Two Dist inct Mec hanisms for 0r nithine Decar boxy lase Regulation by Polyamines in Rat Hepatoma Cells PETER P.,,MCCANN,CHANTAL TARDIF, JEAN-MARIE HORNSPERGER AND PETER BOHLEN Centre de Recherche Merrell International, 16, rue d'Ankara 67084-Strasbourg Cedex, France
ABSTRACT
Exogenous diamines and polyamines added to rat hepatoma (HTC) cells in culture rapidly decrease ornithine decarboxylase (ODC) activity. Previous evidence has suggested t h a t these amines act either a t the level of blocking new enzyme synthesis or by the induction of a non-competitive protein inhibitor, termed antizyme, which complexes with ODC to form an inactive complex. With the use of HMOAcells, a recently cloned rat hepatoma cell line that has a greatly stabilized ODC, it has been possible to demonstrate that M of exogenous putrescine blocks the increase in ODC activity, but unlike in the parent HTC cell line, without induction of the antizyme or formation of any inactive ODC-antizyme complex. However, complete blockade of ODC at M putrescine is effected by induction of antizyme and formation of the ODC-antizyme complex, as now evidenced by the isolation of the active enzyme and antizyme components after Sephadex column chromatography in the presence of 250 mM NaC1. These findings indicate clearly that two polyamine-regulatory mechanisms for ODC exist and are separable in this cell line.
Mechanisms for the regulation of ornithine decarboxylase (E.C. 4.1.1.17; ODC), the ratelimiting enzyme which controls polyamine biosynthesis, include transcriptional (McCann et al., '77b), translational (Clark and Fuller, '75), and post-translational models (Sedory and Mitchell, '77; Obenrader and Prouty, '77). Among the most interesting of these mechanisms is that by which the end product of the enzyme reaction, putrescine, regulates the levels of ODC activity. Suppression of ODC activity by diamines had been noted in a number of cell lines (Clark and Fuller, '75; Pett and Ginsberg, '68; Kay and Lindsay, '73) and in vivo in regenerating rat liver (Janne and Holtta, '74; Poso and Janne, '76). The evidence seemed to suggest that the polyamine regulatory effect was post-transcriptional and a t the level of enzyme synthesis (Clark and Fuller, '75; Canellakis and Theoharides, '76; Kallio e t al., '77). Seemingly in opposition to these findings, however, was the discovery of the non-competitive protein inhibitor or antizyme which was induced by the presence of putrescine or other polyamines and specifically interacted with ODC to form an inactive enzyme-antizyme complex (Fong J. CELL.PHYSIOL. (1979) 99: 183-190.
e t al., '76; Heller et al., '76; Heller et al., '77). The presence of this ODC antizyme in several cell lines, in rat liver, and in rat thyroid was subsequently confirmed (Clark and Fuller, '76; McCann e t al., '77a; Jefferson and Pegg, '77; Pegg et al., '78; Poso e t al., '78; Friedman et al., '77). However, the antizyme could not be detected in several other r a t tissues, including kidney, prostate and heart, after various diamine treatments, even though ODC was totally blocked (Pegg et al., '78). Moreover, the decrease of immuno-reactive ODC protein seen after amine treatment in HTC cells (Canellakis and Theoharides, '76) and regenerating rat liver (Kallio e t al., '77) also suggested t h a t another ODC regulatory mechanism existed other than the induction of the antiz yme. We have been able to use HMOA cells, a clone of HTC cells with a greatly stabilized ODC (McCann e t al., '77b), as a probe to investigate the polyamine regulation of ornithine decarboxylase. We have found that the antizyme mechanism is the means by which ODC Received Oct. 4, '78.Accepted Dec. 12, '78.
183
184
P. P. MCCANN, C. TARDIF, J:M.
HORNSPERGER AND P. BOHLEN
is completely blocked by high concentrations of exogenous putrescine in these cells. However, when HMOA cells a r e treated by a low concentration of putrescine, ODC activity is still reduced, most likely by a block in new enzyme synthesis, but without induction of antizyme or formation of any ODC-antizyme complex. MATERIALS AND METHODS
Rat hepatoma (HTC1 cells and t h e clone of cells (HMO*) isolated from t h e parent HTC cell line as partially resistant to DL-a-methyl ornithine (Mamont e t al., '78) were grown in spinner culture in Swim's 77 medium supplemented with 10% calf serum as previously described (McCann e t al., '75; Mamont e t al., '76). Ornithine de, .rboxylase activity was measured by the me iod previously described (Mc-
9.0
2ann e t al., '75) in cell extracts, taken and pre3ared as described in t h e figures. Inhibitor or lintizyme activity, where one inhibitor unit is iefined as a n amount which will inhibit one unit of ODC activity i.e., 1 nmole/CO/hr, was lissayed as previously described (Heller et al., '76; McCann e t al., '77a). Cell protein was measured by t h e fluorescamine method (BohLen et al., '72). The half-line (t 1/21 of decrease iin enzyme activity was calculated by linear least-squares reduction. Supernatants of cells described in figures 3, 4 and 5 were prepared from concentrated sonicated cell extracts (1.5 X lo8cells in 10 ml) by centrifugation at 100,000 X g for one hour at 4°C as previously described (McCann et al., '77a). Chromatography and separation of ODC and inhibitor activity on a G-75 superfine Sephadex column in t h e presence of assay 1buffer and 250 mM NaCl was done as previ-
A
B
-
0-0
L
r \
70
0
.-I=
/ /"
/
a
c
2 Q 5.0
\
0/O
n
0 0
/
'O.
Y
/
u
L
n
m A A A \ ~ - A ~
A
Putrescine IO-~M
1.0 ' 0
--or
/"
A/A
0
3.0
Putrescine
lo-' M 1
I
0
2
T I M E
6
(hr)
10
0
2
T I ME
I
6
1
I
to
(hr)
Fig. 1 Effect of cycloheximide, putrescine, and cycloheximide plus putrescine on ornithine decarboxylase activity in HMOA cells. The cells were diluted into fresh medium in spinner culture (1.5 X lo5 cells per ml) and incubated for 12 hours a t 37°C. The cells were then divided into five individual cultures and putrescine and/or cycloheximide were added as below. After a further three hours of incubation aliquots of cells (2.0 x lo6) were then taken a t the indicated times (the first a t time 0) and washed by centrifugation twice with phosphate-buffered saline a t 4"C, and sonicated. After all the samples were taken the sonicates were directly assayed for ODC activity. Activities are means from a t least three different experiments using different cultures. A, Control (0); + M putrescine (A);+ M putrescine ( 0 ) .B, Control (0); + 50 pg/ml cycloheximide (A); 50 pg/ml cycloheximide and lo'* M putrescine ( 0 ) .
+
185
POLYAMINE REGULATORY MECHANISMS FOR ODC
ously described (McCann e t al., '77a) except as noted in the figures. Each assay for inhibitory activity of the column fractions contained a final concentration of 25 mM NaCl since 100 11.1aliquots were tested in a 1-ml assay volume. All control enzyme assays, used to calculate inhibitor concentrations also contained 25 mM NaCl to compensate for the slight inhibitory effect (less than 10%)of NaCl on ODC a t this concentration. When ODC was mixed to complex with free antizyme-inhibitor or when ODC-antizyme complex was already present, whichever was in excess, i.e., ODC or inhibitor, was the only assayable species. Only when column chromatography is done in the presence of 250 mM NaCl can both ODC and inhibitor be detected in the same extract. Bound ornithine decarboxylase inhibitor or antizyme in normal nonpolyamine-treated cells was extracted by the method described by Heller et al. ('77). After two centrifugations a t 100,000 x g in Tris buffer A (50 mM Tris pH 7.3 at 23"C, 50 ml KCL and 1 mM MgClJ the pellets from the cell extracts were stirred for ten minutes in lo-* M putrescine a t room temperature. EDTA (2 mM final concentration) and dithiothreitol (0.77 mg/ml) were added in the manner described (Heller et al., '77). After centrifugation again a t 100,OO x g for one hour, both supernatants and pellets were dialyzed for 15 hour a t 4°C with two buffer changes (500 vol) of Tris buffer B (10 mM Tris pH 7.3 a t 4"C, 50 mM KC1, 1 mM EDTA, 0.05 mM pyridoxal phosphate, 2.5 mM dithiothreitol) and then for four hours in the same buffer without KC1. RESULTS
When HMOAcells were induced to grow by dilution into fresh medium with 10% calf serum ornithine decarboxylase activity greatly increased over a period of hours as shown in figure 1. When cycloheximide was added to HMO* cells (fig. lB), the rise in ODC was blocked and there was a slow decay of enzyme activity. The biological half life of the enzyme (t 1/21 was found to be more than eight hours, in dramatic contrast to the t 112 of 22 minutes for ODC from HTC cells (fig. 2). When M was similarly added to HMOn cells, it also blocked the rise of ODC and a slow decay of enzyme activity was seen for some hours, although to a lesser effect than in the presence of cycloheximide (fig. 1A). This low concentration of putrescine, however, rapidly caused a decrease in ODC activity in HTC cells
CELLS
HMO, 100
CH tlI2=
8.2 hr
0
-
0
2
4
8
6
1 0 h r
I00
HTC CEI 1 S
,
40
put 10%
t,,,= 51 min
20
\A 10
0 '
\ I
60 120 180 m i n Fig. 2 Rates of decrease in ODC activity caused by cycloheximide and putrescine in HMOA cells and HTC cells. Maximal enzyme activity (100%activity) waa taken as the beginning of the phase of linear reduction of ODC activity. The activities for HMOA cells were taken from figure 1. Those for HTC cells were determined separately under the same experimental conditions as described in figure 1. HMOA cells: lo'* M putrescine (0); + 50 pg/ml cycloheximide and M putrescine ( 0 ) ;+ 50 pg/ml cycloheximide (W). HTC cells: + lo-* M putrescine (0); 50 pg/ml cycloheximide (W); M putrescine (A).
+
+
+
with t 1/2 of 51 minutes as shown in figure 2. Also, while a high concentration of putrescine (lo-* M) only moderately altered the t 1/2 of HTC cell ODC (fig. 21, in HMOAcells it had the effect of blocking all ODC (fig. 1A) (t 1/2 = 58 minutes; fig. 2). When M putrescine was added to HMOA cells in the presence of cycloheximide, the decrease of ODC activity was much less rapid than with putrescine alone (t 1/2 more than 3 hours) (fig. 2). We have recently shown that cell extracts from HTC cells which have been treated with M putrescine contain both ODC and antizyme as bound complex. The ODC and antizyme could be separated by chromatography
186
P. P. MCCANN, C. TARDIF, J:M.
8-
HORNSPERGER AND P. BOHLEN
i
.90
t; 4 U
LL
\
.60
w 5
>
N
z W
.30 0 (0
k z 3
0
10
20 FRACTION
30
40
NUMBER
Fig. 3 Chromatography of supernatant from HMOA cells treated for ten hours with loe5 M putrescine after initial stimulation of ODC. HMOA cells were incubated for eight hours after induction of ODC by dilution into fresh medium (1.5 X lo5cellslml). The cells were then incubated for an additional ten hours in the M putrescine. Two milliliters of concentrated cell supernatants (3 X lo7 cells) which were presence of prepared by 100,000 X g centrifugation as described in MATERIALS AND METHODS and which had approximately 20 units of residual ODC activity were layered onto a G-75 superfine Sephadex column and chromatographed with buffer containing 250 mM NaCl as described in MATERIALS AND METHODS. ODC activity ( 0 ) ; inhibitor activity (0).
on a G-75 superfine Sephadex column with assay buffer containing 250 mM NaCl (McCann et al., '77a). (This column procedure normally recovers 35-45%of the ODC and about 60%of the antizyme). When HMOA cells were treated with M for ten hours (twice the time of treatment for the HTC cells), no antizyme a t all could be detected after Sephadex column chromatography in the presence of 250 mM NaCl (fig. 3). Approximately 8 units of ODC were recovered from the original 20 units layered on the column. In contrast, when HMOA cells were treated for ten hours with l o - * M putrescine, they were found to produce antizyme in excess of ODC. The amount of free antizyme-inhibitor produced under these conditions was about 2 units per milligram protein. When such an extract with 4 units of free antizyme was chromatographed with buffer containing 250 mM NaC1, a total of 6 units of antizyme as well as about 0.6 units of ODC were recovered from the column (fig. 4). At least 2 of these 6 units of antizyme and all of the ODC must have been derived from bound antizyme-ODC complex. The fraction of antizyme derived from the bound complex was probably higher than 2 of the 6 units as recovery of the antizyme on the column is normally only 60%and we would only expect a yield of about 2.5 units of the original 4 units if there was no bound antizyme-ODC complex in the cells. The recovery
of 0.6 units of ODC was somewhat lower than would be expected from the amount of antizyme recovered.' When a similar experiment was done with HTC cell extracts with 19 units of free antizyme, 11 or 60%of these units were recovered from the column (fig. 4). Thus, in this second case the free antizyme was equivalent to the total antizyme of the cell extracts. As a control to show that the NaCl treatment would indeed completely dissociate t h e HMOA cell ODC-antizyme complex, we mixed 12.5 units of HMOAcell ODC and 7.5 units of HMOA free antizyme with 5 total units of detectable ODC activity remaining after mixing. We also mixed 10.6 units of HTC cell ODC and 6.6 units of HMOA free antizyme with 4 total units of detectable ODC activity remaining after mixing. When we chromatographed these two different mixtures with buffer containing 250 mM NaCl as shown in figure 5, we recovered in the first instance 8.2 units of antizyme and 7.2 units ODC, and in the second 12.8 units of antizyme and 6.1 units of ODC. In both cases we recovered more antizyme than 1 We have found that the HMOA cell ODC tends to show a loss in activity after NaCl treatment and column chromatography that is proportionately greater than after similar treatment of the HTC cell ODC. On occasion after such treatment the HMOA cell ODC has shown up to a 50%loss in activity after 24 hours at 4°C while under similar conditions the HTC cell ODC would only loose 10.20% of its activity. The multiple peaks of antizyme-inhibitor activity seen in figure 4 and in figure 5 are typical of the earlier results of Heller et al. ('76) and McCann et al. ('77a).
187
POLYAMINE REGULATORY MECHANISMS FOR ODC
HMO,
W
2
>
N
z
CELLS TREATED WITH
I
t
-
'lI
--I
I
P
W LL
I 0 2 M PUTRESCINE
.90
0
-
0
1.2
a \
11
-
I]
D 0
2 0
I
-
z .60 0
v
\
-
.30
To
'0-0
4 I0
the original free form amount and also more ODC than remained after mixing. Antizyme is bound to cell membranes of both cell lines. This antizyme could be isolated from the membrane fractions of sonicated cells by extraction with buffer containing M putrescine (table 1). The amounts found in both cell lines represent only 2.5%t o 7.5% of the total amount of antizyme found in maximally induced cells. HMOA cells contain appreciably less of both bound and free antizyme
0
20
than HTC cells, even after maximal induction of antizyme by treatment of the cells with lo-' M putrescine. Buffer containing M putrescine would not extract the bound antizyme from sonicates of either HMO* or HTC cells. DISCUSSION
Our earlier work on ODC regulation in HTC cells suggested that there might be two separate mechanisms for ODC regulation by
188
P. P. MCCANN. C. TARDIF, J.-M. HORNSPERGER AND P. BOHLEN HMOA C E L L ODC M I X E D WITH HMOA C E L L INHIBITOR
:
1.2
z
0 L
v
a
a
LL
90
W
f
N
z
W
R
.60
I
z
0
D
LL
.30
f? -
z
3
m
1
z
0
-
-
a
HTC CELL ODC MIXED WITH HMO,
CELL INHIBITOR
0
z
.90
0
n
a
D
a
k 0
0
a
-r0
LL 1
-60
z
W
f
N
z W
.30
V
c I LL
0
t z 3
0 10 20 FRACTION
30
40
NUMBER
Fig. 5 Chromatography of NaCl treated ODC-inhibitor complexes. HMOA cell ODC (12.5 units) was mixed with HMOA cell inhibitor (7.5 units) resulting in five total units of detectable OM: activity remaining after mixing. The total mixture (1.5ml) was chromatographed with buffer containing 250 mM NaCl as described in figure 3 and MATERIALS AND METHODS. HTC cell ODC (10.6 units) was mixed with HMOA cell inhibitor (6.6 units) resulting in four total units of detectable ODC activity remaining after mixing. The total mixture (1.5 ml) was chromatographed as above.
polyamines (McCann et al., '77a). One was that proposed by several workers (Clark and Fuller, '75; Canellakis and Theoharides, '76; Kallio et al., '77; Poso et al., '78) whereby ODC synthesis would be regulated by putrescine and/or spermidine. The second mechanism was the induction of the antizyme-inhibitor which directly complexes with the enzyme (Fong et al., '76; Heller et al., '76; McCann e t al., '77a). Our evidence in HTC cells showed M putrescine efficiently decreased that ODC activity without inducing any significant amount of antizyme. However, there was
some antizyme present as inactive ODC-inhibitor complex (McCann et al., '77a). The isolation of the HMOA cells and the discovery that it has a greatly stabilized ODC has given us a powerful tool to more precisely investigate the polyamine regulation of ODC. We are now able to conclude that there are indeed two distinct polyamine-controlled regulatory mechanisms. The slow and partial efM putrescine on HMOAcell ODC is fect of clearly similar to the effect seen when we used the general protein synthesis inhibitor cycloheximide (fig. 1) and suggests that both are
189
POLYAMINE REGULATORY MECHANISMS FOR ODC TABLE 1
Comparison of amounts of ornithine decarboxylase inhibitor found a8 membrane-bound form in sonicates of confluent cells, and as free inhibitor in maximally induced cells Ornithine decarboxylase antizyme-inhibitor Bound and extracted inhibitor Total units 12 x lo8 cells
HTC cells (% of induced cells)
HMOA cells (% of induced cells)
3.68 (2.6%) 2.80 (3.3%)
Specific activity Img protein
0.20 (6.9%) 0.16 (7.5%)
Induced inhibitor Total units 12 x lo8 cells
Specific activity Img protein
141.0
2.93
84.8
2.13
Parallel spinner cultures of HTC and HMOA cells were grown to confluence (8.5 X lo5 to 1.1 x lo6cells/ mi) in Swim's 5-77 medium in the absence of putrescine. The cells 12 x 10"total per culture) were washed in phosphate-buffered saline, sonicated and suspended in 5 ml of Tris buffer A. The extracts were centrifuged a t 100.000 x g for one hour, and then the pellets were resuspended in 5 ml of Tris buffer A and again centrifuged a t 100,000 X g for one hour. Bound ornithine decarboxylase antizyme was extracted from this pellet by 10.' M putrescine and then measured as described in MATERIALS AND METHODS. If antizyme activity could be detected of course no ODC activity could be present. In the non-induced (confluent) cells all the cellular ODC activity was removed in the discarded supernatants of the two 100,000 X g centrifugations done before extraction of the antizyme from the pellets. The inhibitor activities shown represent total activity, and specific activity from the final pellet and supernatant together. Free antizyme was induced in cells t h a t had been diluted to 1.5 X lo5 celldm1 (2 x lo8 total cells per culture) in fresh medium containing 10 mM putrescine, and then incubated a t 37°C for 12 hours. After washing and sonication the amount of antizyme was measured in these cell extracts. Each value represents the mean of a t least two separate determinations with less than 6% variation among the individual determinations.
working by a similar mechanism that is by M publocking new ODC synthesis. When trescine is used to block HMOAcell ODC, the M effect is so rapid as compared t o either putrescine or cycloheximide, t h a t clearly another mechanism is being brought into play. This is induction of the antizyme-inhibitor. It had previously been shown (Fong et al., '76; Heller et al., '76; Poso e t al., '78) that the induction of inhibitor was dependent on new protein synthesis and, indeed, when cycloheximide and M were added together (fig. lB),the immediate blocking effect of l o WM2 putrescine was greatly suppressed. The fact that cycloheximide did not completely block M putrescine effect (fig. 2) can be exthe plained by the fact that some preformed membrane bound antizyme-inhibitor was already present and released by the high concentration of putrescine in situ (table 1). Thus, we can argue that the low concentration of putrescine M) only blocks ODC synthesis and is not sufficient to either induce new antizyme or release the preformed membrane bound inhibitor from HMOA cells. In contrast, lo-* M putrescine induces and releases antizyme, which explains why ODC activity rapidly drops from the partially stable level seen M putrescine after cycloheximide or treatment to zero (fig. 1). Further evidence for this is that unlike in
HTC cells, when concentrated cell extracts M putrescine-treated HMOA cells from were chromatographed with buffer containing 250 mM NaCl, no inhibitor at all could be detected (fig. 3). This strongly suggests the antizyme induction and/or release mechanism is not activated by this concentration of putrescine in HMOA cells. However, the antizyme is clearly responsible for the complete M putrescine, as block of ODC activity by is shown by the results in figure 4. In striking contrast, similarly treated HTC cells seem to contain only free antizyme and no ODC-antizyme complex (fig. 4).This suggests that in HTC cells under the conditions of high exogenous putrescine, the other repressive mechanism was the primary means by which the ODC was blocked. Several controls were done to insure that the lack of the appearance of the inhibitor in HMOA cells, under conditions of treatment M putrescine, was not just the result with of some difference in the ability of the HMOA cell ODC and inhibitor to form a more tightly bound complex that was less sensitive to dissociation by the salt treatment. As is shown in figure 5, the complex is clearly dissociable by 250 mM NaC1. This can also be seen from the results in figure 4. In addition, the HMOAcell antizyme can complex with and dissociate from both its own ODC and that of HTC cells.
190
P. P. MCCANN, C. TARDIF, J.-M. HORNSPERGER AND P. BOHLEN
Other evidence also shows that the equilibria of the non-competitive association of ODC with the antizyme are the same for both cell lines (P. P. McCann, to be published). Finally, as HMO* and HTC cells both contain the same small fraction of bound as compared to fully induced antizyme (table l),the bound form of the antizyme cannot somehow be responsible for the regulatory effect of M putrescine in the HMOAcells. We have also M putresfound that a concentration of cine is not sufficient to cause bound antizyme to be extracted from the membrane fraction in either cell line. In summary, by using the HMO* cell line we have been able t o demonstrate that two different mechanisms exist for the polyamine regulation of ODC. This supports a recent hypothesis that intracellular and exogenous polyamines may regulate ODC by different mechanisms (Heller et al., '78). However, we have not yet been able to demonstrate specifically that polyamines block ODC synthesis, although our evidence and that from other laboratories strongly suggests that this is the case. ACKNOWLEDGMENTS
We thank Doctors J. S. Heller, E. S. Canellakis, H. Poso and J. Janne for making their manuscripts available to us before publication. We also thank Doctors N. Seiler and J. Koch-Weser for their very helpful suggestions with this manuscript. LITERATURE CITED Bijhlen, P., S. Stein, W. Dairman and S. Udenfriend 1972 Fluorometric assay of proteins in the nanogram range. Arch. Biochem. Biophys., 155: 213-220. Cannellakis, 2. N., and T. C. Theoharides 1976 Stimulation of ornithine decarboxylase synthesis and its control by polyamines in regenerating rat liver and cultured rat hepatoma cells. J. Biol. Chem., 251: 4436-4441. Clark, J. L., and J. L. Fuller 1975 Regulation of ornithine decarboxylase in 3T3 cells by putrescine and spermidine: indirect evidence for translational control. Biochemistry, 14: 4403-4409. 1976 Protein inhibitor of ornithine decarboxylase does not account for effect of putrescine on 3T3 cells. Biochem. Biophys. Res. Commun., 73: 785-790. Fong, W. F., J. S. Heller and E. S. Canellakis 1976 The appearance of a n ornithine decarboxylase inhibitory protein upon th e addition of putrescine to cell cultures. Biochim. Biouhvs. Acta. 428: 456-465. Friedman, Y., S. Park, S. Levasseur and G. Burke 1977 Regulation of thyroid ornithine decarboxylase by the
Polyamines. Induction of a protein inhibitor of ornithine decarboxylase by t h e end-products of t h e reaction. Biochim. Biophys. Acta, 500: 291-303. Heller, J. S., W. F. Fong and E. S. Canellakis 1976 Induction of a protein inhibitor to ornithine decarboxylase by the end products of its reaction. Proc. Natl. Acad. Sci. (U.S.A.), 73: 1858-1862. Heller, J. S., K. Y. Chen, D. A. Kyriakidis, W. F. Fong and E. S. Canellakis 1978 The modulation of the induction of ornithine decarboxylase by spermine, spermidine and diamines. J. Cell. Physiol., 96: 225-234. Heller, J. S., D. Kyriakidis, W. F. Fong and E. S. Canellakis 1977 Ornithine decarboxylase antizyme is a normal component of uninduced H-35 cells and ra t liver. Eur. J. Biochem., 81: 545-550. Janne, J., and E. Holtta 1974 Regulation of ornithine decarboxylase activity by putrescine and spermidine in rat liver. Biochem. Biophys. Res. Commun., 61: 449-456. Jefferson, L. S., and A. E. Pegg 1977 Studies on ornithine decarboxylase activity in the isolated perfused rat liver. Biochim. Biophys. Acta, 484: 177-187. Kallio, A., M. Ltifman, H. POSOand J. J h n e 1977 Inhibition of ornithine decarboxylase by diamines in regenerating rat liver. FEBS Lett., 79: 195-199. Kay, J. E., and V. J. Lindsay 1973 Control of ornithine decarboxylase activity in stimulated human lymphocytes by putrescine and spermidine. Biochem. J., 132: 791-796. Mamont, P. S., P. Bohlen, P. P. McCann, P. Bey, F. Schuber andC. Tardif 1976 a-Methyl ornithine, a potent competitive inhibitor of ornithine decarboxylase, blocks proliferation of rat hepatoma cells in culture. Proc. Natl. Acad. Sci. (U.S.A.), 73: 1626-1630. Mamont, P. S., M. C. Duchesne, J. Grove and C. Tardif 1978 Initial characterization of an HTC cell variant partially resistant to the anti-proliferative effect of ornithine decarboxylase inhibitors. Exptl. Cell Res., 115: 387-393. McCann, P. P., C. Tardif and P. S. Mamont 1977a Regulation of ornithine decarboxylase by ODC-antizyme in HTC cells. Biochem. Biophys. Res. Commun., 75: 948-954. McCann, P. P., C. Tardif, M. C. Duchesne and P. S. Mamont 1977b Effect of a-methyl ornithine on ornithine decarboxylase activity of ra t hepatoma cells in culture. Biochem. Biophys. Res. Commun., 76: 893-899. McCann, P. P., C. Tardif, P. S. Mamont and F. Schuber 1975 Biphasic induction of ornithine decarboxylase and putrescine levels in growing HTC cells. Biochem. Biophys. Res. Commun., 64: 336-341. Obenrader, M. F., and W. F. Prouty 1977 Detection of multiple forms of rat liver ornithine decarboxylase. J. Biol. Chem., 252: 2860-2865. Pegg, A. E., C. Conover and A. Wrona 1978 Effects of aliphatic diamines on ra t liver ornithine decarboxylase activity. Biochem. J., 170: 651-660. Pett, D. M., and H. S. Ginsberg 1968 Metabolism of polyamines in KB cells. Fed. Proc., 27: 615. Poso, H., and J. Janne 1976 Inhibition of ornithine decarboxylase activity and spermidine accumulation in regenerating ra t liver. Biochem. Biophys. Res. Commun., 69: 885-892. Pbso, H., S. K. Guha and J. Janne 1978 Stabilization of ornithine decarboxylase in ra t liver. Biochim. Biophys. Acta, 524: 466-473. Sedorv. M. J.. and J. L. A. Mitchell 1977 Remlation of ornithine decarboxylase activity during the phys a rum mitotic cycle. Exptl. Cell Res., 107: 105-110.
ABSTRACT
Exogenous diamines and polyamines added to rat hepatoma (HTC) cells in culture rapidly decrease ornithine decarboxylase (ODC) activity. Previous evidence has suggested t h a t these amines act either a t the level of blocking new enzyme synthesis or by the induction of a non-competitive protein inhibitor, termed antizyme, which complexes with ODC to form an inactive complex. With the use of HMOAcells, a recently cloned rat hepatoma cell line that has a greatly stabilized ODC, it has been possible to demonstrate that M of exogenous putrescine blocks the increase in ODC activity, but unlike in the parent HTC cell line, without induction of the antizyme or formation of any inactive ODC-antizyme complex. However, complete blockade of ODC at M putrescine is effected by induction of antizyme and formation of the ODC-antizyme complex, as now evidenced by the isolation of the active enzyme and antizyme components after Sephadex column chromatography in the presence of 250 mM NaC1. These findings indicate clearly that two polyamine-regulatory mechanisms for ODC exist and are separable in this cell line.
Mechanisms for the regulation of ornithine decarboxylase (E.C. 4.1.1.17; ODC), the ratelimiting enzyme which controls polyamine biosynthesis, include transcriptional (McCann et al., '77b), translational (Clark and Fuller, '75), and post-translational models (Sedory and Mitchell, '77; Obenrader and Prouty, '77). Among the most interesting of these mechanisms is that by which the end product of the enzyme reaction, putrescine, regulates the levels of ODC activity. Suppression of ODC activity by diamines had been noted in a number of cell lines (Clark and Fuller, '75; Pett and Ginsberg, '68; Kay and Lindsay, '73) and in vivo in regenerating rat liver (Janne and Holtta, '74; Poso and Janne, '76). The evidence seemed to suggest that the polyamine regulatory effect was post-transcriptional and a t the level of enzyme synthesis (Clark and Fuller, '75; Canellakis and Theoharides, '76; Kallio e t al., '77). Seemingly in opposition to these findings, however, was the discovery of the non-competitive protein inhibitor or antizyme which was induced by the presence of putrescine or other polyamines and specifically interacted with ODC to form an inactive enzyme-antizyme complex (Fong J. CELL.PHYSIOL. (1979) 99: 183-190.
e t al., '76; Heller et al., '76; Heller et al., '77). The presence of this ODC antizyme in several cell lines, in rat liver, and in rat thyroid was subsequently confirmed (Clark and Fuller, '76; McCann e t al., '77a; Jefferson and Pegg, '77; Pegg et al., '78; Poso e t al., '78; Friedman et al., '77). However, the antizyme could not be detected in several other r a t tissues, including kidney, prostate and heart, after various diamine treatments, even though ODC was totally blocked (Pegg et al., '78). Moreover, the decrease of immuno-reactive ODC protein seen after amine treatment in HTC cells (Canellakis and Theoharides, '76) and regenerating rat liver (Kallio e t al., '77) also suggested t h a t another ODC regulatory mechanism existed other than the induction of the antiz yme. We have been able to use HMOA cells, a clone of HTC cells with a greatly stabilized ODC (McCann e t al., '77b), as a probe to investigate the polyamine regulation of ornithine decarboxylase. We have found that the antizyme mechanism is the means by which ODC Received Oct. 4, '78.Accepted Dec. 12, '78.
183
184
P. P. MCCANN, C. TARDIF, J:M.
HORNSPERGER AND P. BOHLEN
is completely blocked by high concentrations of exogenous putrescine in these cells. However, when HMOA cells a r e treated by a low concentration of putrescine, ODC activity is still reduced, most likely by a block in new enzyme synthesis, but without induction of antizyme or formation of any ODC-antizyme complex. MATERIALS AND METHODS
Rat hepatoma (HTC1 cells and t h e clone of cells (HMO*) isolated from t h e parent HTC cell line as partially resistant to DL-a-methyl ornithine (Mamont e t al., '78) were grown in spinner culture in Swim's 77 medium supplemented with 10% calf serum as previously described (McCann e t al., '75; Mamont e t al., '76). Ornithine de, .rboxylase activity was measured by the me iod previously described (Mc-
9.0
2ann e t al., '75) in cell extracts, taken and pre3ared as described in t h e figures. Inhibitor or lintizyme activity, where one inhibitor unit is iefined as a n amount which will inhibit one unit of ODC activity i.e., 1 nmole/CO/hr, was lissayed as previously described (Heller et al., '76; McCann e t al., '77a). Cell protein was measured by t h e fluorescamine method (BohLen et al., '72). The half-line (t 1/21 of decrease iin enzyme activity was calculated by linear least-squares reduction. Supernatants of cells described in figures 3, 4 and 5 were prepared from concentrated sonicated cell extracts (1.5 X lo8cells in 10 ml) by centrifugation at 100,000 X g for one hour at 4°C as previously described (McCann et al., '77a). Chromatography and separation of ODC and inhibitor activity on a G-75 superfine Sephadex column in t h e presence of assay 1buffer and 250 mM NaCl was done as previ-
A
B
-
0-0
L
r \
70
0
.-I=
/ /"
/
a
c
2 Q 5.0
\
0/O
n
0 0
/
'O.
Y
/
u
L
n
m A A A \ ~ - A ~
A
Putrescine IO-~M
1.0 ' 0
--or
/"
A/A
0
3.0
Putrescine
lo-' M 1
I
0
2
T I M E
6
(hr)
10
0
2
T I ME
I
6
1
I
to
(hr)
Fig. 1 Effect of cycloheximide, putrescine, and cycloheximide plus putrescine on ornithine decarboxylase activity in HMOA cells. The cells were diluted into fresh medium in spinner culture (1.5 X lo5 cells per ml) and incubated for 12 hours a t 37°C. The cells were then divided into five individual cultures and putrescine and/or cycloheximide were added as below. After a further three hours of incubation aliquots of cells (2.0 x lo6) were then taken a t the indicated times (the first a t time 0) and washed by centrifugation twice with phosphate-buffered saline a t 4"C, and sonicated. After all the samples were taken the sonicates were directly assayed for ODC activity. Activities are means from a t least three different experiments using different cultures. A, Control (0); + M putrescine (A);+ M putrescine ( 0 ) .B, Control (0); + 50 pg/ml cycloheximide (A); 50 pg/ml cycloheximide and lo'* M putrescine ( 0 ) .
+
185
POLYAMINE REGULATORY MECHANISMS FOR ODC
ously described (McCann e t al., '77a) except as noted in the figures. Each assay for inhibitory activity of the column fractions contained a final concentration of 25 mM NaCl since 100 11.1aliquots were tested in a 1-ml assay volume. All control enzyme assays, used to calculate inhibitor concentrations also contained 25 mM NaCl to compensate for the slight inhibitory effect (less than 10%)of NaCl on ODC a t this concentration. When ODC was mixed to complex with free antizyme-inhibitor or when ODC-antizyme complex was already present, whichever was in excess, i.e., ODC or inhibitor, was the only assayable species. Only when column chromatography is done in the presence of 250 mM NaCl can both ODC and inhibitor be detected in the same extract. Bound ornithine decarboxylase inhibitor or antizyme in normal nonpolyamine-treated cells was extracted by the method described by Heller et al. ('77). After two centrifugations a t 100,000 x g in Tris buffer A (50 mM Tris pH 7.3 at 23"C, 50 ml KCL and 1 mM MgClJ the pellets from the cell extracts were stirred for ten minutes in lo-* M putrescine a t room temperature. EDTA (2 mM final concentration) and dithiothreitol (0.77 mg/ml) were added in the manner described (Heller et al., '77). After centrifugation again a t 100,OO x g for one hour, both supernatants and pellets were dialyzed for 15 hour a t 4°C with two buffer changes (500 vol) of Tris buffer B (10 mM Tris pH 7.3 a t 4"C, 50 mM KC1, 1 mM EDTA, 0.05 mM pyridoxal phosphate, 2.5 mM dithiothreitol) and then for four hours in the same buffer without KC1. RESULTS
When HMOAcells were induced to grow by dilution into fresh medium with 10% calf serum ornithine decarboxylase activity greatly increased over a period of hours as shown in figure 1. When cycloheximide was added to HMO* cells (fig. lB), the rise in ODC was blocked and there was a slow decay of enzyme activity. The biological half life of the enzyme (t 1/21 was found to be more than eight hours, in dramatic contrast to the t 112 of 22 minutes for ODC from HTC cells (fig. 2). When M was similarly added to HMOn cells, it also blocked the rise of ODC and a slow decay of enzyme activity was seen for some hours, although to a lesser effect than in the presence of cycloheximide (fig. 1A). This low concentration of putrescine, however, rapidly caused a decrease in ODC activity in HTC cells
CELLS
HMO, 100
CH tlI2=
8.2 hr
0
-
0
2
4
8
6
1 0 h r
I00
HTC CEI 1 S
,
40
put 10%
t,,,= 51 min
20
\A 10
0 '
\ I
60 120 180 m i n Fig. 2 Rates of decrease in ODC activity caused by cycloheximide and putrescine in HMOA cells and HTC cells. Maximal enzyme activity (100%activity) waa taken as the beginning of the phase of linear reduction of ODC activity. The activities for HMOA cells were taken from figure 1. Those for HTC cells were determined separately under the same experimental conditions as described in figure 1. HMOA cells: lo'* M putrescine (0); + 50 pg/ml cycloheximide and M putrescine ( 0 ) ;+ 50 pg/ml cycloheximide (W). HTC cells: + lo-* M putrescine (0); 50 pg/ml cycloheximide (W); M putrescine (A).
+
+
+
with t 1/2 of 51 minutes as shown in figure 2. Also, while a high concentration of putrescine (lo-* M) only moderately altered the t 1/2 of HTC cell ODC (fig. 21, in HMOAcells it had the effect of blocking all ODC (fig. 1A) (t 1/2 = 58 minutes; fig. 2). When M putrescine was added to HMOA cells in the presence of cycloheximide, the decrease of ODC activity was much less rapid than with putrescine alone (t 1/2 more than 3 hours) (fig. 2). We have recently shown that cell extracts from HTC cells which have been treated with M putrescine contain both ODC and antizyme as bound complex. The ODC and antizyme could be separated by chromatography
186
P. P. MCCANN, C. TARDIF, J:M.
8-
HORNSPERGER AND P. BOHLEN
i
.90
t; 4 U
LL
\
.60
w 5
>
N
z W
.30 0 (0
k z 3
0
10
20 FRACTION
30
40
NUMBER
Fig. 3 Chromatography of supernatant from HMOA cells treated for ten hours with loe5 M putrescine after initial stimulation of ODC. HMOA cells were incubated for eight hours after induction of ODC by dilution into fresh medium (1.5 X lo5cellslml). The cells were then incubated for an additional ten hours in the M putrescine. Two milliliters of concentrated cell supernatants (3 X lo7 cells) which were presence of prepared by 100,000 X g centrifugation as described in MATERIALS AND METHODS and which had approximately 20 units of residual ODC activity were layered onto a G-75 superfine Sephadex column and chromatographed with buffer containing 250 mM NaCl as described in MATERIALS AND METHODS. ODC activity ( 0 ) ; inhibitor activity (0).
on a G-75 superfine Sephadex column with assay buffer containing 250 mM NaCl (McCann et al., '77a). (This column procedure normally recovers 35-45%of the ODC and about 60%of the antizyme). When HMOA cells were treated with M for ten hours (twice the time of treatment for the HTC cells), no antizyme a t all could be detected after Sephadex column chromatography in the presence of 250 mM NaCl (fig. 3). Approximately 8 units of ODC were recovered from the original 20 units layered on the column. In contrast, when HMOA cells were treated for ten hours with l o - * M putrescine, they were found to produce antizyme in excess of ODC. The amount of free antizyme-inhibitor produced under these conditions was about 2 units per milligram protein. When such an extract with 4 units of free antizyme was chromatographed with buffer containing 250 mM NaC1, a total of 6 units of antizyme as well as about 0.6 units of ODC were recovered from the column (fig. 4). At least 2 of these 6 units of antizyme and all of the ODC must have been derived from bound antizyme-ODC complex. The fraction of antizyme derived from the bound complex was probably higher than 2 of the 6 units as recovery of the antizyme on the column is normally only 60%and we would only expect a yield of about 2.5 units of the original 4 units if there was no bound antizyme-ODC complex in the cells. The recovery
of 0.6 units of ODC was somewhat lower than would be expected from the amount of antizyme recovered.' When a similar experiment was done with HTC cell extracts with 19 units of free antizyme, 11 or 60%of these units were recovered from the column (fig. 4). Thus, in this second case the free antizyme was equivalent to the total antizyme of the cell extracts. As a control to show that the NaCl treatment would indeed completely dissociate t h e HMOA cell ODC-antizyme complex, we mixed 12.5 units of HMOAcell ODC and 7.5 units of HMOA free antizyme with 5 total units of detectable ODC activity remaining after mixing. We also mixed 10.6 units of HTC cell ODC and 6.6 units of HMOA free antizyme with 4 total units of detectable ODC activity remaining after mixing. When we chromatographed these two different mixtures with buffer containing 250 mM NaCl as shown in figure 5, we recovered in the first instance 8.2 units of antizyme and 7.2 units ODC, and in the second 12.8 units of antizyme and 6.1 units of ODC. In both cases we recovered more antizyme than 1 We have found that the HMOA cell ODC tends to show a loss in activity after NaCl treatment and column chromatography that is proportionately greater than after similar treatment of the HTC cell ODC. On occasion after such treatment the HMOA cell ODC has shown up to a 50%loss in activity after 24 hours at 4°C while under similar conditions the HTC cell ODC would only loose 10.20% of its activity. The multiple peaks of antizyme-inhibitor activity seen in figure 4 and in figure 5 are typical of the earlier results of Heller et al. ('76) and McCann et al. ('77a).
187
POLYAMINE REGULATORY MECHANISMS FOR ODC
HMO,
W
2
>
N
z
CELLS TREATED WITH
I
t
-
'lI
--I
I
P
W LL
I 0 2 M PUTRESCINE
.90
0
-
0
1.2
a \
11
-
I]
D 0
2 0
I
-
z .60 0
v
\
-
.30
To
'0-0
4 I0
the original free form amount and also more ODC than remained after mixing. Antizyme is bound to cell membranes of both cell lines. This antizyme could be isolated from the membrane fractions of sonicated cells by extraction with buffer containing M putrescine (table 1). The amounts found in both cell lines represent only 2.5%t o 7.5% of the total amount of antizyme found in maximally induced cells. HMOA cells contain appreciably less of both bound and free antizyme
0
20
than HTC cells, even after maximal induction of antizyme by treatment of the cells with lo-' M putrescine. Buffer containing M putrescine would not extract the bound antizyme from sonicates of either HMO* or HTC cells. DISCUSSION
Our earlier work on ODC regulation in HTC cells suggested that there might be two separate mechanisms for ODC regulation by
188
P. P. MCCANN. C. TARDIF, J.-M. HORNSPERGER AND P. BOHLEN HMOA C E L L ODC M I X E D WITH HMOA C E L L INHIBITOR
:
1.2
z
0 L
v
a
a
LL
90
W
f
N
z
W
R
.60
I
z
0
D
LL
.30
f? -
z
3
m
1
z
0
-
-
a
HTC CELL ODC MIXED WITH HMO,
CELL INHIBITOR
0
z
.90
0
n
a
D
a
k 0
0
a
-r0
LL 1
-60
z
W
f
N
z W
.30
V
c I LL
0
t z 3
0 10 20 FRACTION
30
40
NUMBER
Fig. 5 Chromatography of NaCl treated ODC-inhibitor complexes. HMOA cell ODC (12.5 units) was mixed with HMOA cell inhibitor (7.5 units) resulting in five total units of detectable OM: activity remaining after mixing. The total mixture (1.5ml) was chromatographed with buffer containing 250 mM NaCl as described in figure 3 and MATERIALS AND METHODS. HTC cell ODC (10.6 units) was mixed with HMOA cell inhibitor (6.6 units) resulting in four total units of detectable ODC activity remaining after mixing. The total mixture (1.5 ml) was chromatographed as above.
polyamines (McCann et al., '77a). One was that proposed by several workers (Clark and Fuller, '75; Canellakis and Theoharides, '76; Kallio et al., '77; Poso et al., '78) whereby ODC synthesis would be regulated by putrescine and/or spermidine. The second mechanism was the induction of the antizyme-inhibitor which directly complexes with the enzyme (Fong et al., '76; Heller et al., '76; McCann e t al., '77a). Our evidence in HTC cells showed M putrescine efficiently decreased that ODC activity without inducing any significant amount of antizyme. However, there was
some antizyme present as inactive ODC-inhibitor complex (McCann et al., '77a). The isolation of the HMOA cells and the discovery that it has a greatly stabilized ODC has given us a powerful tool to more precisely investigate the polyamine regulation of ODC. We are now able to conclude that there are indeed two distinct polyamine-controlled regulatory mechanisms. The slow and partial efM putrescine on HMOAcell ODC is fect of clearly similar to the effect seen when we used the general protein synthesis inhibitor cycloheximide (fig. 1) and suggests that both are
189
POLYAMINE REGULATORY MECHANISMS FOR ODC TABLE 1
Comparison of amounts of ornithine decarboxylase inhibitor found a8 membrane-bound form in sonicates of confluent cells, and as free inhibitor in maximally induced cells Ornithine decarboxylase antizyme-inhibitor Bound and extracted inhibitor Total units 12 x lo8 cells
HTC cells (% of induced cells)
HMOA cells (% of induced cells)
3.68 (2.6%) 2.80 (3.3%)
Specific activity Img protein
0.20 (6.9%) 0.16 (7.5%)
Induced inhibitor Total units 12 x lo8 cells
Specific activity Img protein
141.0
2.93
84.8
2.13
Parallel spinner cultures of HTC and HMOA cells were grown to confluence (8.5 X lo5 to 1.1 x lo6cells/ mi) in Swim's 5-77 medium in the absence of putrescine. The cells 12 x 10"total per culture) were washed in phosphate-buffered saline, sonicated and suspended in 5 ml of Tris buffer A. The extracts were centrifuged a t 100.000 x g for one hour, and then the pellets were resuspended in 5 ml of Tris buffer A and again centrifuged a t 100,000 X g for one hour. Bound ornithine decarboxylase antizyme was extracted from this pellet by 10.' M putrescine and then measured as described in MATERIALS AND METHODS. If antizyme activity could be detected of course no ODC activity could be present. In the non-induced (confluent) cells all the cellular ODC activity was removed in the discarded supernatants of the two 100,000 X g centrifugations done before extraction of the antizyme from the pellets. The inhibitor activities shown represent total activity, and specific activity from the final pellet and supernatant together. Free antizyme was induced in cells t h a t had been diluted to 1.5 X lo5 celldm1 (2 x lo8 total cells per culture) in fresh medium containing 10 mM putrescine, and then incubated a t 37°C for 12 hours. After washing and sonication the amount of antizyme was measured in these cell extracts. Each value represents the mean of a t least two separate determinations with less than 6% variation among the individual determinations.
working by a similar mechanism that is by M publocking new ODC synthesis. When trescine is used to block HMOAcell ODC, the M effect is so rapid as compared t o either putrescine or cycloheximide, t h a t clearly another mechanism is being brought into play. This is induction of the antizyme-inhibitor. It had previously been shown (Fong et al., '76; Heller et al., '76; Poso e t al., '78) that the induction of inhibitor was dependent on new protein synthesis and, indeed, when cycloheximide and M were added together (fig. lB),the immediate blocking effect of l o WM2 putrescine was greatly suppressed. The fact that cycloheximide did not completely block M putrescine effect (fig. 2) can be exthe plained by the fact that some preformed membrane bound antizyme-inhibitor was already present and released by the high concentration of putrescine in situ (table 1). Thus, we can argue that the low concentration of putrescine M) only blocks ODC synthesis and is not sufficient to either induce new antizyme or release the preformed membrane bound inhibitor from HMOA cells. In contrast, lo-* M putrescine induces and releases antizyme, which explains why ODC activity rapidly drops from the partially stable level seen M putrescine after cycloheximide or treatment to zero (fig. 1). Further evidence for this is that unlike in
HTC cells, when concentrated cell extracts M putrescine-treated HMOA cells from were chromatographed with buffer containing 250 mM NaCl, no inhibitor at all could be detected (fig. 3). This strongly suggests the antizyme induction and/or release mechanism is not activated by this concentration of putrescine in HMOA cells. However, the antizyme is clearly responsible for the complete M putrescine, as block of ODC activity by is shown by the results in figure 4. In striking contrast, similarly treated HTC cells seem to contain only free antizyme and no ODC-antizyme complex (fig. 4).This suggests that in HTC cells under the conditions of high exogenous putrescine, the other repressive mechanism was the primary means by which the ODC was blocked. Several controls were done to insure that the lack of the appearance of the inhibitor in HMOA cells, under conditions of treatment M putrescine, was not just the result with of some difference in the ability of the HMOA cell ODC and inhibitor to form a more tightly bound complex that was less sensitive to dissociation by the salt treatment. As is shown in figure 5, the complex is clearly dissociable by 250 mM NaC1. This can also be seen from the results in figure 4. In addition, the HMOAcell antizyme can complex with and dissociate from both its own ODC and that of HTC cells.
190
P. P. MCCANN, C. TARDIF, J.-M. HORNSPERGER AND P. BOHLEN
Other evidence also shows that the equilibria of the non-competitive association of ODC with the antizyme are the same for both cell lines (P. P. McCann, to be published). Finally, as HMO* and HTC cells both contain the same small fraction of bound as compared to fully induced antizyme (table l),the bound form of the antizyme cannot somehow be responsible for the regulatory effect of M putrescine in the HMOAcells. We have also M putresfound that a concentration of cine is not sufficient to cause bound antizyme to be extracted from the membrane fraction in either cell line. In summary, by using the HMO* cell line we have been able t o demonstrate that two different mechanisms exist for the polyamine regulation of ODC. This supports a recent hypothesis that intracellular and exogenous polyamines may regulate ODC by different mechanisms (Heller et al., '78). However, we have not yet been able to demonstrate specifically that polyamines block ODC synthesis, although our evidence and that from other laboratories strongly suggests that this is the case. ACKNOWLEDGMENTS
We thank Doctors J. S. Heller, E. S. Canellakis, H. Poso and J. Janne for making their manuscripts available to us before publication. We also thank Doctors N. Seiler and J. Koch-Weser for their very helpful suggestions with this manuscript. LITERATURE CITED Bijhlen, P., S. Stein, W. Dairman and S. Udenfriend 1972 Fluorometric assay of proteins in the nanogram range. Arch. Biochem. Biophys., 155: 213-220. Cannellakis, 2. N., and T. C. Theoharides 1976 Stimulation of ornithine decarboxylase synthesis and its control by polyamines in regenerating rat liver and cultured rat hepatoma cells. J. Biol. Chem., 251: 4436-4441. Clark, J. L., and J. L. Fuller 1975 Regulation of ornithine decarboxylase in 3T3 cells by putrescine and spermidine: indirect evidence for translational control. Biochemistry, 14: 4403-4409. 1976 Protein inhibitor of ornithine decarboxylase does not account for effect of putrescine on 3T3 cells. Biochem. Biophys. Res. Commun., 73: 785-790. Fong, W. F., J. S. Heller and E. S. Canellakis 1976 The appearance of a n ornithine decarboxylase inhibitory protein upon th e addition of putrescine to cell cultures. Biochim. Biouhvs. Acta. 428: 456-465. Friedman, Y., S. Park, S. Levasseur and G. Burke 1977 Regulation of thyroid ornithine decarboxylase by the
Polyamines. Induction of a protein inhibitor of ornithine decarboxylase by t h e end-products of t h e reaction. Biochim. Biophys. Acta, 500: 291-303. Heller, J. S., W. F. Fong and E. S. Canellakis 1976 Induction of a protein inhibitor to ornithine decarboxylase by the end products of its reaction. Proc. Natl. Acad. Sci. (U.S.A.), 73: 1858-1862. Heller, J. S., K. Y. Chen, D. A. Kyriakidis, W. F. Fong and E. S. Canellakis 1978 The modulation of the induction of ornithine decarboxylase by spermine, spermidine and diamines. J. Cell. Physiol., 96: 225-234. Heller, J. S., D. Kyriakidis, W. F. Fong and E. S. Canellakis 1977 Ornithine decarboxylase antizyme is a normal component of uninduced H-35 cells and ra t liver. Eur. J. Biochem., 81: 545-550. Janne, J., and E. Holtta 1974 Regulation of ornithine decarboxylase activity by putrescine and spermidine in rat liver. Biochem. Biophys. Res. Commun., 61: 449-456. Jefferson, L. S., and A. E. Pegg 1977 Studies on ornithine decarboxylase activity in the isolated perfused rat liver. Biochim. Biophys. Acta, 484: 177-187. Kallio, A., M. Ltifman, H. POSOand J. J h n e 1977 Inhibition of ornithine decarboxylase by diamines in regenerating rat liver. FEBS Lett., 79: 195-199. Kay, J. E., and V. J. Lindsay 1973 Control of ornithine decarboxylase activity in stimulated human lymphocytes by putrescine and spermidine. Biochem. J., 132: 791-796. Mamont, P. S., P. Bohlen, P. P. McCann, P. Bey, F. Schuber andC. Tardif 1976 a-Methyl ornithine, a potent competitive inhibitor of ornithine decarboxylase, blocks proliferation of rat hepatoma cells in culture. Proc. Natl. Acad. Sci. (U.S.A.), 73: 1626-1630. Mamont, P. S., M. C. Duchesne, J. Grove and C. Tardif 1978 Initial characterization of an HTC cell variant partially resistant to the anti-proliferative effect of ornithine decarboxylase inhibitors. Exptl. Cell Res., 115: 387-393. McCann, P. P., C. Tardif and P. S. Mamont 1977a Regulation of ornithine decarboxylase by ODC-antizyme in HTC cells. Biochem. Biophys. Res. Commun., 75: 948-954. McCann, P. P., C. Tardif, M. C. Duchesne and P. S. Mamont 1977b Effect of a-methyl ornithine on ornithine decarboxylase activity of ra t hepatoma cells in culture. Biochem. Biophys. Res. Commun., 76: 893-899. McCann, P. P., C. Tardif, P. S. Mamont and F. Schuber 1975 Biphasic induction of ornithine decarboxylase and putrescine levels in growing HTC cells. Biochem. Biophys. Res. Commun., 64: 336-341. Obenrader, M. F., and W. F. Prouty 1977 Detection of multiple forms of rat liver ornithine decarboxylase. J. Biol. Chem., 252: 2860-2865. Pegg, A. E., C. Conover and A. Wrona 1978 Effects of aliphatic diamines on ra t liver ornithine decarboxylase activity. Biochem. J., 170: 651-660. Pett, D. M., and H. S. Ginsberg 1968 Metabolism of polyamines in KB cells. Fed. Proc., 27: 615. Poso, H., and J. Janne 1976 Inhibition of ornithine decarboxylase activity and spermidine accumulation in regenerating ra t liver. Biochem. Biophys. Res. Commun., 69: 885-892. Pbso, H., S. K. Guha and J. Janne 1978 Stabilization of ornithine decarboxylase in ra t liver. Biochim. Biophys. Acta, 524: 466-473. Sedorv. M. J.. and J. L. A. Mitchell 1977 Remlation of ornithine decarboxylase activity during the phys a rum mitotic cycle. Exptl. Cell Res., 107: 105-110.
