Cytosol induces apparent selectivity of glucocorticoid receptor binding to nucleic acids of different secondary structure

Biochimi~ et BiophysicaActa, 699 (1982) 53-59 Elsevier Biomedical Press

53

BBA 91128

C Y T O S O L I N D U C E S A P P A R E N T S E L E C T I V I T Y OF G L U C O C O R T I C O I D R E C E P T O R BINDING T O N U C L E I C ACIDS OF D I F F E R E N T S E C O N D A R Y S T R U C T U R E GEORGE A. ROMANOV and BORIS F. VANYUSHIN

A.N. Belozersky Laborato~' of Molecular Biology and Bioorganic Chemistr),', Moscow State Universi(v, Moscow 117234 (U. S. S. R.) (Received February 9th, 1982) (Revised manuscript received June 8th, 1982)

Key words: Glucocorticoid receptor," DNA binding," RNA binding," (Rat liver)

Unpurified rat liver glucocorticoid-receptor complexes within cytosol show a distinct binding preference for double-stranded DNA over single-stranded DNA; the binding to Escherichia coli rRNA is negligible. Extensive purification of the receptor abolishes its ability to distinguish among DNAs of different secondary structure and the affinity of the purified receptor toward RNA is greatly enhanced, reaching 30-50% of that of DNA. The purification effect is reversible: after cytosol addition to purified receptor preparation the binding preference restores. NaCI does not mimic the effect of cytosol. The flow-through fraction of a phosphocellulose column retains the ability of crude cytosol to produce selective decrease in the receptor binding to single-stranded DNA. This effect may also be observed by using two types of DNA-cellulose bearing double-stranded or denatured DNA, pretreated with crude cytosol. Additionally, pretreatment of immobilized DNA with even low cytosol concentrations has been shown to markedly enhance receptor binding, although this enhancement was lacking specificity with respect to DNA secondary structure. The nature of cytosolic active principle and some possible regulatory implications are discussed.

Introduction Steroid-receptor complexes from various target tissues have been shown to be DNA-binding proteins (for review, see Ref. 1). D N A seems to be necessary for translocation of steroid receptors from cytoplasm to the nucleus [1-3] and may prove to be their general acceptor in the cell. Unfortunately, the nature of the interaction between the steroid receptor and D N A is not yet known. Double-stranded D N A has been reported to have a preference in binding over single-stranded D N A [4-8] and D N A over R N A [4,5,7,9-11]. However, other data indicate that steroid-receptor complexes can bind denatured D N A and R N A at least as well as native D N A [3,12-14]. The present study is concerned with unraveling these discrepancies. Cytosol contaminants were found to 0167-4781/82/0000-0000/$02.75 © 1982 Elsevier Biomedical Press

exert a selective inhibitory influence on the interaction of glucocorticoid receptor with single-stranded D N A and RNA.

Materials and Methods Radioactive [1,2 - 3 H]dexamethasone (spec. act. 27 C i / m m o l ) and unlabeled dexamethasone were from Amersham and Sigma, respectively. The purity of labeled and unlabeled dexamethasone was tested by thin-layer chromatography on Silufol (chloroform/absolute ethanol, 9:1, v / v ) and was found to be close to 98% in both cases. The detailed procedure for D N A isolation, shearing and renaturation has been described previously [5,15]. The partially fragmented (s20,w = 12-13 S) E. coli ribosomal R N A was a gift from Prof. A.A. Bogdanov. For DNA-cellulose preparation, we

54 used Sigma Cell type 20 fine crystalline cellulose and high molecular weight double-stranded calf DNA. The coupling of D N A to the cellulose was accomplished by a combination of lyophilization [16] and ultraviolet irradiation [17]. A very stable product containing 20-30 mg D N A per g dry cellulose has been obtained. A similarly prepared DNA-cellulose carrying denatured calf D N A was used in some experiments. All the operations for the labeling and purification of glucocorticoid-receptor complexes were carried out at 0 4°C. Male white noninbred rats were killed by cervical dislocation. The livers were perfused in situ via the interior vena with ice-cold buffer A (0.02M Tris-HC1 (pH 7.4 at 20°C)/1.5 mM EDTA), finely minced and homogenized in a Teflon-glass motor-driven Potter-Elvehjem homogenizer with an equal volume of buffer A, using ten strokes. The soluble cytoplasmic fraction was obtained by centrifugation at 100000 X g for 120 rain at 0-2°C. The lipid-free upper portion of the supernatant containing approx. 20 m g / m l protein was used as cytosol. For receptor purification a two-step DNA-cellulose batch procedure alternating with ammonium sulfate precipitation was employed [11]. Briefly, 15 18 ml of cytosol were incubated with a great amount of DNA-cellulose (about 15-20 mg DNA, DNA-cellulose I) for 10 min and centrifuged. The supernatant was incubated with an excess of [3H]dexamethasone (usually (20 30) • 106 dpm) for 1.5-2 h. The radioactive steroid-receptor complexes thus formed were precipitated (and activated) by the addition of ammonium sulfate to 30% saturation. The pellet

was dissolved in l0 ml buffer A and incubated with the second portion of DNA-cellulose ( 1-2 mg DNA, DNA-cellulose II). Steroid receptors bound to DNA-cellulose II were then extensively washed three times with buffer A / 4 0 mM NaC1 and eluted with a small volume of buffer A / 0 . 4 M NaC1. This procedure allows us to obtain the 1000 2000-fold purified glucocorticoid-receptor complexes of a total radioactivity of (1-2). 106 dpm from liver of 3-5 rats (Table I). The use of the batch procedure made possible fast purification of the receptor without any heating. The preparation was tested for nuclease activity by comparison of the sedimentation coefficients of the polynucleotides before and after incubation with partially purified dexamethasone-receptor complexes and by measuring the quantity of DNA that could be released from DNA-cellulose after the incubation with the receptors. No nuclease activity in the purified receptor preparations could be detected by either procedure. The binding of [3H]dexamethasone-receptor complexes to nucleic acids was estimated by DNA-cellulose equilibrium competition experiments. 0.5 ml of the radioactive complexes (20000-60000 dpm) was added to the mixture of DNA-cellulose (usually 60-100 /~g DNA) and various amounts of sheared D N A or RNA. The final volumes of the binding mixtures were 0.7 0.8 ml in buffer A, NaC1 concentration was, as a rule, about 40 mM. The tubes were incubated in the cold for 30-50 min with periodical shaking and centrifuged at 3000 × g for 5 rain. The radioactivity in the DNA-cellulose pellet was extracted by

TABLE I PURIFICATION OF THE [3H]DEXAMETHASONE-RECEPTORCOMPLEX FROM RAT LIVER CYTOSOL Cytosol from three rat livers was used. The amount of receptor protein was calculated as the difference between the charcoal-resistant bound radioactivity in the presence or absence of unlabeled dexamethasone [5]. Source of glucocorticoid-receptorcomplex

Total volume (ml)

Total protein (mg)

Receptor purification (-fold)

Recovery (%)

Cytosol Cytosol, treated by DNA-cellulose 1 Buffer A extract of ammonium sulfate precipitate DNA-cellulose lI eluate

18 18 10 2

360 317 20.2 0.06

1 1.05 7.6 1632

100 92.5 42.6 27.2

55 ethanol and measured in a Mark III liquid scintillation spectrometer (Searle-Nuclear Chicago) with a counting efficiency of about 40%. The background counting with plain cellulose was subtracted. In some cases, the amount of the [3H]dexamethasone-receptor complexes in the supernatant was determined by the hydroxyapatite adsorption technique. The co-precipitation of free nucleic acids with DNA-cellulose and the elution of D N A from DNA-cellulose during the incubation period were negligible. The analytical procedures were as previously described [11,15]. For the kinetic analysis, the suspension of DNA-cellulose II with bound purified [3H]glucocorticoid-receptor complexes was transferred into standard tubes (0.2 ml slurry per tube). At zero time, different amounts of nucleic acids in 0.5 ml buffer A were added and the tubes were incubated for 60 min at 0°C. Then the samples were processed as in the equilibrium competition experiments described above. The binding of the purified glucocorticoid-receptor complexes to the two different DNA-cellulose preparations pretreated with cytosol was also performed. For this purpose the tubes containing equal amounts (45 /~g DNA) of DNA-cellulose (double-stranded or denatured) or pure cellulose were supplemented with 0.9 ml initial cytosol, diluted cytosol or buffer A, respectively. The suspensions were thoroughly mixed and incubated at 0°C for 15 min. Then DNA-cellulose or cellulose was pelleted by brief centrifugation, the supernatant was quantitatively discarded. 0.5 ml of purified [3H]glucocorticoid-receptor complexes in buffer A containing 0.1 m g / m l human serum albumin were added to the pellets. The mixtures were vortexed and incubated for another 15 min at 0°C. At the end of incubation, DNA-cellulose or cellulose was pelleted, washed with 1 ml buffer A and the bound radioactivity was determined.

Results Competition analysis is a good means to evaluate the affinity of steroid-receptor complexes for various nucleic acids allowing rapid processing of multiple samples. We have examined the ability of E. coli ribosomal R N A and calf D N A in two different forms, single- and double-stranded to

prevent glucocorticoid receptor from binding to DNA-cellulose. When unpurified receptor within crude cytosol was used, all the nucleic acids tested drastically differ in this regard (Fig. 1), The double-stranded D N A competes most strongly, the total denatured D N A and its unique sequences (Cot > 400) are much less effective, whereas R N A is essentially inactive at the concentrations used. But if the purified glucocorticoid-receptor complexes were employed, striking changes in the relative competitive efficiency of these nucleic acids were observed (Fig. 2). The binding activity of denatured D N A becomes nearly equal to that of double-stranded DNA. The binding activity of R N A is enormously enhanced and becomes comparable to that of DNA, accounting for as much as 30-50% of the latter (Fig. 2). In the other type of experiment, the capability of nucleic acids to favor the release of the purified [3H]glucocorticoid-receptor complexes from DNA-cellulose has been investigated (Fig. 3). The results are just the same as in the equilibrium competition experiments. There is no difference in the effectiveness of double-stranded or denatured

%

100 : . . . . . . . . . . . . .

i

:

~

2

i

'

g

R

Fig. 1. Inhibition of binding of [3H]dexamethasone-receptor within crude liver cytosol to DNA-cellulose by soluble DNA and RNA. Increasing amounts of soluble nucleic acids were mixed with DNA-cellulosein buffer A and 0.5 ml of activated [3H]dexamethasone-labeled cytosol was added. After incubation for 30 min at 0°C the samples were centrifuged, and the radioactivity in the DNA-cellulosepellet was determined. The radioactivity of [3H]dexamethasone nonspecifically (with 300fold excess of unlabeled dexamethasone) bound to DNA-cellulose was subtracted. R means the monomer ratio of soluble competitive nucleic acid to DNA coupled on cellulose. Results are expressed as percentage of binding without any soluble nucleic acid added. Each experiment was performed in duplicate. Competitioncurves: double-stranded total calf DNA (O); denatured total calf DNA (D); denatured unique (Cot >400 ) calf DNA sequences (I); E. coli rRNA (e).

56 1

80~\\\\\ 4,0. ~ ,

\\ ~d~

R

Fig. 2. Inhibition of purified [3H]dexamethasone-receptor binding to DNA-cellulose by soluble D N A and RNA. Various amounts of soluble nucleic acids were mixed with DNA-cellulose in buffer A and then [3H]dexamethasone-receptor complexes purified about 1000-fold were added. The mixture was incubated for 40 rain, and centrifuged. The radioactivity of [3H]dexamethasone-receptor binding to DNA was determined by subtracting the dpm bound to uncoupled cellulose from the dpm bound to DNA-cellulose. Each experiment was performed in duplicate. Axis designations are as in Fig. 1. Competition curves: double-stranded (O) and denatured (1~) total calf DNA; E. coli rRNA (0).

D N A in p r o m o t i n g the release of the r e c e p t o r f r o m D N A - c e l l u l o s e . M o r e o v e r , u n i q u e (C0t > 400) calf DNA sequences, lacking any long

d o u b l e - s t r a n d e d D N A stretches, d i s p l a y a n activity similar to those of o t h e r D N A s tested. A n e q u i m o l a r m i x t u r e of four m a j o r d e o x y r i b o n u c leoside m o n o p h o s p h a t e s fails to release 3 H - l a b e l e d c o m p l e x e s f r o m D N A - c e l l u l o s e (Fig. 3). F o r a b e t t e r u n d e r s t a n d i n g of the n a t u r e of the c y t o s o l effect, s o m e a d d i t i o n a l e x p e r i m e n t s with the p u r i f i e d r e c e p t o r have b e e n a c c o m p l i s h e d (Fig. 4). As a b o v e (Figs. 2,3), Fig. 4 A shows that the p u r i f i e d g l u c o c o r t i c o i d - r e c e p t o r c o m p l e x e s b i n d to single- or d o u b l e - s t r a n d e d D N A a l m o s t equally, b u t w h e n cytosol is a d d e d to the p u r i f i e d receptor, the b i n d i n g of the latter to d e n a t u r e d D N A is s i g n i f i c a n t l y r e d u c e d (Fig. 4C). T h e cytosol effect c a n n o t be i m i t a t e d b y a salt: in spite of NaC1 a d d i t i o n to 0.1 M, the p u r i f i e d r e c e p t o r disp l a y s s i m i l a r affinities to single- a n d d o u b l e s t r a n d e d D N A s (Fig. 4B). A n a t t e m p t to lay a finger o n the active p r i n c i p l e has b e e n m a d e b y p a s s i n g cytosol t h r o u g h a p h o s p h o c e l l u l o s e colu m n . T h e f l o w - t h r o u g h r e t a i n s the a b i l i t y of total cytosol to i n h i b i t the b i n d i n g of the glucoc o r t i c o i d - r e c e p t o r c o m p l e x e s to d e n a t u r e d D N A (Fig. 4D). In o r d e r to find o u t w h e t h e r the cytosol active p r i n c i p l e c a n b i n d directly to nucleic acid, the

o~400~

A

-

C

°/. ,oo,~. ~

60

~X

40. 60 r

10;

"

x

~

=

•

2O IG

$

6 20

40~

20I1

Fig. 3. Transfer of purified [3Hldexamethasone-receptor from DNA-cellulose to soluble DNAs. Soluble DNAs were added to purified [3H]dexamethasone-receptor complexes bound to DNA-cellulose II (see Materials and Methods). The mixture was incubated in buffer A at 0°C for 60 min, and centrifuged. The radioactivity remaining bound to DNA-cellulose is expressed as per cent of binding without soluble DNA added. Axis designations are as in Fig. 1. Transfer to: double-stranded (O) and denatured ( ~ ) total calf DNA; denatured unique calf DNA sequences (11); mixture of the four major deoxyribonucleoside 5'-monophosphates (A).

0

\\

~

D

x~x

N o 2

*

R

i

i

Fig. 4. Cytosol effect on the competition for purified [3H]glucocorticoid-receptor by double-stranded and single-stranded DNA. Purified [3H]dexamethasone-receptor complexes were mixed with DNA-cellulose and various amounts of soluble D N A s in buffer'A, and the mixture was supplemented by equal volume of: buffer A (A); NaCI to final concentration of 0.1 M (B); rat liver cytosol (C) and cytosol, passed through phosphocellulose (D). Axis designations are as in Fig. 1. Competition curves: double-stranded calf DNA (O, solid line) and denatured calf DNA (B, dotted line).

57

DNA-celluloses bearing equal amounts of doubleor single-stranded DNA were pre-incubated with rat liver cytosol at different concentrations. Then DNA-cellulose with the bound cytosol components (proteins) were used in the study of purified glucocorticoid receptor-DNA interaction (Table II). The use of plain cellulose and the inactivated (60°C, 15 rain) receptor shows that both intact glucocorticoid-receptor complexes and cellulosecoupled D N A are necessary for substantial binding. One can see (Table II) that cytosol pretreatment of denatured DNA-cellulose decreases its receptor binding ability as compared to double-stranded DNA-cellulose ( P < 0.01). The amount of cytosol protein bound to immobilized DNA exceeded the latter by a factor of 2 (w/w); the protein: DNA ratio was slightly (by 10-15%) higher in the case of denatured DNA-cellulose. The cytosol inhibition effect was evident only if the initial cytosol (20 m g / m l protein) was employed. Diluted cytosol or buffer A produces no such effect (Table II). It should be noted that the pretreatment of DNA-cellulose with even very diluted cytosol resulted in an approx. 2-fold increase of the glucocorticoid receptor binding to both DNA-cellulose

preparations. This effect must be due to the direct interaction of cytosol with DNA, since no enhancement was observed in the case of plain cellulose (Table II). Although the nature of this stimulatory influence remains to be elucidated, this proves that cytosol components may exert a profound and complex effect on the process of receptor-DNA interaction. Discussion

Numerous studies, over a wide spectrum of cell systems, indicate that the activated steroid-receptor complex exhibits a distinct preference to certain nucleic acids, which depends on their primary a n d / o r secondary structure [1,4-14,16,19-21]. However, none of these experiments, carried out with the use of poorly purified or unpurified receptor proteins, can be regarded as convincing, since cytosol contaminants can alter the binding of steroid-receptor complexes to nucleic acids. In order to elucidate if cytosol has any influence on the binding preference observed, we compared the binding of highly purified and unpurified rat liver glucocorticoid-receptor complexes to D N A and R N A . It was shown that after extensive purifica-

T A B L E II E F F E C T OF CYTOSOL P R E T R E A T M E N T ON T H E B I N D I N G OF P U R I F I E D G L U C O C O R T I C O I D - R E C E P T O R COMPLEXES TO D O U B L E - S T R A N D E D A N D S I N G L E - S T R A N D E D D N A - C E L L U L O S E DNA-cellulose (45/Lg DNA) or plain cellulose were pretreated with 0.9 ml of initial cytosol (20 m g / m l protein), cytosol diluted with buffer A, or buffer A alone. Then 0.5 ml purified [3 H]glucocorticoid-receptor complexes were added to the DNA-cellulose or cellulose pellets. The solution of purified receptor contained 0.1 m g / m l h u m a n serum albumin (DN A)-cellulose pretreated with:

Binding (dpm) of: activated complexes to:

Cytosol Cytosol, diluted 3-fold Cytosol, diluted 9-fold Buffer A

double-stranded DNA-cellulose

denatured DNA-cellulose

plain cellulose

1758 1713 1772 1909 1866 1774 833 971

1429 1424 1767 1788 1701 1609 881 879

234 281 338 357 293 288 372 412

inactivated (60°C, 15 min) complexes to: double-stranded DNA-cellulose 115 144 117 135 127 111 184 214

58 tion the binding preference of glucocorticoid receptor toward double-stranded DNA is essentially lost (Figs. 1-4), whereas its ability to discriminate between the avialable nucleotide bases, i.e., recognize the primary structure of a nucleic acid, is retained [16]. The relative affinity of glucocorticoid-receptor complexes toward denatured DNA and especially toward RNA was observed to be greatly enhanced after purification. Thus tyrosol contaminants may produce a prominent specific effect on the interaction of glucocorticoid receptor with nucleic acids by reducing the receptor binding to RNA and single-stranded DNA as compared to double-stranded DNA. In addition, the cytosol influence was shown to be complex, depending on the cytosol protein concentration. Specific inhibition of the receptor interaction with single-stranded DNA was shown to occur only if concentrated cytosol was employed (Table II). Also, cytosol at a much lower concentration can enhance the glucocorticoid receptor binding to DNA irrespective of the DNA secondary structure (Table II). The discrepancies between some data cited [ 1-14,16,19-21 ] may arise from uncontrolled cytosol contaminants in the receptor preparations. Earlier works where essentially unpurified receptors were used showed a binding preference for double-stranded DNA over single-stranded DNA and negligible RNA binding [4-7,9-11]. On the other hand, recent reports, dealing as a rule with more purified receptor preparations, showed a significant and comparable affinities toward all the three nucleic acids mentioned above [12-14]. The indication in one of these reports [13] that RNA, purified from MTW9 rat mammary tumor cytosol, is more potent than the initial crude preparation in preventing estradiol receptor binding to DNA-cellulose, is in good agreement with the present study. One must also bear in mind that cytosols from different tissues may contain dissimilar active factors or the same factor but in different concentrations. Another problem posed by this work is the nature of the cytosolic factor(s) influencing the interaction of glucocorticoid-receptor complexes with single-stranded DNA and RNA. Further work is needed to settle this question, but some preliminary suggestions can be made. Pretreatment of DNA-cellulose with cytosol showed that this fac-

tor may be directly associated with DNA without any degradation of the latter. So, it is hardly probable that this factor is some receptor subunit or nuclease. Other, less direct, evidence also agrees with this standpoint. It appears reasonable to assume that cytosolic nucleic acid binding proteins are those that inhibit the receptor binding to RNA and single-stranded DNA. This possibility is supported by the known fact that cytosol contains proteins that can bind to RNA and DNA in the experimental conditions close to those reported here [11,24,25]. The RNA-binding proteins may account for as much as 1-5% of the total cytosolic protein [24]. These proteins may evidently also bind to denatured DNA, the latter being similar in structure to RNA molecules. In our experiments, both types of immobilized DNA efficiently bound cytosolic protein, denatured DNA-cellulose doing somewhat better than the double-stranded one. It is unlikely that the cytosol effect is a result of binding to and blocking of DNA (or RNA) phosphate groups, since: (i) the cytosolic active principle does not adsorb on phosphocellulose, phosphate groups of the latter being analogous to DNA phosphates; (ii) NaC1 is unable to imitate the cytosolic effect, although Na + produces a screening ion atmosphere around each charged group; and (iii) the interaction of steroid receptor with nucleic phosphates seems to be of only minor significance in the process of receptor-DNA association [12 14,16,19-21]. A note should also be made that the bulk of DNA-binding proteins in rat liver cytosol, as reported recently [25], does not bind to phosphocellulose. The third problem pertains to the properties of glucocorticoid receptor to be not only a DNA-binding but also a RNA-binding protein (Fig. 2). This feature may implicate RNA and RNA-binding proteins in the in vivo process of steroid hormone action. The greater cellular RNA concentration and the lower the concentration of RNA-binding proteins, the smaller part of steroid-receptor complexes will reach nuclear DNA. Since the latter event is believed to be a key one in the mechanism of steroid hormone action, the cellular concentrations of RNA and RNA-binding proteins together with the concentration of the specific receptor proteins may determine the magnitude of steroid response. The RNA-binding property of gluco-

59

corticoid receptor can provide a negative feedback mechanism allowing the attenuation of glucocorticoid-receptor complexes influence on chromatin after massive hormone uptake. Indeed, hormonal induction leads to transcription enhancement in the liver cells with the RNA concentration in the entire cell and especially in the nucleus being increased, which ensures the conditions for retranslocation of glucocorticoid-receptor complexes from chromatin to ribonucleoprotein particles. The transcription might therefore become less active. It should be noted that steroid-receptor complexes were already found to be a minor component of cellular ribonucleoprotein [26]. Finally, a similar mechanism of transcription regulation with the participation of an activator protein that binds both DNA template and its RNA product, has been recently described [27].

Acknowledgements The authors are grateful to Professors V.B. Rozen and A.A. Bogdanov for fruitful discussions and N.A. Romanova for technical assistance.

References 1 Yamamoto, K.R. and Alberts, B. (1976) Annu. Rev. Biochem. 45, 721-746 2 Milgrom, E., Atger, M. and Bailly, A. (1976) Eur. J. Biochem. 70, 1-7 3 Romanova, N.A., Romanov, G.A., Rozen, V.B. and Vanyushin, B.F. (1978) Dokl. Akad. Nauk SSSR 243, 15851588 4 Rousseau, G.G., Higgins, S.J., Baxter, J.D., Gelfand, D. and Tomkins, G.M. (1975) J. Biol. Chem. 250, 6015-6021 5 Romanov, G.A., Sokolova, N.A., Rozen, V.B. and Vanyushin, B.F. (1976) Biokhimiya 41, 2140-2141

6 Andr& J., Pfeiffer, A. and Rochefort, H. (1976) Biochemistry 15, 2964-2969 7 Kallos, J. and Hollander, V.P. (1978) Nature 272, 177-179 8 Thrall, C.L. and Spelsberg, T.C. (1980) Biochemistry 19, 4130-4138 9 Milgrom, E., Atger, M. and Baulieu, E.-E. (1973) Biochemistry 12, 5198-5205 10 Yamamoto, K.R. and Alberts, B. (1974) J. Biol. Chem. 249, 7076-7086 11 Romanova, N.A., Romanov, G.A., Rozen, V.B. and Vanyushin, B.F. (1979) Biokhimiya 44, 529 542 12 Liao, S., Smythe, S., Tymoczko, J.L., Rossini, G.P., Chen, C. and Hiipakka, R.A. (1980) J. Biol. Chem. 255, 5545-5554 13 Feldman, M., Kallos, J. and Hollander, V.P. (1981) J. Biol. Chem. 256, 1145-1148 14 Lin, S. and Ohno, S. (1981) Biochim. Biophys. Acta 654, 181-186 15 Romanov, G.A. and Vanyushin, B.F. (1981) Biochim. Biophys. Acta 653, 204-218 16 Romanov, G.A., Romanova, N.A., Rozen, V.B. and Vanyushin, B.F. (1981) Molek. Biolog. 15, 857 874 17 Alberts, B. and Herrick. G. (1971) Methods Enzymol. 21, 198-217 18 Litman, R.M. (1968) J. Biol. Chem. 243, 6222-6233 19 Simons, S.S. (1977) Biochim. Biophys. Acta 496, 349-358 20 Kallos, J., Fasy, T.M., Hollander, V.P. and Bick, M.D. (1979) FEBS Lett. 98, 347-350 21 Kumar, S.A., Beach, T.A. and Dickerman, H.W. (1980) Proc. Natl. Acad. Sci. U.S.A. 77, 3341 3345 22 Wrange, 6 , Carlstedt-Duke, J. and Gustafsson, J.-A. (1979) J. Biol. Chem. 254, 9284-9290 23 Govindan, M.V. (1980) Biochim. Biophys. Acta 631, 327333 24 Preobrazhensky, A.A. and Spirin, A.S. (1978) in Progress in Nucleic Acid Research and Molecular Biology (Cohn, W.E., ed.), pp. 1-38, Academic Press, New York 25 Westtphal, H.M. and Beato, M. (1981) FEBS Lett. 124, 189-192 26 Liang, T. and Liao, S.J. (1974) J. Biol. Chem. 1974, 279, 4671-4678 27 Pelham. H.R.B. and Brown, D.D. (1980) Proc. Natl. Acad. Sci. U.S.A. 77, 4170-4174