Molecular and CellularBiochemistry 100: 9-23, 1991. © 1991KluwerAcademic Publishers. Printedin the Netherlands. Original Article

Modulation of insulin induced ornithine decarboxylase by putrescine and methylputrescines in H-35 hepatoma cells Judith Frydman, Oscar Ruiz, Eduardo Robetto, Juan M. Dellacha and Rosalia B. Frydman Facultad de Farmacia y Bioqufmica, Universidad de Buenos Aires, Junin 956, Buenos Aires, Argentina Received20 November1989; accepted19 March 1990

Key words: polyamines, methylputrescines, hepatoma cell line, ornithine decarboxylase regulation, insulin

Summary The effect of several methylputrescines on the activity of insulin-induced ornithine decarboxylase (ODC) was examined in H-35 hepatoma cells. The induction involved both protein and m-RNA synthesis. Actinomycin D inhibited O D C activity when given up to 1 h after insulin treatment. When added to the medium 2 h or 3 h after the insulin, the activity was increased 100% and 80% respectively. Insulin-induced ODC from H-35 cells had a biphasic half-life, a shorter one of 46 min and a longer one of 90 min. 1-Methylputrescine and 2-methylputrescine were found to be competitive inhibitors of the ODC from H-35 cells with Ki values of 2.8 and 0.1 mM respectively. Putrescine itself was found to have a Ki = 2.4 mM. N-Methylputrescine was a very poor inhibitor of the cell free ODC while 1,4-dimethylputrescine did not show any inhibitory effect. When cellular ODC activity was measured, the four methylputrescines assayed as well as putrescine entirely abolished its activity in the H-35 cells when given at a 1 mM concentration together with insulin. 1-Methylputrescine and 1,4-dimethylputrescine abolished 60% of the activity at a 0.1tzM concentration. All the methylputrescines given at 0.1 mM concentrations decreased the putrescine content of the stimulated cells to the levels found in quiescent cells, but only 1-methyl and 2-methylputrescines decreased spermidine and spermine content. 1,4-Dimethyl and 1-methylputrescines showed a strong inhibition of ODC synthesis, while the other diamines were less inhibitory. At concentrations that abolished ODC activity, 1,4-dimethylputrescine decreased 70% of the total immunoreactive ODC bands, while 1-methyl and 2-methylputrescine decreased them by 50%, and N-methylputrescine and putrescine decreased them by 20%. The lack of decrease in immuno-reactive ODC with the latter two compounds was mainly due to the appearance of immunoreactive degradation products of O D C of low molecular weight. Putrescine and N-methylputrescine affected protein synthesis to a small extent in stimulated cells, while 1-methylputrescine decreased it to the level of non-stimulated cells. Insulin (1/zM concentration) stimulated D N A synthesis in the cells, and this stimulation was doubled in the presence of 2-methylputrescine or putrescine. It can be concluded that, among the methylputrescines assayed, 2-methylputrescine was the best inhibitor of cell-free ODC activity, while 1,4-dimethylputrescine and 1-methylputrescine were the best inhibitors of cellular ODC activity. Abbreviations: O D C - Ornithine Decarboxylase, T L C - T h i n Layer Chromatography, D N E M - D u l b e c c o ' s Modified Essential Medium, PBS - Phosphate Buffered saline, PEG - Polyethyleneglycol. Enzyme. Ornithine decarboxylase (EC 4.1.1.17)

10

H2N~INH2 PUTRESClNE~

HzN~NHCH3 N-METHYLPUTRESCINE 2

CH3

CH3

H 2 N ~ N H

2

I-METHYLPUTRESCINE ~ H2N~

2 H21,..~ A v ~,,,~/NH v

2-METHYLPUTRESCINE NH2

1,4-DIMETHYLPUTRE$CIN5E Fig. 1. Structures of putrescine and of the methylputrescines used in this study.

Introduction It is well established that polyamines (putrescine, spermidine and spermine) as well as their biosynthetic enzymes increase in proliferating tissues and are essential for all growth and cell division processes [1, 2]. In the regulation of the polyamine biosynthetic pathway, ornithine decarboxylase (ODC) - the first enzyme of the pathway - has a fundamental regulatory role. It is present in very small amounts in normal non-proliferating tissues and in quiescent cells, and its activity is greatly increased and modulated by the exposure of the organism or the cells to trophic stimuli [3]. In addition of being highly inducible it is an enzyme with the shortest half-life known, strictly regulated by many metabolites and products of polyamine biosynthesis and by an inducible proteic inhibitor called antizyme [4] which is involved in the modulation of its half-life and turnover [5]. ODC is a permanent target for studies on the inhibition of the proliferative and neoplasic processes. The inhibitors comprise ornithine and polyamine derivatives and are usually of three types. The 'suicide' or mechanism-based enzyme inhibitors among which the substrate analogue difluoromethylornithine has achieved benchmark status, and product analogues of the 1,4-diaminealkyne type [6, 7]. To the second type belong the polyamine analogues which do not directly inhibit the enzyme but

exert a feed-back repression; among them the diaminealkanes were extensively studied [8], while lately N-alkyl and N,N'-dialkylspermidines and spermines have gained widespread attention [9, 10]. Finally, are the substrate and product analogues which are competitive inhibitors of ODC; among them 2-methylornithine [11] as well as the N-alkyl, 1-alkyl and 2-alkylputrescines [12, 13]. Among the latter we found that the methylputrescines were the best in vivo inhibitors of liver ODC in rats treated with the hepatocarcinogen thioacetamide or with the glucocorticoid dexamethasone. It was also found that, although the methylputrescines were efficient in vivo inhibitors of the ODC from rat liver, they did not markedly deplet the liver polyamine pools except for its putrescine content [13]. It was therefore of interest to examin their effect on the Reuber H-35 rat hepatoma cell line in order to gain further insight into their mode of action. These cells were chosen as a potential model of a growing liver cell since they are one of the few cell lines in culture known to respond to physiological concentrations of a growth factor such as insulin, and since this hepatoma cell line is known to have a number of liver-like functions [14, 15]. The study revealed that N-methyl 2 and 1,4dimethylputrescine 5 (Fig. 1) did not inhibit the activity of the ODC isolated from the cells, while 1-methyl 3 and 2-methylputrescine 4 as well as putrescine 1 itself were competitive inhibitors of the enzyme. All the methylputrescines were however, good inhibitors of ODC activity when added to the cell cultures. All of them decreased ODC synthesis, as well as total protein synthesis with the exception of putrescine and N-methylputrescine 2 which affected the latter to a small extent.

Experimental procedure Materials

Synthetic procedures were used to obtain N-methylputrescine 2 [16], 1-methylputrescine 3 [17], 2-methylputrescine 4 [18] as well as 1,4-dimethylputrescine 5 (2,5-diaminehexane) (Fig. 1). The diamines were purified as their bis- (benzyloxycarbo-

11 nyl) derivatives and were liberated from the latter by hydrogenolysis in the presence of hydrochloric acid. The hydrochlorides obtained were hygroscopic and their purity was checked by TLC on cellulose coated plates (Merck, Darmstadt) with isopropanol : acetic acid : water (8 : 3 : 1) as developing solvent. They were spotted with a ninhydrin spray. Putrescine dihydrochloride, spermidine trihydrochloride, spermine tetrahydrochloride, pyridoxal-phosphate, dithiothreitol (DTT), phenylmethylsulphonylfluoride (PMSF), polyethyleneglycol (PEG), EDTA, and dansyl chloride were purchased from Sigma Chem. Co. Glucagon-free insulin was from either Hoechst or Sigma. L-[14C]ornithine (specific radioactivity 52-59 mCi/mmol), [3H-methyl]-thymidine (18.2Ci/mmol), [14C]-leucine (300mCi/mmol)and Protosol were obtained from New England Nuclear Corp. (Boston, MA). [35S]-Methionine (1151 Ci-mmol) was from Amersham. All other reagents were of analytical grade. Silica gel plates (0.25 mm thickness) as well as the solvents were from Merck (Darmstadt). Vectastain ABC reagent was from Vector Labs. (Burlingame).

Tissue culture materials Dulbecco's modified minimal essential medium (DMEM), fetal bovine serum, and Tryptan Blue were from Gibco (Grand Island, NY). Compounds to be added to cell cultures were dissolved in phosphate buffered saline (pH 7.4) (PBS) and sterilized by filtration through 0.22/zm pore size membrane filters (Millipore SA.).

Methods Cell culture Regularly 400,000 cells were cultured as monolayers in 75 cm 2 plastic flascs (NUNC, Denmark), in Dulbecco's minimal essential medium supplemented with 10 mM Hepes, glutamine (0.3 g/l), sodium bicarbonate (0.9g/I), penicilline (2.5 x 105UI/1), streptomycine (0,05g/I) and fetal bovine serum (10%). The cells were routinely grown under an atmosphere of 5% CO2: 95% air at 37°C. The medium was renewed every 48 h. Upon reaching

conftuency the cells were subcultured by trypsin transfer (2.5 g trypsin/1 and 0.2 g/1 EDTA, pH 7.5). The cell suspension was then diluted 1 : 1 0 and plated in either 60 or 100 mm tissue culture dishes and incubated as described above and additional supplement of 1.3 g/l sodium bicarbonate. The cell number was determined by trypsinizing and dispersing the cells before being counted in a hemocytometer chamber. Viability Was determined using Trypan blue.

Induction of ODC activity To study the induction of ODC, cells at 70-80% of confluence were used to avoid the inhibitory effect that confluency might exert on cell proliferation. The cultures were serum deprived and kept in the DMEM for 24-48 h prior to their use. At the end of this period cells were rinsed once and refurbished with fresh serum-free medium containing the insulin at a 10 -6 M concentration. At the indicated time intervals the plates of H-35 hepatoma cells were harvested. They were rinsed with PBS and harvested with 0.4-0.7 ml of a solution containing 50mM phosphate buffer (pH 7.4), 5mM NaF, 0.1% PEG, 0.1mM EDTA, l m M PMFS, l m M DTT and 60/zM pyridoxal phosphate (ODC) buffer). The cells were then disrupted by sonication and the homogenates were centrifuged at 12,000 x g for 10 min in an Eppendorf microfuge. The supernatants thus obtained were used for enzyme assays either with or without dialysis.

Assay for ornithine decarboxylase (ODC) activity ODC activity was determined as described elsewhere [13]. The incubation mixture contained, in a final volume of 130 ~1, 50 mM Tris-HCl buffer (pH 7.4) 0.1mM EDTA, 5mM dithiothreital, 50/zM pyridoxal phosphate, 5 mM NaF, 0.2% PEG and 50 ~1 of enzyme (2 mg protein/ml). The reaction was started by the addition of 0.6raM L-[1-14C]ornithine (65,000 dpm). Incubation were run for 60 rain at 37 ° C with constant shaking. The reaction was stopped by addition of 100/zl of 4 M citric acid and the incubation was continued for additional 60rain with shaking. The 14CO2 released was trapped by Protosol adsorbed on Whatman filter strips. The latter were placed in 10 ml of an Om-

12 nifluor Toluene scintillation solution (New England Nuclear) and counted with a liquid scintillation counter. One unit of enzyme activity was defined as the amount catalyzing the formation of 1 nmol of CO2/h at 37° C. Protein was determined by the method of Bradford [19] using bovine serum albumin as a standard.

Polyamine analysis For the preparation of the perchloric extracts, cells were washed twice with PBS and 0.3 M perchloric acid was added to the washed cell pellets which were then disrupted by sonication with a microtip. All the manipulations were performed between 0-4 ° C. After 2 h at 4 ° C the precipitates were pelleted by centrifugation in a microfuge. The supernatants were analyzed for polyamine content by dansylation following the described techniques [20[. The dansyl derivatives were separated and identified by TLC on silica gel plates using two solvent systems; chloroform: triethylamine, 9 : 1 (solvent A) and cyclohexane : ethyl acetate, 3 : 2 (solvent B). The dansylated polyamines were scrapped off the plates, eluted with benzene and the fluorescence was determined in an Aminco Bowman fluorometer. Standard concentration curves of the dansylated natural and synthetic polyamines used in this study were run simultaneously with each of the polyamine analysis, and the amounts were calculated from standard curves run under identical conditions. Each sample was dansylated with the inclusion of appropriate amines as standards.

Determination of the half-life of O DC activity The half-life of ODC activity was measured by inhibiting the protein synthesis with cycloheximide (50/~g/ml). Insulin was used to induce ODC of the H-35 hepatoma cells maintained in DMEM. After a 4 h induction period, the cells were rinsed and refed with fresh MEM to which cycloheximide (50 t~g/ml) was added. Samples were harvested at different time intervals after the addition of cycloheximide and enzyme activities were assayed as described above. The time required after the addition of cycloheximide for a 50% decrease of the induced O D C activity was defined as the half-life

(tl/2) of the enzyme. Cell extracts were prepared from at least three experiments in duplicate and the half-lives were calculated by the least-squares method.

Determination of ODC protein and of total soluble protein synthesis in H-35 cell cultures The apparent rate of ODC synthesis was determined by measuring the incorporation of [35S]methionine into the enzyme. The labelled aminoacid (50 ~Ci) was added to 4 ml of culture in the induction medium and incubated for 30min at 37°C under 5% CO2. The medium was then aspirated and the cells were washed with an ice-cold solution of 25 mM Tris-HC1 buffer (pH 7.5) containing 10mM L-methionine, 0.1mM EDTA, l mM dithiothreitol and 0.2M sucrose. The cells were suspended in 0.3 ml of ODC buffer and homogenized and centrifuged as described above. In order to measure the radio-activity incorporated into the cellular soluble protein, aliquots (10~1) were spotted on Whatman 3 MM paper strips (1 x 1 cm), the strips were boiled for 3 min in 10% trichloroacetic acid and were then washed succesively with ethanol, acetone and ether. They were dried, and radioactivity was measured in a toluene scintillant containing 0.4% diphenyloxazole. When leucine was used to measure protein synthesis, an identical procedure to that described above was employed, except that [35S]-methionine was substituted by [14C]-leucine. The remaining portions of the supernatants (200 ~g of protein) were used for determinations of the radioactivity incorporated into the newly synthesized ODC. The supernatants (equal amounts of extracts), were incubated with lt~g of anti-ODC immunoglobulin at room temperature for 2h with gentle shaking. After that time formalin-fixed S. aureus strain Cowan I cells (10 mg wet weight) were added and the incubations were continued with shaking either for additional 4 h or overnight. The mixtures were then centrifuged at 12,000g for 5min and the precipitates were washed four times with 1 ml each of 25 mM Tris-HCl buffer (pH 7.5) which contained 0.1% sodium dodecylsulfate (SDS), 0.1% Triton X-100, 0.1% deoxycholate (DOCA), 2mM EDTA and 1 mM dithiothreitol (DTI~). The pellets were then

13 dissolved in 100/A of 100 mM Tris-HC1 buffer (pH 6.8) containing 4.6% SDS, 10% 2-mercaptoethanol and 20% glycerol. The samples were heated for 5 min at 100° C. After centrifugation, the supernatants were applied to SDS-polyacrylamide slab gel electrophoresis according to Laemmli [21]. After electrophoresis the gels were impregnated with diphenyloxazole and dried under vacuum. The ODC was visualized by fluorography [22]. After localization, the O D C spots were integrated and the labelled enzyme bands were cut and counted for radioactivity in the toluene scintillant.

Immunobloning. Slab SDS-PAGE was performed in a vertical unit as described above. An aliquot of the cell extracts (200/xg of protein) was mixed with an equal volumen of cracking buffer. The mixture was incubated in a boiling water bath for 5 min. After electrophoresis, protein were visualized by staining with silver nitrate as described by Morrissey [23]. For the immunoblot experiments, the samples (equal amounts of extracts, 100/xl, 200/zg protein) were subjected to electrophoresis under non-denaturing conditions or SDS-PAGE as described above. A replica of the gel was prepared by electrophoretic transfer (2 h at 1 A) of the proteins to nitrocellulose paper in buffer Tris-HC125 mM (pH 8.3), glycine 150mM, and 0.1% SDS. After completion of the transfer the nitrocellulose sheets were washed with TBS (Tris 50 mM pH 7.5, NaC1 150 raM), and blocked with 3% BSA, 2% glycine, 1 mM PMSF in TBS for one hour at room temperature. The filters were then probed with anti-ODC rabbit serum/diluted 1 : 300 to 1 : 700 in 0.2% BSA, 2% glycine, 1 mM PMSF and 0.02% azide in TBS. Unbound antibody was washed with 3 changes of TBS containing 0.05% Nonidet P-40 for 30min each. Anti-ODC rabbit immunoglobulin-ODC immune complexes were detected as described in the procedure provided by the Vecta Stain ABC reagent manufacturers (Vector Labs Burlingame) and peroxidase activity was detected by incubation with 3,3'-diaminobenzidine in the presence of H202. Dot blot analysis was performed by using gravity flow. The samples (50/xl of the cell extracts) were loaded onto prewetted nitrocellulose paper in

a BRL Hybridot apparatus. The filter was then treated as described above for immunodetection in Western blots. Separation of ODC and antizyme was carried out in a Sephadex G-75 column (1 x 20 cm) in the presence of buffer and 250 mM NaCt as described [24]. Antizyme activity was assayed by adding 100/xl of the Sephadex G-75 fractions to known amounts of a partially purified rat liver ODC as suggested by Heller et al. [25].

Analysis of the effects of the methylputrescines on DNA synthesis The H-35 cells cultured as described above were maintained in serum free DMEM for 24-48 h to induce quiescence. The DNA synthesis was determined by [3H]-thymidine incorporation. The latter was estimated by precipitation with perchloric acid of cells that were treated with a 30 rain pulse of [3H]-thymidine (50/zCi/ml) 23.5 h after the insulin treatment. When the effect of the putrescine analogues was assayed, the compounds were added at a 0.5 mM concentration together with the insulin and 30 min before addition of [3H]-thymidine. Tryplicate dishes were run for each assay and for the controls. The cells were washed twice with PBS and twice with TCA containing unlabelled thymidine (0.2mg/ml). Incorporation of [3H]-thymidine in treated cultures was compared to incorporation in control cultures with a similar final cell density (4.3 x 105 cells). Results

Time course of ornithine decarboxylase (ODC) induction in hepatoma H-35 cells The rate of induction of ODC by insulin was measured by the increase of its enzymatic activity. Addition of the hormone (1 ~M) to serum deprived H-35 cells (quiescent cells) resulted in a marked increase of ODC activity which showed a net peak between 4 and 5 h after the addition of the inducer (Fig. 2). Under similar conditions addition of fetal bovine serum increased the enzymatic activity to about 20% of the activity induced by the hormone, while the basal activity of the cells in the DMEM was very low (0.1-0.2 unit/mg protein). It is known

14

7,0

6.0

.=_

o

e~

3£ o~

E c

4.0 I--

2.C

V- 5.0

>

¢D

I-(..)

(.~ o'-'

5.0

IJ)

o o

1.0

2.0

1.0

"~

0

I

2

5

4

TIME,

hours

5

6

c

[

I

T

I

Ao

At

Az

A3

Fig. 3. Time course of the effect of actinomycin D on the H-35

(ODC) induced by insulin in H-35 hepatoma cells. The H-35 cells kept in DMEM were serum-deprived for 24 h; they were then rinsed and refed with fresh medium containing 1/xM insulin. Individual plates (4 ml) of the ceils were harvested at the indicated times and assayed for ODC activity as described in Methods. The results are the mean values of three runs in triplicate and the variation was 10% within experiments.

cell ODC activity induced by insulin. The H-35 cells were induced with insulin as indicated in Methods (I). Actinomycin (4 ~g/ml) was added either together with the inducer (Ao) or 1 (A 0, 2 (A2) or 3 h (A3) after insulin treatment. All the cells were harvested 4 h after induction and the ODC activity was assayed as described in Methods. Control cells (C) were kept in the MEM during the same period (4 h). ODC activity was assayed as described in Methods. The results are the mean of three experiments in triplicate.

that dibutyryl cyclic-AMP, dexamethasone, and asparagine induce ODC activity in H-35 cells which reach a maximum after 4 h [26]. Insulin induction of ODC was found to be very small 2 h after the addition of the hormone and had decreased to about 60% of its maximum activity 6 h after addition. The mode of ODC induction by insulin in H-35 cells was examined. Actinomycin D blocked the increase in ODC activity when added together with the insulin or 1 h after (Fig. 3). When actinomycin D was added to the medium 2 or 3 h after the addition of the hormone and the cells were harvested 4 h after the start of the induction, ODC activity was increased to almost twice the values obtained in the absence of the antibiotic (Fig. 3). These data strongly suggest that the induction effect of insulin

is due to a de n o v o synthesis of the mRNA of ODC and that this synthesis is very likely complete in the first 2 h after addition of the hormone. Once the mRNA is synthesized the antibiotic produces a great increase in the activity of ODC. This could be due to changes in either the stability or the translational efficiency of the mRNA. It is also possible that the great increase in ODC activity is due to an inhibition by actinomycin D of proteolytic enzymes which degrade ODC. Alternatively, the mRNA could also be stabilized if its degradation is inhibited by the antibiotic. Cycloheximide (50/xg/ml) inhibited ODC induction when it was added to the insulin treated cultures up to 2 h after the addition of the hormone. When added after that time, it inhibited about 70% of the enzymatic activity. An explanation of these

Fig. 2. Time course of the activity of ornithine decarboxylase

15 results could be found in the biphasic half life-time of the enzyme (see below).

ODC turnover ODC is an enzyme with an unusual short half-life in most animal cells, a fact which results in fast changes in the content of the enzyme [1]. It was reported that in transformed cells the ornithine decarboxylases have longer half-lives than in the corresponding normal cells [27, 28]. It was therefore of interest to compare the half-life of the ODC from H-35 cells with that of the enzyme isolated from livers of thioacetamide treated rats [13]. This was done by inhibition of protein synthesis in the cells with cycloheximide (50/xg/ml) administrated at different intervals. The insulin induced ODC activity showed a biphasic decay with half-lives of 46 rain and 90 min. Such a biphasic ODC decay following the treatment with cycloheximide indicates that insulin is able to produce at least two forms of ODC in the H-35 cells. A similar biphasic ODC decay following cycloheximide treatment was found in livers of dexamethasone or thioacetamide treated rats. The shorter half-life of rat liver ODC was 24 min [13]; therefore, the degradation rate of the cell enzyme is half of that of the liver enzyme. It is however very similar to the half-life of the ODC induced by dibutyryl cyclic-AMP in the same hepatoma cell line [26]. When cycloheximide was added 5 to 10 min before the cell harvest, a 25-30% increase of ODC activity was found as compared to controls (data not shown), probably due to an inhibition of the protein breakdown processes [29].

Table 1. Inhibitor constants of putrescine and of methylputrescines for ornithine decarboxylase from H-35 cells

Inhibitor

Ki (mM)

Putrescine 1 1-Methylputrescine 3 2-Methylputrescine 4 N-Methylputrescine 2

2.4 2.8 0.1 30.0

The K i constants were calculated from the data obtained by replotting the Kmappvalues vs concentrations of the inhibitors.

Effect of putrescine and of methylputrescines on the in vitro activity of ODC from H-35 cells The effect of the methylputrescines 2-5 (Fig. 1) on ODC from H-35 cells was examined in view of their inhibitory effect on the liver enzyme from dexamethasone and thioacetamide treated rats [12, !3]. The ODC from the insulin treated cells had a specific activity of 3.7 + 0.3 units/rag protein. The Kmapp determined for the enzyme was 0.25 mM, lower than the Kmappdetermined for the rat liver enzyme (0.35-0.4mM). N-Methyl 2, 1-methyl 3 and 2-methylputrescine 4 were found to be competitive inhibitors of the ODC from H-35 cells with respect to ornithine. 2-Methylputrescine 4 was the best inhibitor among them, while putrescine and 1-methylputrescine 3 were less inhibitory and Nmethylputrescine 2 inhibited only 20% of the activity at a 20 mM concentration. 1,4-Dimethylputrescine 5 did not inhibit ODC activity even at a 25 mM concentration. The Ki's of the inhibitory methylputrescines are given in Table 1. The Ki for putrescine was lower than the Ki found for the liver enzyme (2.4mM as compared to 3.3 mM). 1-Methylputrescine 3 inhibited the enzyme from both sources to a similar extent (see Table 1 and ref. 13) while 2-methylputrescine 4 was ten times more inhibitory of the hepatoma enzyme than of the liver enzyme (K~= 1.0 mM).

Effect of methylputrescines on the activity of ODC in H-35 cells The effect of putrescine and methylputrescines on ODC activity in H-35 cells is shown in Fig. 4. When added together with insulin, 1-methyl-putrescine 3 and 1,4-dimethylputrescine 5 were the best inhibitors among those tested since they inhibited 60% of ODC activity at a 0.1/~M concentration. Putrescine I was very effective only at a 10/~M concentration but not at lower concentrations. It showed a biphasic inhibitory effect as a function of concentration. Although 2-methylputrescine 4 was the best in vitro inhibitor (Table 1) it was a weak inhibitor in the in vivo assays (Fig. 4). Just the opposite was

16

o°t _J

0 n~ FZ

o

8O

LI_ 0

r

>2 6C

F-

~- 40

o

a o

20k

L

4C

20 ---M

0L U

i

i

1.0 ANALOGUE

I

IO

I00

1,000

CONCENTRATION , juM

Fig. 4. Dose response curves showing the effect of increasing concentrations of putrescine (@), 1-methylputrescine (A), 2methylputrescine (A), 1,4-dimethylputrescine (O) and N-methylputrescine (*) on ODC activity in H-35 cell cultures. The inhibitor was added together with the insulin and the cells were harvested 4 h after induction of the ODC. The enzyme was obtained and assayed as described in Methods. Each point is the average of two runs in triplicate and vary by less than 10%.

true for 1,4-dimethylputrescine 5. T h e latter was not inhibitory w h e n assayed directly on the e n z y m e but strongly decreased O D C activity w h e n assayed directly on the cells. H e n c e , the latter belongs to the c a t e g o r y of n o n - c o m p e t i t i v e O D C inhibitors and was recently s h o w n to inhibit O D C activity in H e L a cells [30]. T o eliminate the possibility that the a f o r e m e n tioned in vivo inhibitions w e r e due to a direct inhibition o f the H-35 O D C , the cells of several plates which had b e e n i n d u c e d with insulin in the presence of the inhibitors (at a 10 .4 M c o n c e n t r a t i o n ) were p o o l e d and the h o m o g e n a t e s ( p r e p a r e d as described in M e t h o d s ) were dialyzed to eliminate the a d d e d diamines. W h e n O D C activity was then assayed, essentially the s a m e inhibitions w e r e f o u n d as for the non-dialyzed extracts. W h e n the rate of inhibition of O D C activity by the a b o v e m e n t i o n e d diamines was m e a s u r e d , it

OL 0

I 2 3 4 TIME AFTER DIAMINE ADDITION,hours

Fig. 5. Time course of the effect of putrescine and of methylputrescines on the activity of ornithine decarboxylase. Putrescine (@), 1-methylputrescine (A), 2-methylputrescine (A), 1,4-dimethylputrescine (O) and N-methylputrescine (*) at 10 4M concentrations were added either together (time = 4 h) or 1 h (time = 3 h), 2 h (time = 2 h) or 3 h (time = 1 h) after the insulin (I/xM) addition. In the case of 1-methyl and 1,4-dimethylputrescine an addition at 3.55 h was also made. The cells were harvested 4 h after induction and the enzyme was obtained and assayed as described in Methods. ODC activity of insulininduced cells was taken as 100%. The values are the average of three runs in duplicate and varied less than 10% within experiments.

was f o u n d that the strongest inhibitory effect was exerted w h e n t h e y were a d d e d t o g e t h e r with the insulin and that the inhibitory effect decreased w h e n they were a d d e d 1 to 3 h after addition of the h o r m o n e (Fig. 5). W h e n given 1 h before harvest and 3 h after the insulin, O D C activity was still strongly inhibited by 1-methylputrescine 3 (85% inhibition), while putrescine I given u n d e r identical conditions was a w e a k inhibitor (Fig. 5). T h e inhibitory effects described a b o v e could be due to the induction of antizyme or to the reduction in the a m o u n t of O D C present in the cells. T h e synthesis o f O D C was investigated by measuring the i n c o r p o r a t i o n of [3sS]-methionine into O D C

17 and by carrying out immunoprecipitations of the enzyme with polyclonal antibodies. [35S]-Labelled cell extracts were prepared, immunoprecipitated and subjected to SDS-PAGE as described in Methods. Gel autoradiography exhibited a major radioactive band with a molecular mass of 52 kDa corresponding to the ODC subunits (Fig. 6). This band was present in the insulin induced cells (Fig. 6, lane A) as well as in the cells treated with the diamines at 1 mM concentrations (Fig. 6, lanes B, C, D, E and F). The immunoprecipitated ODC bands obtained from the diamine treated cells were of weaker intensity than the insulin induced control (Fig. 6). At the aforementioned concentration the diamines suppressed all the ODC activity in the cells. When the amount of incorporated [35S]-methionine into the 52 kDa immunoreactive ODC was measured, it was found that 1,4-dimethylputrescine 5 exerted the strongest inhibitory effect on ODC synthesis and was followed by 1-methylputrescine 3, N-methylputrescine 2 and putrescine I itself. 2-Methylputrescine 4 exerted the weakest inhibition on ODC synthesis (Table 2). When the amount of the total immunoreactive ODC protein was investigated by making use of dot blots, it was found by semiquantitative integration Table 2. Effect of the methylputrescines on the de novo synthesis of ODC in H-35 hepatoma cells

Treatment

(35S)-Methionine incorporation into ODC (% of control)

Insulin Insulin + Insulin + Insulin + Insulin + Insulin +

100 45 43 40 62 36

putrescine 1 N-methylputrescine 2 1-methylputrescine 3 2-methylputrescine 4 1,4-dimethylputrescine 5

The cells were stimulated with insulin and the indicated diamines ( l m M ) were added together with the inducer as described in Methods. (3~S)-Methionine (50/~Ci) was added to each dish 30 min before harvest. The cells were disrupted and the ODC was immunoprecipitated and processed as described in Methods. The 52kDa radioactive bands were cut-out, dissolved with Protosol and counted. The incorporation of (35S)methionine into ODC of the insulin induced H-35 cells was taken as 100%. The data are the average of three experiments in duplicate which had a variation of less than 10%.

Fig. 6. Effects of putrescine and of methylputrescines on the de novo ODC synthesis in H-35 cells. The diamines (1 mM) were added together with insulin. (35S)-Methionine (50 txCi/dish) was added 3.5h and left in the incubation medium for another 30 rain. The cells were then disrupted and the labelled ODC was immunoprecipitated and analyzed by SDS-polyacrylamide slab gel electrophoresis and fluorography as described in Methods. Lane A: Insulint reated cells. Lane B: Insulin plus putrescine. Lane C: Insulin plus N-methylputrescine. Lane F: Insulin plus 1,4-dimethylputrescine. The molecular mass markers used were bovine serum albumin (68 kDa), ovalbumin (45 kDa), 3-phosphoglyceraldehyde dehydrogenase (36kDa) and carbonic anhydrase (29 kDa). Runs were carried out with pooled extracts of three experiments in duplicate.

of the immunoreactive spots that, if the amount of the enzyme induced by insulin was taken as 100%, the addition of 1,4-dimethylputrescine 5 decreased the immunoreactive enzyme by about 70% while the other C-methylputrescines 3 and 4 gave inhibition values between 50 and 60%. Putrescine i and N-methylputrescine 2 only decreased about 20% of the immunoreactive ODC protein (Fig. 7A). The presence of immunoreactive ODC-protein bands

18

o

lOO

W r~ Z

_z BC _.l U')

_z l.z_ 0

~ 6o

03 Ld

4o Z >O3 Z

~ z0 0 rr" 13._

Fig. 7. Effect of putrescine and methylputrescines on the immunoreactive ODC. (A) Dot blot analysis performed as described in Methods. The cells were treated with insulin (C); and insulin plus l mM of putrescine 1; N-methylputrescine 2, 1-methylputrescine 3, 2-methylputrescine 4 and 1,4-dimethylputrescine 5. (B) Western blots of the ODC immunoreactive proteins, were performed and detected as described in the Experimental Section. Lanes 1-7, from right to left, represent cells grown in: (1) MEM, (2) in the presence of insulin, (3) insulin plus putrescine, (4) insulin plus N-methylputrescine, (5) insulin plus 1methylputrescine, (6) insulin plus 2-methylputrescine and (7) insulin plus 1,4-dirnethylputrescine. Runs were carried out with pooled extracts of three experiments in duplicate.

was then investigated by Western blots (Fig. 7B). As can be seen, in the insulin induced cells (lane 2) the ODC was mainly present as the 52 kDa monomer although several minor faint bands of lower molecular weight also appeared. They were also present in the cells kept in DMEM where the 52 kDh band was completely absent (lane 1). In the cells were putrescine and N-methylputrescine were added (lanes 3 and 4) there was a marked decrease in the 52 kDa ODC band while the lower molecular immunoreactive ODC bands (Mr 22,000 _+ 3,000 and 13,000 _+ 2,000) were much increased. In the cells treated with 1-methylputrescine 3, the 52 kDa ODC band was also decreased (lane 5). 1,4-Dimethylputrescine 5 produced a great decrease in all the ODC immunoreactive bands. The lower molecular weight proteins arose from the ODC 52 kDa

C

.I

I+CH I+[

I+2

1'3_ I+4

1+5

Fig. 8. Effect of putrescine and of methylputrescine on protein synthesis in H-35 cells. Cells were serum deprived during 24 h in DMEM and induced with insulin as described. (35S)-Methionine (50~Ci) was added to each dish 3.5h after insulin and its incorporation into proteins was measured as described in Methods. The incorporation of (35S)-methionine into total protein of insulin induced cells was taken as 100%. Putrescine 1 and the methylputrescines 2-5 (see Fig. 1) (1 mM) were added together with the insulin C): contol; I): insulin induced ceils; I + CH): insulin plus cycloheximide. The data are the mean of three runs in duplicate.

subunit, a fact which was established by [3H]DFMO labelling and by activity measurements and are most likely degradation products of the enzyme (Frydman, J and Frydman, R.B., unpublished results). The appearance of ODC-immunoreactive proteins with lower molecular weights and the increase of the latter found in the cells treated with putrescine and N-methylputrescine falls in line with the induction of antizyme by the two diamines, since the complex of the former with ODC was shown to be degraded faster than ODC itself [5]. ODC immunoreactive bands with lower Mr than 52 kDa were also found in the in vitro translation of the ODC-mRNA and were considered to be premature termination proteins [31, 32]. Faint bands of smaller Mr than 52 kDa were also detect-

19 ed when the degradation of prelabelled ODC in CHO cells by polyamines was studied [33].

Effect of methylputrescines on the polyamine content of the H-35 cells The methylputrescines given to the cells at a 0.1 mM concentration together with the insulin decreased the putrescine levels to thouse found in the quiescent cells, l-Methyl 3 and 2-methylputrescine 4 decreased the spermidine and spermine contents while N-methyl 2 and 1,4-dimethylputrescine 5 were weaker inhibitors of the former (Table 3). All the methylputrescines entered the hepatoma cells, and about 20-36% of the added compounds could still be detected in the cells 4 h after addition (Table 3). In general it can be stated that the intracellular level of the exogenous methylputrescines in the H-35 cells was lower than that found in the livers of thioacetamide treated rats. However, they affected more the polyamine levels in the cell system than in the whole animal [13].

Effect of the methylputrescines on protein and DNA synthesis in the H-35 cells The insulin treatment increased protein synthesis in H-35 cells when compared with the effect of the medium itself (Fig. 8). Addition of cycloheximide

entirely suppressed the incorporation of [35S]methionine or [14C]-leucine (data not shown). Addition of putrescine slightly increased protein synthesis, while N-methylputrescine 2 decreased protein synthesis by about 20%. When the C-methylputrescines were given together with the insulin they completely blocked the stimulation of protein synthesis evoked by the latter (Fig. 8). Since 1methyl 3 and 2-methylputrescine 4 inhibits spermidine and spermine synthesis (see above), and since it is known that the latter affect ribosome assembly in polyribosomes in hepatoma cells [34], it is conceivable that 3 and 4 should also blocked protein synthesis. The behaviour of 1,4-dimethylputrescine 5 does not fit however in this rather simple explanation. The weak inhibitory effect of N-methylputrescine 2 on protein synthesis can be correlated not only with its rather weak inhibition of spermidine and spermine content, but also with its known ability to substitute for putrescine in maintaining protein synthesis [35]. The effect of methylputrescines on DNA synthesis was also explored. Insulin itself strongly promoted the incorporation of [3H]-thymidine into DNA (Table 4). The enhancement of DNA synthesis caused by the insulin was very strong during the first 24 h of growth, which was the doubling period for the cells under the conditions of the assay. When N-methyl 2 and 1-methylputrescine 3 were added together with the insulin no major change in the thymidine incorporation was found,

Table 3. Effect of putrescine and methylputrescines on the polyamine content of H-35 cells

Treatment

Polyamine content (nmol/107 cells) Putrescine

-

26

Insulin

Insulin + Insulin + Insulin + Insulin + Insulin +

putrescine 1 N-methylputrescine 2 1-methylputrescine 3 2-methylputrescine 4 1,4-dimethylputrescine 5

50 330 21 ]8 24 21

(100) (660) (42) (36) (48) (42)

Spermidine

Spermine

190 270 189 183 ]40 162 180

165 365 310 288 197 201 259

(100) (70) (68) (52) (60) (67)

Methyl-Putrescine -

(100) (85) (79) (54) (55) (71)

360 260 200 300

The cells (10 7cells) were induced with insulin and the indicated diamines were added together with the inducer. The cells were harvested 4 h after treatment and the soluble polyamines were extracted and measured as described in Methods. The data are the average of three determinations which agreed within 10%. The percentage of polyamines is given in brackets. The polyamine content of the insulin treated cells were taken as 100%.

20 while the incorporation of the latter was strongly enhanced by putrescine and by 2-methylputrescine 4.

Discussion

It was known that insulin induces cell proliferation in quiescent H-35 cells [36]. It was also known that it leads to an increase in the activity of two short lived enzymes, ODC and tyrosine aminotransferase of which the latter is induced by a mechanism that does not involve new m R N A synthesis [37, 38]. The results reported in this paper show that the increase in O D C activity produced by insulin involves m R N A synthesis. Hence, insulin induces ODC in these hepatoma cells at a transcriptional level as it does in primary cultured hepatocytes [39]. The great increase observed in the ODC activity in stimulated cells when actinomycin D was added 2 or 3 h after the addition of the hormone (Fig. 3) is similar to the result obtained with the same antibiotic when it was added to density inhibited 3T3 mouse fibroblast cultures 2 or 3 h after stimulation [40]. This 'superinduction' of ODC was attributed by the latter author to a stabilization of the enzyme due to a decrease in its degradation rate. More recently a similar increase in ODC induction by actinomycin D was observed in ras transfected 3T3 cells [28]. In this case an increase in mRNA was detected and it was attributed to the possible presence of labile proteins which could be involved in the stability of ODC-mRNA or in its translation. If such an explanation is valid for the H-35 cells, remains to be demonstrated. It has been shown that the unusual short half-life of ODC from animal cells increases when the cells undergo either viral or tumorogenic transformations [27, 41, 42]. The finding that O D C from hepatoma H-35 cells has a half-life of 46 min (see Resuits) while O D C from rat liver has a half-life of 24 min [13] falls in line with the above mentioned results. In addition, a biphasic decay was found for O D C activity in cycloheximide treated H-35 ceils (see Results) similar to the biphasic decay detected for the liver enzyme when cycloheximide was given to dexamethasone [13] or diethyl maleate [43]

treated rats. These results strongly suggest that ODC is present in the hepatoma cells as well as in rat liver in more than one form, each with a different turnover rate and very likely with different molecular features; v.g., one form could be the phosphorylated enzyme [44, 45]. Although it has been reported [40, 46] that the half-life of ODC varies with the growth cycle of the cells, it is important to state that both half-lives of the enzyme were found in the H-35 cells grown under identical growth conditions. It is known that diaminoalkanes inhibit ODC activity and affect its synthesis and degradation [8]. These effects are complex and it is still impossible to draw general conclusions about the relation between structure and activity in these compounds. Thus, 1,3-diaminopropane, 1,3-diamino-2-propanol, 1,5-diaminopentane, 1,6-diaminohexane and 1,7-diaminoheptane decrease ODC activity in vivo and in cell cultures and this inhibition was partially explained by induction of antizyme. They are devoid however of inhibitory effect on ODC itself and their in vivo effects are undoubtedly more complex than merely antizyme induction [8]. Competitive inhibitors of ODC on the other hand, apparently lead to a stabilization of the enzyme and thus increase the amount of the ODC protein [8]. In order to obtain a more systematic information Table 4. Effect of insulin and diamines on the incorporation of (3H)-thymidine into H-35 cells DNA

Addition

(3H)-Thymidine incorporated into DNA (cpm/4.5 x 105 cells)

Control Insulin Insulin + Insulin + Insulin + Insulin +

6,580 38,100 81,100 41,100 42,000 92,900

putrescine 1 N-methylputrescine 2 1-methylputrescine 3 2-methylputrescine 4

+ 1,360 + 4,500 + 13,400 _+ 2,350 + 4,300 + 14,500

The cells were grown and stimulated with insulin as described in Methods and were harvest 24h after insulin addition. (3H)Thymidine was added 30 min before harvest. Putrescine and methylputrescines (1 mM) were added in two portions, together with the insulin and i h before harvest. Estimated doubling time of the cells was 24h. The data are the mean values + SD obtained from six separate experiments.

21 on the effect of diamines on ODC activity we carried out the present study with diamines where the 1,4-diaminobutane structure was kept invariant while small modifications were introduced by placing methyl residues at different positions. The results showed the complexity of the structure-activity relation of these diamines. Thus, N-methylputrescine 2 does not affect the ODC activity in vitro while it decreased the enzyme activity when added to cell cultures (Table 1 and Fig. 4). This decrease in the ODC activity was partially due to a decrease in the ODC protein synthesis (Fig. 6). As is the case with putrescine, it induced formation of antizyme and it increased the degradation of ODC as shown by the formation of lower molecular weight ODC immunoreactive bands (Fig. 7B, lane 4). When the methyl group was placed at the C-1 position, the resulting 1-methylputrescine 3 was found to be a competitive inhibitor of ODC. It also very efficiently increased the decay in ODC activity in the hepatoma cells (Figs. 4 and 5). The decrease in ODC activity was simultaneous with a reduction in ODC synthesis and a decrease in the immunoreactive enzyme (Fig. 7B, lane 4). This diamine did not induce antizyme in the cells. When the 2-methylputrescine 4 was assayed it was found to be a more efficient competitive inhibitor of the cell enzyme than 3. It had however a weaker effect than 3 in inhibitory ODC in the cell cultures (Figs. 4 and 5). It was also found to affect to a smaller degree ODC synthesis (Table 4) and the immunoreactive ODC protein (Fig. 7, lane 6). It did not induce antizyme formation. N-Methylputrescine 2 inhibited more ODC synthesis than total protein synthesis. Both, 3 and 4 inhibited both synthesis to a similar degree (Figs. 3 and 6). When two methyl residues were introduced in the putrescine molecule at C-1 and C-4, the diamine completely lost its inhibitory in vitro effect on the enzyme but was the most potent inhibitor of the ODC activity among the methylputrescines when assayed on the cell cultures. This inhibition was due to a decrease both in ODC synthesis (Table 4) and in the immunoreactive O D C protein. The increase of [3H]-thymidine incorporation into D N A found for putrescine I and 2-methylputrescine 4 (Table 4) is noteworthy since it is known that putrescine itself does not

interact with either DNA or R N A [47]. Their promoting effect on D N A synthesis could be attributed to the in vivo formation of spermidine and spermine in the case of putrescine or their methyl homologue in the case of 2-methylputrescine. The formation of the latter has been demonstrated both in vitro and in chick embryos [48]. N-Methylputrescine and 1-methylputrescine are not aminopropylated neither in vitro nor in vivo to give the spermine derivative. It is well known that the aminopropyl derivatives interact with the above mentioned macromolecules [49, 50].

Acknowledgments We are deeply grateful to Professor Shim-ichiHayashi (School of Medicine, Jikei University, Japan) for a generous gift of polyclonal antibodies. This research was supported by the National Institutes of Health under Grant GM-11973 and by the Consejo Nacional de Investigaciones Cientificas y T6cnicas (CONICET). J.F. and E.R. are grateful to CONICET for fellowships.

References 1. Tabor CW, Tabor H: Polyamines, Ann Rev Biochem 53: 749-790, 1984 2. Bachrach U, Kaye A, Chayen R: Advances in Polyamine Research, Raven Press, (eds.) New York, 1983, 4 3. McCann PP: Regulation of ornithine decarboxylase in eukaryotes. In: JM Gaugas (ed) Polyamines in Biomedical Research. J. Wiley and Sons, Chichester, 1980, pp 109-123 4. Fong WF, Heller JS, Canellakis ES: The appearance of an ornithine decarboxylase inhibitory protein upon the addition of putrescine to cell cultures. Biochim Biophys Acta 428: 456--462, 1976 5. Murakami Y, Hayashi S: Role of antizyme in degradation of ornithine decarboxylase in HTC cells. Biochem J 226: 893-900, 1985 6. Metcalf BB, Bey P, Danzin C, Jung MJ, Casara P, Veverti JP: Catalytic irreversible inhibition of mammalian ornithine decarboxylase. J Am Chem Soc 100: 2551-2553, 1978 7. Danzin C, Bey P, Schirlin D, Claverie ct-Monofluoromethyl and ct-difluoromethyl putrescine as ornithine decarboxylase inhibitors: In vitro and in vivo biochemical properties. Biochem Pharmac 31: 3871-3878, 1982 8. Danzin Ch, Mamont PS: Polyamine inhibition in vivo and

22

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20. 21.

22.

23.

24.

in organ growth and repair. In: PP McCann, AE Pegg, A Sjoerdsma (eds.). Inhibition of Polyamine Metabolism, Academic Press, New York, 1987, pp. 141-164. Porter CW, McMannis J, Casero RA, Bergeron RJ: The relative abilities ofbis(ethyl)derivatives of putrescine, spermidine and spermine to regulate polyamine biosynthesis and inhibit cell growth. Cancer Res 47: 2821-2827, 1987 Bergeron RJ, Neims AH, McManis JS, Hawthorne ThR, Vinson JRT, Portell R, Ingeno MJ: Synthetic polyamine analogues as antineoplastics. J Med Chem 31: 1183-1190, 1988 Mamont PS, Bohlen P, McCann PP, Bey P, Schuber F, Tardif C: a-Methyl ornithine, a potent competitive inhibitor of ornithine decarboxylase, blocks proliferation of rat hepatoma cells in culture. Proc Natl Sci, USA 73; 16261630, 1976 Ruiz O, Alonso Garrido DO, Buldain G, Frydman RB: Effect of N-alkyl and C-alkylputrescines on the activity of ornithine decarboxylase from rat liver and E. coli. Biochim Biophys Acta 873, 53-61, 1986 Ruiz O, Buldain G, Alonso Garrido DO, Frydman RB: Interaction of alkylputrescines with ornithine decarboxylase from rat liver and E. coli: an in vitro and in vivo study. Biochim Biophys Acta 954: 114-125, 1988 Kelley DS, Becker JE, Potter VR: Effect of insulin, dexamethasone and glucagon on the amino acid transport ability of four rat hepatoma cell lines and rat hepatocytes in culture. Cancer Res 38: 4591-4600, 1978 Munns TW, Johnston MF, Liszewski MK, Olson RE: Vitamin K-dependent synthesis and modification of precursor prothrombin in cultured H-35 hepatoma cells. Proc Natl Acad Sci, USA 73: 2803-2807, 1976 Alonso Garrido DO, Buldain G, Frydman B: Concerning the synthesis of N-methylputrescine and its homologues. J Org Chem 49: 2021-2023, 1984 Alonso Garrido DO, Buldain G, Frydman B: 1,4-Diaminoalkanes from pyrroles. A new synthetic approach to substituted putrescines. J Org Chem 49: 2619-2622, 1984 Alonso Garrido DO, Buldain G, Ojea MI, Frydman B: Synthesis of 2-alkylputrescines from 3-alkylpyrroles. J Org Chem 53: 403-407, 1988 Bradford M: Rapid and sensitive method for quantitation of microgram quantities of protein utilising the principle of protein-dye binding. Anal Biochem 72: 248-254, 1976 Dreyfuss G, Duir R, Harell A, Chayen R: Determination of polyamines in urine. Clin Chim Acta 49: 65-72, 1973 Laemmli UK: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680-685, 1970 Bonner WM, Laskey RA: A film detection method for tritium labelled proteins and nucleic acids in polyacrylamide gels. Eur J Biochem 46: 83-88, 1974 Morrissey JH: Silver stain for proteins in polyacrylamide gels: a modified procedure with enhanced uniform sensitivity. Anal Biochem 117: 307-310, 1981 McCann PP, Tardif C, Mamont PS: Regulation of ornithine

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

decarboxylase by ODC-antizyme in HTC cells. Biochim Biophys Res Commun 75: 948-954, 1977 Heller JS, Chen KY, Kyriakidis DA, Fong WF, Canellakis ES: The modulation of the induction of ornithine decarboxylase by spermine, spermidine and diamines. J Cell Physiol 96: 225-234, 1978 Chen KY, Liu AYC: Differences in the mode of regulation of ornithine decarboxylase and tyrosine amino-transferase in H-35 rat hepatoma cells shown by varying the medium composition. Biochim Biophys Acta 755: 244-252, 1983 Bachrach U: Polyamines and neoplastic growth: stabilization of ornithine decarboxylase during transformation. Biochim Biophys Res Commun 72: 1008-1013, 1976 Hollta E, Sistonen L, Alitalo K: The mechanism of ornithine decarboxylase deregulation in C-Ha-ras oncogene transformed NIH 3T3 cells. J Biol Chem 263: 4500-4507, 1988 Epstein D, Elias-Bishko S, Hershko A: Requirement for protein synthesis in the regulation of protein breakdown in cultured hepatoma cells. Biochemistry 14: 5199-5204, 1975 Holm I, Persson L, Heby O, Seiler N: Feedback regulation of polyamine synthesis in Ehrlich ascites tumor cells. Analysis using non-metabolizable derivatives of putrescine and spermine. Biochim Biophys Acta, 972: 239-248, 1988 Kameji T, Pegg AE: Inhibition of translation of mRNAs for ornithine decarboxylase and s-adenosylmethionine decarboxylase by polyamines. J Biol Chem 262: 2427-2430, 1986 Persson L, Holm I, Heby O: Regulation of ornithine decarboxylase mRNA translation by polyamines. J Biol Chem 263: 3528-3533, 1988 Holtta E, Pohjanpelto P: Control of ornithine decarboxylase in chinese hamster ovary cells by polyamines. J Biol Chem 261: 9502-9508, 1986 Rudkin BB, Mamont PS, Seiler N: Decreased proteinsynthesis activity is an early consequence of spermidine depletion in rat hepatoma tissue culture cells. Biochem J 217: 731-741, 1984 Goldemberg SH, Algranati ID, Miret JJ, Alonso Garrido DO, Frydman B: Effect of putrescine derivatives on growth and protein synthesis. In: U Bachrach, A Kaye, R Chayen (eds.) Raven Press, New York. Advances in Polyamine Research, vol. 4, 1984, pp. 233-244. Mottola C, Czech MP: The type II insulin-like growth factor receptor does not mediate increased DNA synthesis in H-35 hepatoma cells. J Biol Chem 259: 12705-12713, 1984 Granner DK, Hangrove JL: Regulation of the synthesis of tyrosine aminotransferase; the relationship to mRNA TAT. MO1 Cell Biochem 53/54, 113-128, 1983 Reel JR, Lee KL, Kenney FT: Regulation of tyrosine -ketoglutarate transaminase in rat liver. IX Studies of the mechanism of hormonal induction in cultured hepatoma cells. J Biol Chem 245:5806--5812 Hiramatsu Y, Oomoto H, Eguchi K, Sekiba K: Regulation of ornithine decarboxylase by various hormones and aminoacids in primary cultured hepatocytes (Abstr,) Interna-

23

40.

41.

42.

43.

44.

45.

tional Conference on Polyamines in Life Science, Lake Yamanaka, Japan, 1986, pp. 173-174 Clark JL: Specific induction of ornithine decarboxylase in 3T3 mouse fibroblasts by pituitary growth factors: celldensity dependent biphasic response and alteration of half life. Biochemistry 13: 4668-4674, 1974 Astrup EG, Boutwell RK: Ornithine decarboxylase activity in chemically induced mouse skin papillomas. Carcinogenesis 3: 303-308, 1982 Poso H, Karvonen E, Svomalainen H, Anderson LC: A human neuroblastoma cell line with an altered ornithine decarboxylase. J Biol Chem 259: 12307-12310, 1984 Numazawa S, Oguro T, Yoshida T, Karoiva Y: The mode of induction of ornithine decarboxylase by diethylmaleate (Abstr.) International Conference on Polyamines in Life Science, Lake Yamanaka, Japan, 1986, pp. 187-188 Mitchell JLA, Qasba P, Stofko RE, Franzen MA: Ornithine decarboxylase modification and polyamine-stimulated enzyme inactivation in HTC cells. Biochem J 228: 297308, 1985 Worth LL, Haddox MK: Evidence for activation of ornithine decarboxylase by differential phosphorylation (Abstr.) J Cell Biol 107: 855a

46. Hogan BLM, Murden S: Effect of growth conditions on the activity of ornithine decarboxylase in cultured hepatoma cells. J Cell Physiol 83: 345-352, 1974 47. Frydman B, Frydman RB, de los Santos C, Alonso Garrido DO, Goldemberg SH, Algranati IA: Putrescine distribution in E. coli studied in vivo by 13C-NMR. Biochim Binphys Acta 805: 337-344, 1984 48. Sarhan S, Dezeuve F, Seiler N: Putrescine derivatives as substrates of spermidine synthase. Int J Biochem 19: 10371047, 1987 49. Feuerstein BG. Marton LJ: Specificity and binding in polyamine nucleic acid interaction. In: U Bachrach, YM Heimer (eds.) The Physiology of Polyamines, vol. I, CRC Press, Boca Raton, 1989, pp. 109-124 50. Quigley GT, Teeter MM, Rich A: Structural analysis of spermine and magnesium ion binding to yeast phenylalanine transfer RNA. Proc Natl Acad Sci USA, 75: 64-68, 1978

Address for offprints: R.B. Frydman, Facultad de Farmacia y Bioqufmica, Junin 956, Universidad de Buenos Aires, Buenos Aires (1113) Argentina