprin KM, Levine V: Histamine (H2) reeptor-adenylate ylase system in pig skin (epidermis). Biohim Biophys Ata 437:15-157, Iizuka H, Adahi K, Halprin KM, Levine V: Adenosine and adenine nuleotide stimulation of skin (epidermal) adenylate ylase. Biohim Biophys Ata 444: , Adahi K, Iizuka H, Halprin KM, Levine V: Speifi refratoriness of adenylate ylase in skin to epinephrine, prostaglandin, histamine and AMP. Biohim Biophys Ata 497: , imer! S, Neufeld G, Korner M, Shramm M: Funtional implantation of a solubilized,8-adrenergi reeptor in the membrane of a <: ll. Pro Nat! Aad Si USA 77:76-764, Green H: Cyli AMP in relation to proliferation of the epidermal ell: a new view. Cell 15:81-811, Marelo CL: Differential effets of AMP and GMP on in vitro epidermal ell growth. xp Cell Res 12:21-21, Okada N, Kitano Y, Ihihara K: ffets of holera toxin on proliferation of ultured human keratinoytes in relation to intraellular yli AMP levels. J Invest D ermatol 79:42-47, 1982

6 136 CONNOR AND LOW 24. Steiner AL, Parker CW, Kipnis DM: Radioimmunoassay for yli nuleotides. I. Preparation of antibodies and iodinated yli nuleotides. J Bioi Chern 247: , Adahi K, Iizuka H, Halprin KM, Levine V: pidermal AMP is not dereased in psoriasis lesions. J Invest Dermatol 74:74-76, Lowry OH, Rosebrough NJ, Farr AL, Randall RJ: Protein measurement with the Folin phenol reagent. J Bioi Chern 193: , Marks F, Rebien W: The seond messenger system of mouse epidermis. I. Properties and,8-adrenergi ativation of adenylate ylase in vitro. Biohim Biophys Ata 284: , Salomon Y: Adenylate ylase assay. Adv Cyli Nuleotide Res 1:35-55, Nikols GA, Brooker G: Potentiation of holera toxin stimulated AMP prodution in ultured ells by inhibitors of RNA and protein synthesis. J Bioi Chern 255:23-26, Rodbell M, Birnbaumer L, Pohl SL, Krans HMJ: The gluagon- Vol. 81, No.2 sensitive adenylate ylase system in plasma membrane of rat liver. J Bioi Chern 246: , Wisnieski BJ, Barnhall JS: Photolabelling of holera toxin subunits during membrane penetration. Nature 289: , Haigler H, Ash JF, Singer SJ, Cohen S: Visualization by fluoresene of the binding and internalization of epidermal growth fator in human arinoma ell A-431. Pro Natl Aad Si USA 75: , Shleifer LS, Kahn la, Hanski, Northrup JK, Sternweis PC, Gilman AG: Requirement for holera toxin dependent ADPribosylation of the purified regulatory omponent of adenylate ylase. J Bioi Chern 257:2-23, Sternweis PC, Northup JK, Smigel MD, Gilman AG: The regulatory omponent of adenylate ylase. Purif.ation and properties. J Bioi Chern 256: , Sternweis PC, Gilman AG: Aluminum: a requirement for ativation of the regulatory omponent of adenylate ylase by fluoride. Pro Nat! Aad Si USA 79: , X/ 83/ $2./ TH JOURNAL OF INVSTIGATIV DRMATOLOGY, 81: , 1983 Copyright 1983 by The Williams & Wilkins Co. Vol. 81, No.2 Printed in U.S.A. pidermal Pyridoxal 5'-Phosphate Depletion, Inhibition of DNA Synthesis, and Inhibition of the xpression of Ornithine Dearboxylase Ativity by the Vitamin B-6 Antagonist 4'-Deoxypyridoxine MICHAL J. CONNOR, PH.D. AND NICHOLAS J. LOW, M.D., M.R.C.P. Division of Dermatology, UCLA Shool of Mediine, Los Angeles, California, U.S.A. The indution of ornithine dearboxylase (ODC) ativity may be an essential omponent of skin tumor promotion. ODC requires pyridoxal5'-phosphate (PLP) as a ofator. We have measured the epidermal PLP onentration and investigated its relationship to DNA synthesis and ODC ativity in the hairless mouse. The epidermal PLP onentration was approximately g/ g. When tape-stripping was used to indue ODC ativity in the epidermis the onentration of PLP was signifiantly elevated 4.5 h later at the time of peak ODC ativity and when DNA synthesis was redued. Systemi treatment with the vitamin B-6 antagonist 4'-deoxypyridoxine (4DOP) signifiantly redued the epidermal PLP onentration and DNA synthesis. The ODC ativity indued in the epidermis 4.5 h after tape-stripping in 4-DOP-treated mie was only 17% of that indued in untreated tape-stripped ontrols. In in vitro experiments it was shown that while 4-DOP does not inhibit ODC ativity, a major metabolite of 4-DOP, 4-DOP-phosphate (K;.6 mm), does. In mixing experiments it was shown that the epidermal extrats from 4-DOP-treated mie did not ontain signifiant amounts of ODC inhibi- Manusript reeived January 25, 1983; aepted for publiation April 7, This work was supported by grants from the Derma to logi Researh Foundation of California. Reprint requests to: Dr. Mihael J. Connor, Division of Dermatology, UCLA Shool of Mediine, Los Angeles, California 924. Abbreviations: 4-DOP: 4'-deoxypyridoxine 4-DOP-phosphate: 4'-deoxypyridoxine-phosphate ODC: ornithine dearboxylase (C , L-ornithine arboxylase) PLP: pyridoxal 5' -phosphate tors. 4-DOP may inhibit ODC indution in the epidermis by depleting the PLP ontent. The pyridoxine (vitamin B-6) derivative, pyridoxal 5' -phosphate (PLP), is a ofator for several mammalian dearboxylases inluding the polyamine synthesis enzyme ornithine dearboxylase (ODC). ODC ativity is indued in mouse skin by tumor promoters [1] and polyamine levels are elevated in proliferating. tissues (2]. Current evidene indiates that the indution of ODC ativity, while not neessarily speifi for tumor promotion, may be an essential omponent of the proess (3-5]. If the indution of ODC ativity is essential for arinogenesis, then inhibitors of ODC should be good andidates as antitumor agents. In this respet the irreversible, highly speifi (suiide) inhibitor of ODC, alpha-difluoromethyl-ornithine, has reently been shown to inhibit tumor promotion in mouse skin [5,6]. Hepatoarinogenesis is redued [7,8] and the growth of hepatomas severely impaired [9] in pyridoxine defiieny. The pyridoxine antagonist 4'-deoxypyridoxine (4-DOP) irihibits tumor growth [1] and has been used in the treatment of adult lymphati leukelnia [11]. Tissue putresine and spermidine levels are redued in pyridoxine defiieny [12], and the PLP onentration may therefore regulate putresine synthesis. Murray and Frosio [13] found that redued levels of ODC ativity were indued in the epiderlnis of pyridoxine-defiient ompared to pyridoxine-replete Inie after treatment with the tumor promoter tetradeanoylphorbol-13-aetate. Thus epidermal ODC ativity is influened in vivo by the pyridoxine status and this may explain in part the antianer and antiproliferative properties of pyridoxine defiieny. We have been investigating the possible use of 4-DOP for regulating the supply of PLP to the epidermis to ontrol ODC ativity more diretly, and to establish a new route for inhibiting