421

Biochimica et Biophysica Acta, 561 (1979) 421--434 © Elsevier/North-Holland Biomedical Press

BBA 99396

FRACTIONATION AND CHARACTERIZATION OF RAT LIVER POLY(A)-CONTAINING RNA

JAMES DENNIS and ROBERT KISILEVSKY

Departments of Pathology and Biochemistry, Queen's University and Kingston General Hospital, Kingston, Ontario (Canada)

(Received July 14th, 1978)

Key words: Poly(A)-containing RNA; Poly(U) sequence; mRNA

Summary

Three fractions of poly(A)-containing RNA were separated from total rat liver RNA using poly(U)-Sepharose 4B affinity chromatography. The poly(A)- containing RNA fractions were released by thermal elution. Fraction 1, eluted under the mildest conditions, and had poly(A) tracts of approx. 200 AMP units in length which appeared to be associated with poly(U) sequences of 20-- 50 UMP in length. Fraction 1 appeared to be present mainly in the nucleus and, its size distribution was similar to that of fractions 2 and 3. Fractions 2 and 3 eluted at higher temperatures and were associated mainly with polysomal and microsomal fractions. Poly(U) sequences were absent in fractions 2 and 3 while their poly(A) sequences had a size distribution characteristic of those reported in the mRNA of other organisms.

Introduction

Isolation of poly(A)-containing messenger RNA by affinity chromatography has resulted in the rapid proliferation of literature concerning the structure and function of messengers in a variety of cells. In the course of studying mRNA metabolism in an induced pathological state, we examined several techniques for isolating and characterizing poly(A)-containing mRNA from total rat liver RNA. The procedure described herein has allowed us to separate two major poly(A)-containing RNA fractions. One fraction has characteristics of a nuclear species, while the other appears to be cytoplasmic in origin. The present paper describes the isolation and characterization procedures.

422

Materials and Methods

Animals. Sprague-Dawley female rats fasted overnight and weighing 200-- 250 g were used in all experiments. Animals were killed by decapitation using a guillotine.

RNA extraction. The livers were homogenized in 10 volumes of homogeniz- ing buffer (25 mM NaC1, 5 mM MgC12, 25 mM Tris-HC1, pH 7.5) using 5--10 strokes of a Potter-Elvehjem homogenizer. RNA was extracted by the sodium dodecyl sulfate (SDS)-phenol-chloroform method of Palmiter [1], and precipi- tated overnight at --20°C using 2 volumes of ethanol. It was pelletted at 48 000 X g for 15 rain at --10°C in a Sorval RC-2B centrifuge. The pellet was drained for 15 min and redissolved in SDS buffer (0.5% SDS, 0.1 M sodium acetate, pH 5.0) to a concentra t ion of about 1 mg/ml. A second phenol- chloroform extract ion followed and the RNA precipitated and pelletted as before. The drained pellets were washed once more in SDS buffer followed by a final ethanol precipitation.

Subcellular fractionation. Rat livers were homogenized in 3 volumes of 0.25 M sucrose, 50 mM Tris-HC1, pH 7.5, 25 mM KC1, 5 mM MgC12 (Buffer A) as described above. The homogenates were spun in a Sorval RC-2B centrifuge at 19 000 ~ g for 10 min at 4°C. The post-mitochondrial supernatant was layered on a discontinuous sucrose gradient to separate polysomes, membrane-contain- ing fractions and cytosol as described by Staehelin et al. [2]. The cell debris, nuclear and mitochondrial pellet was mixed with 2.1 M sucrose in Buffer A and spun at 54 500 • g for 1 h at 4~C in a Beckman Model L ultracentrifuge. A nuclear pellet was thereby isolated. The pellet was washed in 5~ Tri ton X-100 in Buffer A and the nuclei were repelletted by spinning at 1000 :y g for 10 rain at 4°C [3]. A membrane-free nuclear pellet was obtained as verified by electron microscopy. RNA was extracted from each cell fraction as described above.

Isolation of poly(A)-containing RNA. RNA isolated from liver homogenates, or cell fractions, was taken up to a concentrat ion of 1--2 mg/ml in binding buffer (0.2 M NaC1, 0.2% SDS, 5 M EDTA, 10 mM Tris-HC1, pH 7.5). Samples were applied to a poly(U)-Sepharose 4B column (0.9 ~' 15 cm, Pharmacia) previously equilibrated with binding buffer. The sample was stirred with approx. 2/3 of the column bed volume for 30 rain at room temperature (20°C). The remainder of the poly(U)-Sepharose was left in the column. The column was repacked and washed with approx. 5 column volumes of binding buffer at a rate of 25 ml/h at 20~C. Poly(A)-containing RNA was eluted in three steps. Fract ion 1 was released when the column was washed with no salt buffer (5 mM EDTA, 0.2~5 SDS, 10 mM Tris-HC1, pH 7.5) at 20cc. The column was then placed in a water bath heated to 42'~C and cont inued elution with the no salt buffer released fraction 2. The temperature was then raised to 50°C releas- ing a third fraction. Raising the temperature again did not result in the recovery of further fractions. The eluate was continuously moni tored at 254 nm using an ISCO Model UA-2 ultraviolet analyzer.

Isolation and sizing of poly(A)-sequences. Poly(A)-containing RNA was digested with 10 ,g /ml of pancreatic RNAase (Nutritional Bioehemieals Co.) and 10 units/ml of T~ RNAase (Sigma) in 0.3 M NaC1, 10 mM Tris-HC1, pH 7.5, for 30--45 min at 37°C. Digestion was terminated by the addition of SDS to a

423

concentration of 0.2% and 10 t~g/ml of protease V (Sigma) for 15 min at 37°C [4]. The poly(A) sequences were then applied to the poly(U) column as above and eluted at 42°C only. The size of the poly(A) sequences were determined on 10, 8 and 4% polyacrylamide gels in 98% formamide as described by Wu and Wilt [5].

RNA base composition. RNA samples were taken up in 50 pl 5 mM MgC12, 50 mM Tris-HC1, pH 8.5, and incubated for 4 h at 37°C with 10 ~g alkaline phosohatase and 25 ~g snake venom phosphodiesterase (Worthington Biochemical Corp.) [6]. The nucleosides produced by enzymatic digestion were treated with NaIO4 and NaB3H4 (200 Ci/mol New England Nuclear) [7]. Nucleoside standards (Sigma Biochemical Corp.) were treated in the same manner except that unlabelled NaBH4 was used. Samples were spot ted on strips of Whatman No. 1 paper and nucleosides separated by descending paper chromatography in water-saturated butanol [8]. The standards were located visually by flourescence under ultraviolet light. The paper was cut into 1-cm pieces, and placed in 10 ml of a counting mixture of 0.4% omnifluor (New England Nuclear) in toluene for determination of radioactivity.

Detection o f poly(U) sequences by [3H]poly(A) hybridization. Various quantities of synthetic poly(U) (Sigma Biochemical Corp.), or isolated fractions of poly(A)-containing RNA, were taken up in 50 pl 0.1 M NaC1, 2 mM Na2EDTA, 50 mM Tris, pH 7.2, and heated to 90°C for 10 min. 1 pCi [3H]- poly(A) (2.9 Ci/mmol New England Nuclear) was added to a final volume of 100 pl and the sample cooled at room temperature for 30 min. Samples were made up to 0.5 M NaC1 by the addition of 0.95 M NaC1, 2 mM Na2EDTA, 50 mM Tris, pH 7.2. Digestion of single-stranded RNA was carried out with 10 units of T: RNAase (Sigma Biochemical Corp.) and 10 ~g RNAase A for 30 min at 28°C [9]. RNA was then precipitated by the addition of 0.5 ml 10% trichloroacetic acid and 200 t~g carrier RNA. Precipitated RNA was pelleted at 10 000 × g for 10 min and washed four times in 10% trichloroacetic acid. Pellets were solubilized in 0.7 ml. Protosol (New England Nuclear) and 10 ml 0.4% Omnifluor in toluene was added followed by determination of radioactiv- ity.

Electron microscopy. A small sample of the nuclear pellet was fixed in 3% buffered glutaraldehyde in cacodylate buffer at pH 7.4 and post-fixed in 1% OsO4 in phosphate buffer, pH 7.4. The sample was then dehydrated in graded alcohol solutions, propylene oxide and embedded in Epon 812. Sections 0.25 pm thick were cut and stained with uranyl acetate and lead citrate [10,111.

RNA determinations. These were performed by the method of Fleck and Munro [12]. Where alkaline hydrolysis was not performed one Ae60nm unit was considered to be equal to 50 ug/ml of RNA [13].

Results

Affinity chromatography on poly(U)-Sepharose 4B The conditions necessary for the binding and elution of poly(A)-containing

RNA to a poly(U)-Sepharose 4B column were initially examined using synthe- tic poly(A) (Sigma, molecular weight greater than 10 000). Samples of approx.

424

1 mg were applied, and unbound material was washed from the column with binding buffer. When the eluate no longer showed detectable absorbance at 254 nm the column was washed with the no salt buffer. Poly(A) was not released by the no salt buffer at 20°C, but when the temperature of the column was raised to 42°C, 80--90% of the applied poly(A) was recovered (Fig. 1). In contrast total liver RNA had poly(A)-containing material with varying affinities for poly(U) and could be differentially eluted. Following application of the total cellular RNA, the unbound RNA was washed from the column with binding buffer. Bound RNA was released in three steps (Fig. 2). Fractions 1, 2 and 3 were released from the column at 20, 42 and 50°C and comprised 38, 48 and 12% of the column-bound RNA, respectively. The bound RNA was 0.6% of the total cellular RNA applied to the column. Similar yields of poly(A)- containing RNA have been reported by others [14,15]. Reapplication of the originally unbound material resulted in the additional binding of material equal to only 10% of the originally bound material.

Since synthetic poly(A) sequences could be released only when the tempera- ture of the column was raised to 42°C, it was clear that the binding of frac- tion 1, released at 20°C, was different from a simple poly(A) - po ly (U/dup lex . We considered two possibilities: (1) fraction 1 RNA may have shorter poly(A) tracts or (2) secondary structure in fraction 1 RNA may have caused a weaker poly(A) • poly(U) binding complex.

Sizing of poly(A ) tracts Poly(A) sequences were isolated from poly(A)-containing RNA fractions

with T~ and pancreatic ribonuclease as described in Materials and Methods and applied to 10, 8, and 4% polyacrylamide gels in 98% formamide. Fraction 1 poly(A) remained at or near the origin of the 10% gels but migrated as a doublet into the 8% gels {Figs. 3A and 3D). Similar results were obtained with 4% gels {Fig. 3B) where the poly(A) doublet again migrated behind the 5.5 S

0 2 -

NSB 42°C 50'PC

tO Od

0.1-

0 . 0 ~

IOml froctions Fig . 1. A n e l u t i o n p r o f i l e o f s y n t h e t i c p o l y ( A ) r e l e a s e d f r o m a p o l y ( U ) - S e p h a x o s e 4B c o l u m n . A s m a l l

a m o u n t o f m a t e r i a l f a i l ed t o b i n d to the c o l u m n . N o sa l t b u f f e r ( N S B ) a t 2 0 ° C f a i l e d t o re lease any

p o l y ( A ) . R a i s i n g the t e m p e r a t u r e to 4 2 ° C r e l e a s e d al l t h e b o u n d m a t e r i a l .

0 . 2 -

NSB

~ 0.1-

0 . 0 -

42oc +

1 50oC &

4 2 5

I I I I [ I I I I I

I0 ml fraction

Fig. 2. A n e l u t i o n p ro f i l e o f p o l y ( A ) - c o n t a l n i n g R N A re l eased f r o m a p o l y ( U ) - S e p h a r o s e 4B c o l u m n .

T o t a l ce l lu la r R N A i so l a t ed f r o m o n e r a t l iver was a p p l i e d t o t h e c o l u m n . P o l y ( A ) - c o n t a l n i n g R N A frac- t i o n s we re r e l eased w i th t h e n o sal t b u f f e r (NSB) a t 20, 4 2 a n d 5 0 ° C .

RNA. By the use of markers the length of the tracts was estimated to be about 200 AMP units. Fractions 2 and 3 were combined for this and other experi- ments to simplify the handling of the small amounts of fraction 3 isolated. In 10% polyacrylamide gels fraction (2 + 3) poly(A) tracts (Fig. 3C) showed a heterogeneous size distribution with species ranging from 50 to 160 AMP units in length as estimated by Wu and Wilt [5]. The majority were the shorter species migrating with tRNA. In 8% gels the poly(A) tract of fractions 1 and (2 + 3) could be easily separated (Fig. 3D). Base composit ion analysis con- firmed that the poly(A) of both fractions 1 and (2 + 3) were composed of approx. 95% AMP units (Table I). It appears that the weaker affinity of frac- tion 1 for the poly(U) column is not a result of short poly(A) sequences on the 3'-end of RNA.

Secondary structure in poly(A )-containing RNA In the absence of shorter poly(A) tracts, secondary structure associated with

the poly(A) of fraction 1 and not present in fractions 2 and 3 could produce different poly(U) binding capacities resulting in the separation of the RNA fractions by the poly(U)-Sepharose 4B column. Samples of fraction 1 and (2 + 3) RNA labelled with [3H]orotate were digested with RNAase T1 and A under conditions which allow the survival of double-stranded segments [15], Poly(A)-containing sequences were then reisolated on a poly(U) column. Fraction 1 poly(A)-contalning segments appeared to have more secondary structure than did fraction (2 + 3) as indicated by the greater degree of hyper- chromicity (Table II) and the presence of greater amounts of orotate incor- porated into the RNAase-resistant material of fraction 1 when compared to fraction (2 + 3). Base analysis of the enzyme-resistant RNA indicated 28% uracil, 70% adenine for fraction 1 and 6% uracil, 87% adenine for fraction (2 + 3) with only small amounts of cytosine and guanine present.

Single-stranded poly(A) survives digestion by RNAase T1 and A but only uracil in duplex structure would escape such digestion. Since the uracil in RNAase-resistant RNA of fraction 1 was isolated with poly(A) sequences on a

a

distance (cm)

5.5S 5s 4 S

11 1

426

6cm

d C

5.5S 5S 4S

111

0 distance (¢m) 6¢m

55S 5S 4S

b 28S 18S 4S

0 6cm

distance (cm)

distance (cm) 6 cm

Fig. 3. D e n s i t o m e t e r scans of p o l y ( A ) s e q u e n c e s in p o l y a c r y l a m i d e gels. Po ly (A) t r ac t s were i so la ted f r o m p o l y ( A ) - c o n t a i n i n g R N A f r a c t i o n s and 2 0 - - 6 0 Dg of p o l y ( A ) w a s appl ied to gels p r e p a r e d in 98%

f o r m a m i d e . A f t e r e l e c t ropho re s i s the gels were s t a ined o v e r n i g h t in w a t e r / f o r m a m i d e (1 : 1, v /v) con ta in - ing 0 .005% 'S t a ins All ' , 1 - e t h y l - 2 - [ 3 - ( 1 - e t h y l n a p h t h o - ( 1 , 2 d ) - t h i a z o l i n - 2 - y l i d e n e ) - 2 - m e t h y l p r o p e n y l ] - n a p h t h o - ( 1 , 2 d ) - t h i a z o l i u m b r o m i d e ( E a s t m a n , K o d a k Co.). Exces s s ta in was r e m o v e d by r ins ing the gels

in wa te r . T h e gels were s canned at 618 n m us ing an E-C 810 d e n s i t o m e t e r . (a) F r a c t i o n 1 p o l y ( A ) in an 8% gel. N o t e the d o u b l e t n e a t the or igin . (b) F r a c t i o n 1 p o l y ( A ) in a 4% gel. N o t e the d o u b l e t nea t the

f ron t . (c) F r a c t i o n (2 + 3) p o l y ( A ) s e q u e n c e s in a 10% gel. (d) F r a c t i o n s 1 a n d (2 + 3) p o l y ( A ) s e p a r a t e d on a single 8% gel. 5 S R N A is c o n s i d e r e d to be a p p r o x . 120 n u c l e o t i d e s in length , 4 S R N A is c o n s i d e r e d

to be a p p r o x . 80 n u c l e o t i d e s in length .

427

T A B L E I

BASE COMPOSITION OF POLY(A) ISOLATED FROM POLY(A)-CONTAINING RNA

Nucleosides produced by enzymatic digestion of isolated poly(A) tracts were labelled by oxidation with

NaIO4 and reduction with NaB3H4 followed by paper chromatography separation. The paper strips were cut into 1-cm sections and placed in an Omnifiuor-toluene containing mixture to determine radioactivity.

Expt . A U C G

F r a c t i o n 1 p o l y ( A ) (1) 94.7 0.8 3.2 1.3 (2) 96 .0 0.2 1.0 3.0

F rac t i on (2 + 3 )po l y (A) (1) 91 .0 2.2 3.8 3.0 (2) 95 .8 0.2 0.6 3.4

poly(U) column and is labelled, it is likely to exist in vivo as a poly(U) • poly(A) duplex.

Detection of poly(U) sequences [3H]Poly(A) was used as a hybridization probe for the detection of poly(U)

tracts in the poly(A)-containing RNA fraction. Poly(A) in a duplex structure with synthetic poly(U), or poly(U) sequences from the poly(A)-containing RNA would be stable in 0.5 M NaC1 and resistant to RNAase T2 [9]. The results in Fig. 4 show that single-stranded poly(A) is rapidly degraded by T2 RNAase. The protection of [3H]poly(A) is proportional to the concentration of synthetic poly(U), or fraction 1 RNA, added. Fraction (2 + 3) failed to protect the [3H]poly(A).

Fraction 1 ribonuclease-resistant duplex fragments, isolated in high salt con- centration were labelled in vitro using NaIO4 and NaB3H4 and then subjected to electrophoresis on 10% polyacrylamide gels in 98% formamide. If the poly(A) and poly(U) isolated in this manner were not covalently attached, they should be separated in these gels on the basis of their size. Approx. 60% of the RNA sequences stayed at the origin of the gel as was expected for the long poly(A)

T A B L E II

H Y P E R C H R O M I C I T Y A S S O C I A T E D W I T H T H E P O L Y ( A ) S E Q U E N C E S OF F R A C T I O N S 1 A N D 2

Three rats were each given an in jec t ion of 5 ~Ci [ 1 4 C ] o r o t i c acid. Rats were killed 15 rain later . Po ly(A)- con ta in ing R N A of f rac t ions 1 and 2 having specific act ivi t ies of 45 and 25 d p m / ~ g , respect ive ly , we re sub jec ted to l imi ted hydro lys i s using 2 ~g /ml pancrea t i c RNAase and 2 un i t s /ml T 1 RNAase at 20°C for 15 ra in in 0.3 M NaC1, 10 m M Tris-HC1, p H 7.5 [15 ] . The po ly (A) -con t a in ing RNAase*resis tant ma te r i a l was isolated on po ly (U) -Sepha rose and e lu ted a t 42°C. 1-ml samples were m i x e d wi th Aquaso l II and c o u n t e d . The A 2 6 0 n m of RNAase-res is tan t mater ia l s re leased f r o m the c o l u m n was d e t e r m i n e d be fo re and a f t e r hea t ing to 75°C, f r o m wh ic h h y p e r c h r o m i c i t y was calcula ted. H y p e r c h r o m i c i t y is de f ined as the A 2 6 0 n m a f t e r hea t ing and rapid cool ing divided by the A 2 6 0 n m be fo re heat ing , m in u s 1, (i.e. A 2 6 0 n m D / A 2 6 0 n m O--1) [ 1 6 ] .

Yield Specific ac t iv i ty H y p e r c h r o m i c i t y (gg) (14C)

( d p m / p g )

RNAase- res i s t an t sequences of f rac t ion 1 12 .2 22.0 0 .32 RNAase- res i s tan t sequences of f rac t ion 2 48.6 4.1 0 .05

428

I00-~

90.

8 0 -

7 0 -

6 0 - u

o L. 50- t-I

O o.

~_,j 30- e-

20- Q

10-

X X x

° i - - ' ,b ~o ' 3'o ~o 5'o do , i JJg mRNA i

O.I 0.5 LO

Jug SYNTHETIC POLY (U)

Fig. 4. D e t e c t i o n of p o l y ( U ) s e q u e n c e s in f r a c t i o n 1 R N A by h y b r i d i z a t i o n to [ 3 H ] p o l y ( A ) . Non-

d u p l e x e d [ 3 H ] p o l y ( A ) was d iges t ed wi th R N A a s e T 2 in 0.5 M NaC1 as desc r ibed in Mater ia ls and Me thods . T h e p r o t e c t e d [ 3 H ] p o l y ( A ) • p o l y ( U ) d u p l e x e s were p r e c i p i t a t e d wi th t r i ch lo roace t i c acid. The

e f f e c t o f v a r y i n g q uan t i t i e s o f s y n t h e t i c p o l y ( U ) ( - . ) , f r a c t i o n 1 R N A ( ) - - ~) and f r ac t i on (2 + 3) R N A (× ×) were e x a m i n e d . In the absence of added R N A 99 .95% of [ 3 H ] p o l y ( A ) b e c a m e

t r i ch l o r o acc t i c acid soluble .

sequences of fraction 1 RNA. The remaining sequences moved near the front and were estimated to be 20--50 nucleotides in length (Fig. 5). In contrast digestion of fraction 1 RNA under conditions designed to preserve only the poly(A) tracts resulted in the loss of the ment ioned small RNA sequences from the gel as well as the loss of the uracil content from the isolated poly(A) {Table I).

Binding characteristics of a synthetic poly(A) , poly(U) mixture to a poly(U) column

Fraction I RNA appears to contain a poly(A) • poly(U) duplex structure which weakens its affinity for the poly(U) column. If this is so synthetic poly(A) • poly(U) duplexes may show a similar affinity for poly(U)-Sepharose columns. Binding characteristics of such a duplex were therefore examined by stirring equal amounts of synthetic poly(A) and poly(U) under binding condi- tions for 30 min, followed by their application to a poly(U)-Sepharose 4B column. The elution profile in Fig. 6 shows that most of the poly(A) • poly(U) mixture did no t bind to the column. Bound RNA was released at both 20 and 42°C, whereas poly(A) alone is release from the poly(U) column only at 42°C

429

4° t 3O

5.5S 5S 4S

'7, o x

(3_ (.,..)

8-

6-

4-

2-

,¢ to t~

0 , 2 -

0.1- l 0.0-

NSB 42°C 50°C 1 I I

,~ 2~, 3'o ,~o ~b 8'0 . . . . . . . . . . . . . . IOml fractions GEL SLICES (mm)

Fig. 5. Separat ion o f fract ion 1 RNAase- res i s tan t f ragments on 10% p o l y a c r y l a m i d e gels in 98% folTnamide. R N A res is tant to RNAases A a nd T I in 0.3 M NaCl bu f f e r were isolated as descr ibed in Table II and 3 ' -OH ends were label led wi th NaB3H4 as descr ibed in Materials and Methods for nuc leo - sides. F o l l o w i n g e l ec trophores i s gels were sliced in to 2- ram sect ions and pieces solubl l ized overn igh t in 0.7 m l o f 90% N uc l ear Chicago Sc int i l la t ion ( A m e r s h a m Searle) . Aquaso l II (New England Nuc lear ) and 30 ~1 glacial ace t i c acid were a dde d and radioact iv i ty de t e rmin ed . The pos i t i on o f markers is ind ica ted by arrows: 5 S R N A is cons idered to be approx . 120 nuc leo t ides in length, 4 S RNA is cons ide red to be approx. 8 0 n u c l e o t i d e s in length.

Fig. 6. An elution profile of synthetic polynucleotides released from a poly(U)-Sepharose 4B column.

200 /~g each o f p o l y ( A ) and p o l y ( U ) were m i x e d t o g e t h e r fo~ 3 0 min in 5 m l o f binding buf fer and t h e n appl ied to the p o l y ( U ) c o l u m n . Th e major i ty of the R N A did n o t bind to the c o l u m n . B o u n d R N A was e lu ted wi th no salt bu f f e r (NSB) at 20 and 42°C.

(Fig. 1). A small fraction of the synthetic poly(A) • poly(U) produces com- plexes with the poly(U) column which are eluted under the same conditions as rat liver fraction 1 RNA. Mixing poly(A) and poly(U) in a ratio of 2 : 1 resulted in the retention of 2/3 the applied material and a much larger propor- tion of material eluted in the position of fraction 2 rather than fraction 1. It would appear that more than a partial poly(A) • poly(U) duplex is responsible for the weaker binding of fraction 1.

Cellular location of the various poly(A )-containing RNA fractions Long poly(A) sequences have been found in hnRNA and newly synthesized

m R N A [17] while small poly(U) tracts have been detected in hnRNA [18] and translational control RNA (tcRNA) [9] . The presence of long poly(A) sequences and poly(U) in fraction 1 RNA suggested that it may be a nuclear species.

The proportions of fractions 1 and 2 were examined in polysomes, micro- somes, cytosol, and purified nuclei. The lack of contaminating perinuclear membrane and rough endoplasmic reticulum in the nuclear fraction was con- firmed by electron microscopy (Fig. 7). Both polysomes and microsomes had

430

F i g . 7. A n e l e c t r o n m i c r o g r a p h o f i s o l a t e d r a t l i ve r n u c l e i d e m o n s t r a t i n g t h e a b s e n c e o f p e r i n u c l e a r m e m b r a n e s . M a g n i f i c a t i o n X 11 0 0 0 .

higher proport ions of poly(A)-containing RNA eluting in fraction 2 that the nuclei, which had predominantly fraction 1 (Table III). Similar results were obtained with nuclei whether or not they had been washed with Triton X-100. Cytosol had negligible amounts of poly(A)-containing RNA in either of the fractions. Poly(U) column elution profiles (not shown) indicated that frac- tion 3 is present in RNA isolated from cytoplasmic fractions (i.e. microsomes and polysomes) but absent in nuclear RNA.

Electrophoretic profiles o f fractions 1 and (2 + 3) RNA The size distribution of fractions 1 and (2 + 3) poly(A)-containing RNA

was performed in 4% polyacrylamide gels in 98% formamide (Fig. 8). There was a surprising similarity between fractions 1 and (2 + 3) RNA which was different from the profile of total liver RNA. A spectrum of messengers appears

431

T A B L E I I I

L E V E L S OF F R A C T I O N 1 AND 2 P O L Y ( A ) - C O N T A I N I N G R N A IN V A R I O U S S U B C E L L U L A R L O C A T I O N S

R N A e x t r a c t e d f r o m subcel lu lar f rac t ions of ra t l iver was appl ied to a p o l y ( U ) c o l u m n and po ly (A) -con- ta in ing R N A was f r ac t i ona t ed a nd q u a n t i t a t e d as descr ibed in Materials and Methods .

Expt . R N A appl ied F rac t i on 1 F rac t i on 2 F rac t ion 2 / f r ac t ion 1 No. (rag) (}ig) (pg)

P o l y s o m e s 1 10 .30 15.9 45 .5 2.86 2 16 .13 35 .5 69 .3 1 .95

Mic rosomes 1 17 .47 30.8 98 .3 3.19 2 16 .41 21.2 78.8 3 .72

Cytosol 1 4.03 -- -- --

Nuclei 1 4 .03 130 .5 60.7 0 .46 2 3 .49 88 .0 48 .2 0 .55

Fig. 8. Size d i s t r ibu t ion o f po ly (A) -con t a in ing R N A . Frac t ions 1 and (2 + 3) R N A s were i so la ted and 75 ~ug of each was appl ied to 4% p o l y a c r y l a m i d e gels in 98% f o r m a m i d e . R N A isola ted f r o m pos t - m i t o c h o n d r i a l s u p e r n a t a n t was used to d e t e r m i n e 28-S an d 18-S marke r s . A, F r a c t i o n 1; B, f rac t ion (2 + 3); C, to ta l R N A .

432

as background with a number of distinct bands. Many of the bands are present in both fractions 1 and (2 + 3) suggesting that the fractions have species m common. The very prominant bands may be in RNA species for proteins produced in large amounts by the liver, such as albumin mRNA (18 S) [19]. The size range of the RNA was similar to that reported elsewhere for mRNA [20], 6--28 S, with the highest density near 18 S. Some of these bands however may represent fragments of rRNA which may be trapped on the poly(U) column.

Discussion

Two major fractions of poly(A)-containing RNA were separated from total rat liver RNA using affinity chromatography. Poly(A) sequences isolated from fraction 1 RNA had a length characteristic of poly(A) from hnRNA or newly synthesized mRNA [17]. Fraction (2 + 3) had shorter poly(A) tracts, charac- teristic of polysomal mRNA (Fig. 3). There was very little overlap in the size of poly(A) sequences between the RNA fractions, indicating that the poly(U) column was effectively separating different RNA species.

The weaker affinity of fraction 1 RNA for the poly(U) column did not appear to be due to short poly(A) sequences on the 3'-OH end of the RNA, but was more likely due to greater secondary structure which caused weaker bind- ing to the poly(U) column.

The base composition of RNAase-resistant duplex fragments and the [3H]- poly(A) hybridization studies (Fig. 4) indicate the presence of poly(U) sequences in fraction 1 RNA and their absence in fraction (2 + 3). As the data in Fig. 4 shows a linear relationship between the [3H]poly(A) protected and poly(U) added, it is possible to calculate the poly(U) content of fraction 1. Since fraction 1 poly(A) would compete with [3H]poly(A) for the fraction 1 poly(U), only a minimum estimate of the fraction 1 poly(U) content is possi- ble. Using the data in Fig. 4 fraction 1 RNA is represented by at least 0.28% poly(U).

Additional evidence for the existence of a poly(A)" poly(U) duplex in fraction 1 RNA is provided by the elution characteristic of synthetic poly(A) • poly(U) duplexes. It is possible that preformed poly(A) • poly(U) duplexes are able to form a triplex (poly(A)- poly(U)" poly(U)) with the poly(U) of the column. This structure is more labile than a poly(A) • poly(U) duplex [21] and it could be reasoned that they would be released from the column under milder conditions. However, where poly(A) and poly(U) were mixed in a 1 : I ratio most of the synthetic poly(A) • poly(U) mixture did not bind to the poly(U) column (i.e. the duplex was excluded from the column) indicating that the triplex was not readily formed under the binding conditions used. Increasing the poly(A) • poly(U) ratio resulted in proportionately more material eluting in the position of fraction 2 rather than fraction 1. Some of these mixtures of synthetic poly(A) and poly(U) bound to the poly(U)-Sepharose 4B column and eluted under the same conditions as fraction 1 RNA from rat liver (Fig. 6). Synthetic poly(A) alone, eluted from the column under the same conditions as did fraction 2 RNA from rat liver (Fig. 1). It would appear that more than a po ly(A) , poly(U) duplex, or partial duplex, is responsible for the weaker

433

binding of fraction 1. Since the poly(U) sequences associated with the 200 AMP unit poly(A) sequences of fraction 1 were estimated to be 20--50 UMP units long (Fig. 5) considerable stretches of free poly(A) must be present in this fraction. The free poly(A) should form a duplex with the poly(U) column, but weakened in part, by the internal poly(A) • poly(U) duplex in fraction 1. It might be expected that the latter poly(A) • poly(U) duplex would be as readily broken as a po ly (A) .po ly (U) duplex between fraction 1 poly(A) and the poly(U) column. If this were true one would expect an overlap in the size of the poly(A) sequences in fractions I and (2 + 3). This does not appear to be the case since the column separates two classes of poly(A) ÷ RNA each having a distinct and mutually exclusive size range of poly(A) sequences (Figs. 3c and 3d). Therefore it appears that the fraction 1 poly(A) • poly(U) duplex is eluted intact from the column and something more than poly(A) • poly(U), hybridiza- tion is probably holding the fraction 1 poly(A) and poly(U) sequences together. It is unclear at the moment whether the poly(U) sequence is covalently linked or hydrogen bonded to the mRNA.

The size distribution of fractions 1 and (2 + 3) RNA were both characteristic of cellular mRNA and the banding patterns of polyacrylamide gels were quite similar. Although this suggests that the mRNA species are common to both fractions (Fig. 8) an alternative interpretation is that contaminating rRNA is present in both fractions.

Fraction 1 RNA appears to be present mainly but not exclusively in the nucleus (Table III). It is known that large precursor mRNAs are processed in the nucleus by cleavage [22], addition of 'cap' [23], and poly(A), which may be followed by a lag period before the mRNA appears in the cytoplasm [24]. Messenger coding sequences have been found in hnRNA as well as in mRNA sized molecules in the nucleus [24] indicating that cleavage to mRNA size molecules does not necessarily result in immediate movement of the message to the cytoplasm. It is possible that fraction 1, represents such a population. However attempts to show that fraction 1 is a precursor of fraction 2 have not been successful. Maturation of mRNA has been reported by others [17] and this may still be so in the present case.

Data recently obtained in our laboratory have shown that fraction (2 + 3) but not fraction 1 can stimulate protein synthesis in vitro using a wheat germ system (Dennis, J. and Kisilevsky, R., unpublished). Inhibition of protein syn- thesis by an isolated poly(U)-containing tcRNA has been shown to be depen- dent on the presence of the 3'-OH poly(A) sequence in mRNA [9]. It has been suggested that the poly(U)-containing tcRNA that is found hydrogen bonded to inactive mRNA may result from cleavage of hnRNA and remain attached to both the 3'-poly(A) and a sequence at the 5' end of new mRNA. This tcRNA remains so attached until the mRNA is needed for translation. The poly(U) sequence is then cleaved or removed releasing the poly(A) • poly(U) hybrid and the cyclic structure of the tcRNA • mRNA complex. This model would fit the structural data obtained for fraction 1 RNA. Hybridization of the poly(U)- containing RNA to both the 5'- and 3'-end of fraction 1 poly(A) ÷ RNA would produce a more stable complex than a poly(A) • poly(U) hybrid alone. In a recent report, affinity chromatography with a poly(U) column and a procedure similar to the one reported here was used to isolate an RNA fraction from

434

Artemia salina which inhibited protein synthesis in vitro [25]. The poly(U)- containing RNA may be a tcRNA. Further work is in progress to determine the effect of the poly(U) sequence on protein synthesis.

Acknowledgements

This research was supported by a research grant from the Medical Research Council of Canada MT-4133.

References

1 P a l m i t e r , R . O . ( 1 9 7 4 ) B i o c h e m i s t r y 13, 3 6 0 6 - - 3 6 1 4 2 S t a e h e l i n , T. , B r i n t o n , C., W e t t s t e i n , F .O . a n d Nol!, H. ( 1 9 6 3 ) N a t u r e 199 , 8 6 5 - - 8 7 0 3 B r a w e r m a n , G. ( 1 9 7 3 ) in M e t h o d s in Cell B i o l o g y (P re sco t t , D.M. , ed . ) , Vol . 7, pp . 2 - - 2 1 , A c a d e m i c

Press, N e w Y o r k 4 G r o n e r , B., H y n e s , N. a n d Phil l ips, S. ( 1 9 7 4 ) B i o c h e m i s t r y 13, 5 3 7 8 - - 5 3 8 3 5 Wu, R.W. a n d Writ, F .H . ( 1 9 7 4 ) Dev. Biol . 41 , 3 6 2 - - 3 7 0 6 Moss , B. a n d K o i z o t , F. ( 1 9 7 6 ) J . Virol . 17, 3 8 5 - - 3 9 2 7 R a n d e r a t h , E. , Y u , C.T. a n d R a n d e r a t h , K. ( 1 9 7 2 ) Ana l . B i o c h e m . 48 , 1 7 2 - - 1 7 6 8 L e p p l a , S .H. , B j o r a k e r , B. a n d B o c k , R.M. ( 1 9 6 8 ) in M e t h o d s in E n z y m o l o g y ( G r o s s m a n , L. a n d

Moldave , K. eds.) , Vol . x n B, pp . 2 3 6 - - 2 4 0 , A c a d e m i c Press, N e w Y o r k 9 Bes te r , A . J . , K e n n e d y , D.S. a n d H e y w o o d , S.M. ( 1 9 7 5 ) P roc . Nat l . A c a d . Sci. U.S. 72, 1 5 2 3 - - 1 5 2 7

1 0 L u f t , J .H . ( 1 9 6 1 ) J . B i o p h y s . B i o c h e m . C y t o l . 9, 4 0 9 - - 4 1 4 11 K a t n o v s k y , M.J . ( 1 9 6 1 ) J . B i o p h y s . B i o c h e m . Cy to l . 11, 7 2 9 - - 7 3 9 12 F l eck , A. a n d M u n r o , H .N. ( 1 9 6 2 ) B i o c h i m . B i o p h y s . A c t a 55, 5 7 1 - - 5 8 3 13 S t a e h e l i n , T. , We t t s t e i n , F .O . a n d Nol l , H. ( 1 9 6 3 ) Sc i ence 140 , 1 8 0 - - 1 8 3 14 S c h n e i d e r , E .L . a n d S h o r r , S.S. ( 1 9 7 5 ) Cell 6, 1 7 9 - - 1 8 4 15 J e f f e r y , W.R. a n d B r a w e r m a n , G. ( 1 9 7 5 ) B i o c h e m i s t r y 14, 3 4 4 5 - - 3 4 4 8 16 Mahle r , H .R . a n d Cordes , E .H. ( 1 9 7 1 ) B io log ica l C h e m i s t r y , 2 n d edn . , pp . 1 9 5 - - 2 0 1 , H a r p e r a n d R o w ,

N e w Y o r k 17 She iness , D. , P u c k e t t , L. a n d Darne l l , J .E . ( 1 9 7 5 ) Proc . Nat l . A c a d . Sci. U.S. 72, 1 0 7 7 - - 1 0 8 1 18 D u b r o f f , L.M. ( 1 9 7 7 ) P roc . Nat l . A c a d . Sci . U.S. 74, 2 2 1 7 - - 2 2 2 1 19 T a y l o r , J .M. a n d Tse, T .P .H. ( 1 9 7 6 ) J . Biol . C h e m . 251 , 7 4 6 1 - - 7 4 6 7 2 0 M o o r e , R . E . , Cardel l i , J .A . a n d P i t o t , H.C. ( 1 9 7 7 ) B i o c h e m i s t r y 16, 3 0 3 7 - - 3 0 4 4 21 S p o d h e i m , M. a n d N e w m a n , E. ( 1 9 7 5 ) B i o p h y s . C h e m . 3, 1 0 9 - - 1 2 4 22 B r a w e r m a n , G. ( 1 9 7 4 ) A n n u . Rev . B i o c h e m . 43 , 6 2 1 - - 6 4 2 23 P e r r y , R .P . a n d Ke l ly , D.E. ( 1 9 7 6 ) Cell 8, 4 3 3 - - 4 4 2 24 S p o h r , G., D e t t o r i , G. a n d Manza r i , V. ( 1 9 7 6 ) Cell 8, 5 0 5 - - 5 1 2 25 Slegers , H. , Met t r i e , R . a n d K a n d o , M. ( 1 9 7 7 ) F E B S Le t t . 80, 3 9 0 - - 3 9 4