Cell, Vol. 5, 29-35,

May 1975,

Copyright0

“Superinduction” by Actinomycin

1975 by MIT

of Tyrosine Aminotransferase D: a Reevaluation

Robert A. Steinberg*, Barbara B. Levinson, Gordon M. Tomkins Department of Biochemistry and Biophysics University of California at San Francisco, California 94143

Summary Reexamination of the effects of actinomycin II (AMD) on the intracellular level and rate of synthesis of tyroslne aminotransferase (TAT) in hepatoma tissue culture (HTC) cells reveals that much apparent controversy can be resolved with acknowledgment of the multi-faceted nature of this inhibitor’s action. AMD can slow overall protein synthesis and inhibit the degradation of both TAT and its mRNA as well as block the synthesis of RNA. The extent of these secondary actions of the inhibitor depend somewhat upon the growth condition of the cells. The effects of cordycepin (a’deoxyadenosine) on the metabolism of TAT and its mRNA are also complex, but differ in several respects from those of AMD. Introduction Early studies of the steroid-mediated induction of tyrosine aminotransferase (TAT, E.C. 2.6.1.5) in rat liver and in hepatoma cells revealed that actinomytin D (AMD) and other inhibitors of RNA synthesis not only prevented induction of the enzyme but also prevented its deinduction (Garren et al., 1964; Thompson, Tomkins, and Curran, 1966). This paradoxical maintenance of enzyme level in the absence of RNA synthesis has now been observed in a wide variety of inducible systems (reviewed in Tomkins et al., 1972) and has suggested the possibility of posttranscriptional mechanisms for gene control in eucaryotes. The potential importance of posttranscriptional regulation is also implied by the striking stability of eucaryotic mRNAs whose mean lifetimes approach a cell generation time (Greenberg, 1972; Murphy and Attardi, 1973; Singer and Penman, 1973); the levels of such stable mRNAs could be modulated only very slowly by changing their rates of synthesis. There has been some controversy over the mechanism of the AMD effect on deinduction of TAT. Experiments from this laboratory have demonstrated an AMD-mediated maintenance of the rate of TAT synthesis compared with the usual drop in synthesis following steroid removal (Thompson, Granner, and *Present address: Fund, Burtonhole

Mill Hill Laboratory, Imperial Cancer Lane, London NW7 1 AD, England.

Research

Tomkins, 1970; Tomkins et al., 1972). These studies were interpreted to suggest that deinduction involves an RNA-requiring process in which translation of TAT-specific mRNA is blocked and/or the mRNA destabilized. Other investigators have reported that TAT synthesis falls rapidly in the presence of AMD, and that the principal effect of the inhibitor is a stabilization of the enzyme itself (Reel and Kenney, 1968; Kenney et al., 1973). Using a detailed kinetic study of the steroidmediated induction and deinduction of TAT synthetic activity, we have recently shown that the mRNA for TAT is unusually unstable, having a halflife of about 1-l 5 hr (Steinberg, Levinson and Tomkins, 1974b). The rapid turnover of this mRNA proceeds as well in growing as in serum-deprived nongrowing cells, and it is not apparently altered by dexamethasone, a glucocorticoid inducer of TAT. Other studies from this laboratory have established that steroids have no direct effect on the translation of TAT mRNA (Scott, Shields, and Tomkins, 1972; Steinberg et al., 1974a). Taken together these findings suggest that steroids specifically regulate the rate of production of TAT-specific mRNA and thus render unlikely a direct relationship of the maintenance of TAT in the presence of AMD to the induction of TAT by steroids. Nevertheless, two lines of evidence suggest that the AMD-mediated maintenance of TAT is not simply an inhibitor artifact and support its possible significance for regulation: -induced levels of TAT are maintained in cells which have been physically enucleated by centrifugation in the presence of cytochalasin B (Ivarie, Fan, and Tomkins, 1974); -and Hepatoma cells from which steroid has been removed after the S phase of the cell cycle maintain induced levels of TAT through mitosis and the first 3 hr of Gl (Martin, Tomkins, and Bresler, 1969; Sellers and Granner, 1974). In view of the continuing controversy over the effects of AMD on the metabolism of TAT, the present study reexplores the phenomena in hepatoma tissue culture (HTC) cells. Since previous studies have suggested the relevance of cellular growth conditions to the particulars of the TAT response to AMD treatment (Thompson et al., 1970), special consideration is given them. Moreover, in light of recent findings that some inhibitors of RNA synthesis block induction, but not deinduction of TAT activity (Butcher et al., 1972; Bushnell, Becker, and Potter, 1974) the present report examines in more detail the effects of one of them, cordycepin (3’-deoxyadenosine).

Cell 30

Results Growth Regulation in Hepatoma Cells When HTC cells are deprived of serum, they show a reduction in protein synthesis and an enhancement of protein degradation (Gelehrter and Tomkins, 1969; Hershko and Tomkins, 1971). The reversal of these effects by serum readdition may somehow be related to the changes observed in mature hepatocytes stimulated to grow by partial hepatectomy(Schreiber et al., 1971; Scornik, 1972). Thus, despite their tumorigenicity, HTC cells might show at least part of the “pleiotypic” or coordinated program of growth control observed in untransformed cell lines (Hershko et al., 1971). Considering that mature hepatocytes are blocked in the Gl (or “GO”) portion of the cell cycle (Grisham, 1969) and that untransformed cultured cells generally manifest growth limitation by an extended Gl phase (Pardee, 1974) the cell cycle specificity of HTC cell growth inhibition was investigated. Flow microfluorimetry of cells stained with acriflavine (Van Dilla et al., 1969) was used to follow the response of HTC cultures to removal of serum (Figure 1). Exponentially growing cells were centrifuged, washed, and resuspended in either fresh serum-containing growth medium (Figure la) or

serum-free BSA medium (Figure 1 b). Samples were removed for flow microfluorimetric analysis at various times after resuspension. While the control culture showed slight changes in the distribution of DNA content per cell, the culture incubated without serum showed a marked progressive decrease in the proportion of cells in S, G2, and M, reaching a minimum after about one generation time. Thus under some physiological conditions, hepatoma cells can be blocked in Gl. Effect of Growth State on “Superinduction” by AMD Although induced levels of TAT are maintained upon the administration of AMD to growing HTC cells (Figure 2a; Tomkins et al., 1972), the pronounced increase or “superinduction” of TAT observed after administration of AMD to rat liver (Garren et al., 1964) is observed only in growth-inhibited cells (Figure 2b; Reel and Kenney, 1968; Thompson et al., 1970). For the experiments shown in Figure 2, AMD was simply added to cultures preinduced with steroid to avoid additional effects of handling or change in media (Figures 3, 4; Auricchio, Martin, and Tomkins, 1969). Figure 3 shows a more detailed analysis of the effects of AMD on the metabolism of TAT in growing

10 -

10 -

9-

9-

a-

31

a-

a) GROWTH MEDIUM

': g'

b) BSA MEDIUM

76-

; 6v

5-

5

5-

f

4

s 2

3-

3-

2-

2-

l-

l-

4-

G2+M 0 / I2

/I

0 0.0

0.4 DNA

Figure

1. Distributions

0.8 CONTENT of Cellular

Exponentially growing cells were represent hours after resuspension), Procedures). (a) Control culture (b) Serum-deprived

1.2

1.6

PER CELL (arbitrary DNA Content

2.0

2.4

units)

in Growing

010

014 DNA

and Serum-Deprived

0.8

CONTENT

1.2 PER CELL

1.6 (arbitrary

2.0

2.4

units)

HTC Cells

centrifuged, washed, and resuspended in either growth or SSA medium. At the times shown (numerals cell samples were centrifuged, washed, and fixed for later flow microfluorimetric analyses (Experimental

in growth medium. culture in BSA medium.

Superinduction 31

of TAT

by Actinomycin

D

HTC cells. Under these conditions, AMD not only stabilized the level of TAT activity against the normal fall occasioned by steroid withdrawal (Figure 3a), but also maintained the relative rate of TAT synthesis at approximately induced levels for as long as 6 hr (Figure 3b). When the rates of TAT synthesis were expressed as cpm/mg of extracted protein, a considerable decline was observed (Figure 3c), but this decline was still slower than that observed in the deinduced culture. The effects of AMD on serum-deprived HTC cells (Figure 4) differed in several respects from those observed in growing cells. Despite the considerable AMD-mediated increase in TAT activity over both fully induced and deinduced controls (Figure 4a), the rate of TAT synthesis fell markedly from induced levels following AMD addition (Figure 4b,c). Since there was only slight inhibition of amino acid incorporation by AMD in this experiment, it made little difference whether the immunoprecipitation data were expressed as percent of total incorporation (Figure 4b) or relative to extracted protein (Figure 4~). In either case the decline in rate of synthesis of TAT was considerably more rapid during deinduction than after AMD treatment. As previously demonstrated (Thompson et al., 1970), the effects of AMD were all independent of steroid.

-DEX+AMD

140-

120 -

bl

o.nF

O.OAC

a) GROWTH

I

MEDIUM

I

I

R

I

I

I

I

18

cl

I6

+DEx

b) BSA MEDIUM A

0 Figure 2. Effect TAT by AMD

1 2 3 HOURS AFTER AMD

4 5 ADDITION

of Cell Growth

on “Superinduction”

Condition

6

I 1

0

I

I

2

3

I A

I

I

5

6

HOURS of

HTC cells were induced overnight (about 19 hr) with IO-&M dexamethasone (DEX) in either (a) growth medium or(b) BSA medium. Each culture was then divided and AMD added to one portion (final concentration 5 pg/ml). At 1 hr intervals, samples were taken for determinations of TAT specific activity.

Figure

3. AMD

Effects

on TAT

Synthesis

in Growing

HTC

Cells

Cells preinduced overnight in growth medium with lo-TM DEX were washed and resuspended in either fresh growth medium, growth medium containing lo-TM DEX, or growth medium with 5 pg/ml AMD. Samples were removed at intervals, labeled with 3H-leucine, and analyzed for activity and synthesis of TAT as described in Experimental Procedures.

Cell

32

al

a) 2mc 170 140 120 100 80

+AMD

+ DEX

4 DEX

*yi-z2

E

df’ 0

4

60 50 40 1

$, , ,r\--, a

0.5 r

2or

,

b) b)

0.4 0.3

0.08

+CORDYCEPH

0.06 0.05

0.1

0.04

0.08

0.03/-

0.06

1

,

1

1

l

l

,

,

+CORDYCEPIN

L I

I

I

I

I

I

I

I

012345678

0.6 1 012345678

I

I

I

I

I

1

HOURS

HOURS Figure 4. Analysis of “Superinduction” Deprived HTC Cells

I

of TAT

by AMD

in Serum-

Cells were preinduced as in Figure 3 but in BSA medium, then washed and resuspended in fresh BSA medium with or without DEX (lo-TM) and AMD (5 pg/ml) as indicated in the figure. At various times samples were taken for analyses as in Figure 3.

Figure 5. Effects sis of TAT

of Cordycepin

on the Activity

and Rate of Synthe-

Additional preinduced cells from the experiment shown in Figure 4 were washed and resuspended in BSA medium containing DEX and/or cordycepin (concentration 15 pg/ml). The activity and rate of synthesis of TAT were monitored as in Figures 3 and 4 (control curves without inhibitors are redrawn from Figure 4).

Superinduction 33

of TAT

by Actinomycin

D

Effects of Cordycepin on Levels and Rates of Synthesis of TAT Figure 5 shows the effects of cordycepin on fully induced HTC cells or cells deinduced by removal of steroid. In contrast to AMD, cordycepin allowed TAT activity to decline (Figure 5a; Butcher et al., 1972). The time course of this change in enzymic activity, however, differed from that seen on deinduction. The rapid initial fall in TAT level following addition of cordycepin was apparently the result of a sharp diminution in the rate of enzyme synthesis (Figure 5b,c). The relative rate of TAT synthesis fell rapidly for about 4 hr in the presence of cordycepin, then was stabilized (Figure 5b); the rapid phase of the decline approached that observed after steroid withdrawal but without the lag preceding normal deinduction (Figure 5b; Steinberg et al., 1974b). [The slight differences in rates of decay of relative synthesis of TAT between cordycepin-treated and deinduced cells (Figure 5b) might be a mathematical artifact arising from a decrease in the total pool of cytoplasmic mRNA after cordycepin treatment. Since both inhibitors efficiently prevent the new appearance of mRNA at the concentrations used in these studies (Penman, Rosbash, and Penman, 1970; Peterkofsky and Tomkins, 1967) such an artifact cannot explain the greater maintenance of TAT synthesis by AMD.] Since amino acid incorporation was inhibited considerably by treatment with cordycepin, the total rate of incorporation into immunoprecipitable TAT (Figure 5c) fell even more dramatically than the relative rate. As observed for AMD, the cordycepin-mediated effects on both level of TAT and rate of TAT synthesis were unaffected by the presence or absence of steroid. Discussion Extending previous studies from this laboratory, the experiments presented here demonstrate that AMD stabilizes the rate of TAT synthesis against the more rapid declines observed in cultures from which steroid has been removed (Figures 3, 4). Similar differences in the rates of decline of specific enzyme synthesis following treatment with AMD or cordycepin have also been observed in other systems (Grayson and Berry, 1973; Tilghman et al., 1974). While in principle the maintenance of specific enzyme synthesis by AMD may result from either a stabilization of mRNA or a selective enhancement of mRNA translation (Palmiter and Schimke, 1973) we favor the former mechanism to explain the inhibitor’s effect on relative TAT synthesis. The half-life for TAT mRNA is less than 1.5 hr in the absence of inhibitors (Steinberg et al., 1974b). The translational mechanism would therefore require that the rate of translation of TAT mRNA be increased more

than 15 fold relative to that for other mRNAs to achieve the nearly complete maintenance of TAT synthesis observed in growing cells after 6 hr in AMD (Figure 3). Stabilization by AMD need not be specific for TAT mRNA, however, since an overall stabilization of mRNA would result in a relative increase in the level of this more unstable species. On the other hand, at least one recent report suggests that AMD enhances the degradation of most cellular mRNA (Singer and Penman, 1973), leaving open the possibility of a specific effect of AMD on TAT mRNA. Specificity of cordycepin for synthesis of polyadenylate and ribosomal precursor RNA (Siev, Weinberg, and Penman, 1969; Darnell et al., 1971) is relatively short-lasting since, by 4 hr after addition of this drug, synthesis of HTC cell RNA is nearly totally inhibited (Steinberg, unpublished results). Therefore, the maintenance of the relative rate of TAT synthesis at late times after addition of cordycepin (Figure 5) tends to support the conclusion, suggested by results with AMD and other inhibitors of RNA synthesis (Levinson, Tomkins, and Stellwagen, 1971; Tomkins et al., 1972), that a labile RNA species is required for rapid turnover of the mRNA for TAT. The failure to observe maintenance of TAT synthesis in the first 3 hr after treatment with cordycepin suggests that the mediator of TAT mRNA instability is neither rRNA nor polyadenylate-containing mRNA. Enucleation studies indicate that synthesis of this presumptive regulatory RNA requires the cell nucleus (Ivarie et al., 1974), and experiments with synchronized cells suggest that it is synthesized only during S and the latter portion of Gl in growing cells (Martin et al., 1969). The identification of this element and an understanding of its role in mRNA metabolism await further study. Although our incorporation data have not been corrected for possible changes in intracellular leutine specific radioactivity (Regier and Kafatos, 1971), the progressive polysome disaggregation observed when HTC cells are incubated with AMD (not shown) is consistent with results from other systems showing that initiation of protein synthesis is slowed by the inhibitor (Singer and Penman, 1972; Craig, 1973). Thus the decline in total incorporation into TAT observed after AMD treatment (Figures 3c, 4c) reflects a true fall in the absolute rate of synthesis of TAT as noted by others (Kenney et al., 1973). The increase in the level of TAT observed under some conditions of AMD treatment must therefore result in part from inhibition of the degradation of TAT. AMD-mediated stabilization of TAT has been reported under a number of experimental conditions (Reel and Kenney, 1968; Auricchio, Martin, and Tomkins, 1969; Kenney et al., 1973), and the coordinated regulation of hepatoma

Cell 34

cell growth provides a framework for understanding these observations. As discussed above (Results), when hepatoma cells are exposed to a growth-limiting environment, they exhibit several responses characteristic of normal hepatocytes and untransformed cell lines. Furthermore, as shown in Figure 1, serum-deprivation preferentially blocks HTC cells in the Gl phase of the cell cycle. This result was somewhat unexpected since HTC cells have unusually low levels of adenylate cyclase and CAMP (Granner et al., 1968; Makman, 1971) which have been implicated in the growth regulatory response (Burger et al., 1972; Otten, Johnson, and Pasten, 1972). On the other hand, Gl arrest has also been observed recently in a number of other malignant cell lines (Holley, 1974; Bourne et al., 1974). The observation that “superinduction”, or a true increase in the level of TAT following AMD treatment, occurs in serum-deprived HTC cells (Figures 2b, 4a) and in stationary cultures of H-35 cells maintained in medium with or without serum (Reel and Kenney, 1968; Kenney et al., 1973) but not in growing HTC cells (Figures 2a, 3a; Tomkins et al., 1972) suggests that growth inhibition, rather than simple nutritional step-down, provides the milieu for superinduction. The conclusion that steroids regulate some aspect of the production of TAT mRNA rather than its degradation (Lee, Reel, and Kenney, 1970; Steinberg et al., 1974) is supported by the absence of any steroid effect on the rapid initial decline of TAT synthesis mediated by cordycepin (assuming that the only significant early effect of cordycepin is to inhibit the synthesis of TAT mRNA). Since, in contrast to normal deinduction, the fall in TAT synthesis following cordycepin addition shows no significant lag (Figure 5b), the steroid-regulated step in the production of TAT mRNA could precede polyadenylate addition. The complexity of the TAT response to treatment with inhibitors of RNA synthesis illustrates additional possibilities for controlling the level of an enzyme which turns over rapidly and whose mRNA is also rapidly degraded. The stabilization of TAT mRNA by AMD appears similar to that observed in mitotic cells (Martin et al., 1969), and may indicate a physiological mechanism to prevent large fluctuations in the levels of short-lived messengers during the transcriptional shut-down of mitosis. “Superinduction” of TAT by high levels of AMD occurs under conditions where general protein degradation is enhanced (Hershko and Tomkins, 1971) and probably results from a slowing of this process. Both serum and insulin can slow enhanced protein degradation (Hershko et al., 1971), and, under conditions of

growth inhibition, both these agents promote increases in the specific activity of TAT in HTC cells (Gelehrter and Tomkins, 1969; Gelehrter and Tomkins, 1970). While serum and insulin do not apparently increase the relative rate of TAT synthesis (Steinberg et al., 1974 Mamont and Tomkins, unpublished results), their stimulation of overall protein synthesis may also contribute to the differential rise in TAT level. Experlmental

Procedures

HTC cells were grown in suspension cultures in Swim’s S-77 Medium with 10% calf serum (growth medium) as previously described (Hershko and Tomkins, 1971; Tomkins et al., 1972). For serumdeprivation studies, cells in mid-logarithmic phase growth were centrifuged, washed, and resuspended in serum-free S-77 containing 5 mg/ml bovine serum albumin (BSA medium). Details of hormonal and inhibitor treatments are provided in legends to the figures. Analyses of cell DNA content were performed on the Lawrence Livermore Laboratory’s flow microfluorimeter using cells fixed with 10% formalin and stained by the Feulgen-acriflavine procedure (Van Dilla et al., 1969; Van Dilla et al., 1974). Cell suspensions were sonicated before analyses, and determinations were terminated when 104 counts were recorded in a single channel. Figure 1 is redrawn from computer tracings of the data. Rates of synthesis of TAT were determined by measuring the radioactivity incorporated into immunoprecipitable TAT during a 15 min exposure to ,H-leucine (Tomkins et al., 1972; Steinberg et al., 1974 in press). Extracts were prepared from washed, frozen cell pellets using “sonicating buffer” (0.05 M potassium phosphate buffer, pH 7.6,2 x lo-4M pyridoxal phosphate, lo-3M ethylenediamine-tetraacetate, 5 x lo-‘M P-oxoglutarate) containing 0.5% Nonidet P-40 (Shell Chemicals), and TAT was partially purified by high speed centrifugation, heat precipitation, and DEAE-cellulose chromotography(Tomkins et al., 1972). Relative rates of synthesis, expressed as percent of cpm incorporated into total soluble extracted protein, were corrected for background counts in “second” immunoprecipitates and for recoveries of TAT in the prepurification steps (Tomkins et al., 1972). Total incorporation into TAT was calculated by multiplying this corrected fractional rate of TAT synthesis by the total cpm incorporated into soluble protein. Assays of TAT were by the method of Diamondstone (1966) and extract protein was determined by the procedure of Lowry et al. (1951) using bovine serum albumin as a standard. The specific enzymic activities shown in Figure 2 were determined using the supernatant fraction after a 10 min centrifugation at 30,000 x g. while those for Figures 3-5 were measured using a high speed supernatant fraction (1 hr centrifugation at 100,000 x g). Acknowledgments This work, supported by a grant from the National Institute of General Medical Sciences of the National Institute of Health, was submitted by RAS in partial fulfillment of the requirements for the degree of Doctor of Philosophy. During the course of these studies RAS was supported by a Graduate Fellowship of the National Science Foundation and by a Training Grant from the National Institute of General Medical Sciences of the National Institutes of Health. The authors wish also to thank Joe Gray of the Lawrence Livermore Laboratory for making the flow microfluorimeter available to us. Received

January

13. 1975;

revised

February

14, 1975

Superinduction 35

of TAT

by Actinomycin

D

Relerences

Regier. 6488.

Auricchio, F., Martin, D. W.. Jr., and Tomkins, G. (1969). Nature 224, 806-808. Bourne, H. R., Coffino, P., Melmon, K. L., Tomkins, G. M., and Weinstein, Y. (1974). Advances in Cyclic Nucleotide Res., 5, (New York: Raven Press), in press.

Schreiber, U. (1971).

Burger, M. M., Bombik, B. M., Breckenridge, B. McL., pard, J. Ft. (1972). Nature New Biol. 239, 181-163. Bushnell, Biophys.

D. E., Becker, Res. Commun.

J. E.. and Potter, 56, 815-821.

V. Ft. (1974).

Butcher, F. R.. Bushnell, D. E., Becker, (1972). Exptl. Cell Res. 74, 115-123. Craig,

N. (1973).

Darnell, Science

J. E., Philipson, 174, 507-510.

Diamondstone, Garren, (1964).

J. Cell. Physiol.

J. E., and

Biochem.

Potter,

V. R.

82, 133-150.

L.. Wall,

R..

Anal.

Biochem.

T. I. (1966).

and Shep-

and

Adesnik,

M. (1971).

16, 395-401.

L. D., Howell, R. R., Tomkins, G. M., and Crocco, Proc. Nat. Acad. Sci. USA 52, 1121-1129.

R. M.

Gelehrter. T. D.. and Tomkins. USA 64, 723-730.

G. M. (1969).

Proc.

Nat. Acad.

Sci.

Gelehrter, T. D., and Tomkins, USA 66, 390-397.

G. M. (1970).

Proc.

Nat. Acad.

Sci.

Granner, Science

D., Chase, L., Aurbach, 762, 1018-1020.

Grayson,

S., and Berry,

Greenberg,

G., and Tomkins,

S. J. (1973).

J. R. (1972).

Nature

Science

G. M. (1968).

780, 1071-1072.

A., and Tomkins,

G. M. (1971).

Hershko. A., Mamont, P., Shields, Nature New Biol. 232, 206-211.

J. Biol. Chem.

R., and Tomkins,

246, 710-

G. M. (1971).

Halley, R. W. (1974). In Control of Proliferation in Animal Cells, 8. Clarkson, and R. Baserga, eds. (Cold Spring Harbor Laboratory), pp. 13-18. Ivarie, R. D., Fan, W., and Tomkins, in press. Kenney, F. T., Lee, K. L., Stiles, Nature New Biol. 246, 208-210. Lee, K. L., Reel, 245, 5806-5812.

G. M. (1974). C. D., and

J. E. (1973).

F. T. (1970).

J. Biol. Chem.

Levinson, B. B., Tomkins, G. M., and Stellwagen, J. Biol. Chem. 246, 6297-6302.

R. H. (1971).

Lowry, (1951).

J. R., and Kenney,

J. Cell. Physiol.,

Fritz,

0. H., Rosebrough, N. J., Farr, J. Biol. Chem. 193, 265-275.

Makman,

M. H. (1971).

Proc.

A. L., and

Nat. Acad.

W., and Attardi,

Otten, J., Johnson, 247, 7082-7087. Palmiter, R. D., and 1502-l 512. Pardee,

A. B. (1974).

G. (1973).

G. S., and Schimke,

M. A. (1969).

Nat. Acad. I, (1972).

R. T. (1973).

Sci. USA 70, J. Biol.

J. Biol.

Chem.

Chem.

Nat. Acad.

Sci. USA

71, 1286-1290.

Penman, S., Rosbash, M., and Penman, Sci. USA 67, 1878-1885.

M. (1970).

Proc.

Peterkofsky,

Proc.

Bresler,

Proc.

Pastan,

R. J.

Sci. USA 68, 2127-2130.

Martin, D. W., Jr., Tomkins, G. M.. and Proc. Nat. Acad. Sci. USA 63, 842-849. Murphy, 115-119.

Randall,

B., and Tomkins,

Reel, J. R., and Kenney, 61, 200-206.

G. M. (1967).

F. T. (1968).

Proc.

248,

Nat. Acad.

J. Mol. Biol. 30. 49-61. Nat. Acad.

Sci.

Scornik, 1066. Scott, Acad.

USA

F. C. (1971).

J. Biol. Chem.

G., Urban, J., Zahringer. J., Reutter, J. Biol. Chem. 246, 4531-4538. 0. A. (1972).

W., and

Frosch.

Res. Commun.

47, 1063-

W. A., Shields, R., and Tomkins, Sci. USA 69, 2937-2941,

G. M. (1972).

Proc.

Sellers,

Biochem.

246, 6480-

Biophys.

L., and Granner,

Siev, M., Weinberg, 51 O-520.

D. (1974).

R.. and

Penman,

Nat.

J. Cell Biol. 60, 337-345. S. (1969).

J. Cell Biol. 47,

Singer,

R. H., and Penman,

S. (1972).

Nature

Singer,

R. H., and Penman,

S. (1973).

J. Mol. Biol. 78, 321-334.

240,

100-102.

Steinberg, R. A., Scott, W. A., Levinson, B. B., Ivarie, R. D., and Tomkins, G. M. (1974a). In Regulation of Gene Expression in Eucaryotic Cells, Fogarty International Center Proceedings, No. 25 (Washington, D.C.: U.S. Government Printing Office), in press. Steinberg, R. A., Levinson, (1974b). Proc. Nat. Acad. Sci. Thompson, Nat. Acad.

B. B., and USA, in press.

E. B., Tomkins, G. M., and Curran, Sci. USA 56, 296-303.

Thompson, E. B., Granner, J. Mol. Biol. 54, 159-175.

D. K.. and

Tomkins,

G.

J. F. (1966).

Tomkins.

M.

Proc.

G. M. (1970).

Tilghman, S. M., Hanson, R. W., Reshef, L., Hopgood, M. F., and Ballard, F. J. (1974). Proc. Nat. Acad. Sci. USA 71, 1304-1308. Tomkins, L. (1972).

240, 102-104.

Grisham. J. W. (1989). In Normal and Malignant Cell Growth, R. J. M. Fry, M. L. Griem, and W. H. Kirsten, eds. (New York: SpringerVerlag), pp. 28-43. Hershko, 714.

J. C., and Kafatos,

G. M., Levinson, B. B., Baxter, Nature New Biol. 239, 9-14.

J. D., and

Dethlefsen,

Van Dilla, M. A., Trujillo. T. T., Mullaney, R. (1989). Science 163, 1213-1214.

P. F., and

Coulter,

Van Dilla, M. A., Steinmetz, L. L., Davis, D. T., Calvert, Gray, J. W. (1974). IEEE Transactions Nuclear Science 714.

J.

R. N., and NS-21, p,