366

Biochimica et Biophysica Acta, 4 7 5 ( 1 9 7 7 ) 3 6 6 - - 3 8 2 © Elsevier/North-Holland Biomedical Press

BBA 98873

THE HETEROGENEOUS NUCLEAR RNA OF CHICKEN ERYTHROBLASTS

PATRICK

L. W I L L I A M S O N

a,* a n d A L L A N J. T O B I N b,* *

a Department o f Biochemistry and Molecular Biology, Harvard University, Cambridge, Mass. 02138 and b Department o f Biology, Harvard University, Cambridge, Mass. 02138 and Department o f Biology and Molecular Biology Institute, University o f California, Los Angeles, Calif. 90024 (U.S.A.) (Received August 16th, 1976)

Summary Heterogeneous nuclear RNA (hnRNA) from chicken erythroblasts has a modal molecular weight of 1.6 • 106 in 99% dimethylsulfoxide. When erythroblasts are labeled continuously with [14C]uridine, nuclear RNA is labeled as a single kinetic c o m p o n e n t with a half-life of 18 min. After a 10--20 min lag, label appears in cytoplasmic RNA at about 1% of the initial rate of total RNA synthesis. Of the h n R N A sedimenting faster than 28 S ribosomal RNA in both an aqueous sucrose gradient and a subsequent fructose gradient in 99% dimethylsulfoxide, about one-third is polyadenylated, although only about one in 2000 (i.e. about four molecules per cell) contain a globin messenger sequence. The hnRNA of erythroblasts isolated from 5.7- and 11-day chick embryos have the same content of globin messenger sequences as erythroblasts from anemic adults.

Introduction The most rapidly labeled RNAs of vertebrate cells are high molecular weight nuclear RNAs, consisting both of large heterogeneous nuclear RNA (hnRNA) and of discrete nucleolar ribosomal RNA precursors [1--3]. In cultured cells, a large fraction of the h n R N A is degraded soon after synthesis, with a half-life of about 20 min, and never leaves the nucleus [4]. A consistent, but circumstantial, body of evidence supports the hypothesis that h n R N A is, or contains, a * Supported in part by a National S c i e n c e F o u n d a t i o n Predoetoral Fellowship. Present address: L ab orato ry of Molecular Biology, NIAMDD, National Ins t i t ut e s of Health, Bethesda, Md. 20014, U.S.A. ** To w h o m correspondence should be sent. Present address: D e p a r t m e n t of Biology, University of California, Los Angeles, Calif. 90024, U.S.A. Abbreviation: HEPES, N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid.

367 precursor to cytoplasmic messenger RNA [5--7,9]. The processes intervening between the transcription of a messenger RNA sequence from genomic DNA and its appearance in the cytoplasm have until recently resisted direct study, largely because of the difficulties inherent in identifying a particular mRNA sequence throughout its lifespan. In erythroid cells, hemoglobin synthesis accounts for more than 90% of total protein synthesis. This specialization makes available highly purified globin messenger RNAs which can be translated in heterologous cell-free systems and can be used to direct the synthesis of single-stranded complementary DNA by RNA-dependent DNA polymerase [10--12]. Since avian erythrocytes retain their nuclei even in the circulation, the events intervening between transciption and the appearance of messenger RNA in the cytoplasm are amenable to investigation. Several investigators have reported that the hnRNA of erythroblasts from anemic ducks [7,8] and mouse fetal liver [13] contain globin mRNA sequences. Imaizumi et al. [7] suggest, however, that a portion of the detectable globin sequences in the giant RNA are removed by treatment with dimethylsulfoxide [7]. In addition, McKnight and Schimke [14] found no evidence of a high molecular weight precursor of ovalbumin mRNA. We have examined the hnRNA from erythroblasts isolated from anemic adult chickens and from chick embryos. Like the hnRNA of HeLa, L, and ascites cells, it is degraded with an apparent half-life of 20 min, is large and heterogeneous, and contains a proportion of polyadenylic acid sequences. We find that most of the globin messenger sequences associated with this hnRNA under non-denaturing conditions are not covalently attached. Of the hnRNA molecules that sediment faster than 28 S RNA under denaturing conditions, no more than 0.05% contain globin messenger RNA sequences. Preliminary communications of this work have appeared [ 15,16]. Materials and Methods

(a) Preparation of cells and labeling methods. Erythroblasts from anemic white Leghorn roosters were produced, collected, and washed as previously described [31]. Cells were smeared, stained 20 rain with 2% Wrights/6% Giemsa, and typed according to the nomenclature of Lucas and Jamroz [17]. Fertile, pathogen-free white Leghorn eggs (SPAFAS, Norwich, Conn.), stored no more than 6 days at 16°C, were incubated at 38°C in a humidified incubator (Model 55, Humidaire Incubator Co., New Madison, Ohio), and turned every hour through about 60 ° C. Embryos at 10--12 days of incubation were bled by openeing the shell at the air sac, tearing the shell membrane at a point away from the large blood vessels, and reaching into the egg with a forceps to the branch point of the chorioallantoic artery. This vessel was hung over the side of the shell and crushed, and the blood was collected with a Pasteur pipette. Using this method, 30 12-day embryos yield 5--6 ml of blood, containing about 101° cells. Embryos incubated 5.7 days were bled by cutting the egg at the narrow end and allowing the yolk and albumin to pour out. The embryo and membranes left behind were then rinsed with cold isotonic saline, containing 0.14 M NaC1, 0.005 M KC1 chloride, 0.0015 M MgC12. They

368 were then removed into a petri dish containing NaC1/KC1/MgC12 solution m~d minced. After 10--15 min embryos had been bled in this way, the mixture was filtered through cheesecloth to remove debris. Yolk and other contaminating materials were removed by subsequent repeated washings with NaC1/KC1/MgC12 solution. Using this method, 120 embryos yield about 1 ml of packed cells. After washing with NaC1/KC1/MgC12 solution, cells were prepared for radioactive labeling by washing once in 10 volumes of incubation medium, consisting of Dulbecco's modified Eagle's medium with 25 mM HEPES buffer (GIBCO), with 10% dialyzed fetal calf serum (GIBCO No. 630, lot No. C12560). They were then resuspended in 5--10 volumes of incubation medium and preincubated for 45 min at 37°C. The cells were labeled for up to 60 min with either [5-3H]uridine (20--30 Ci/mmol, final concentration 100 pCi/ml) or [14C]uridine (50 Ci/mol, final concentration 1 pCi/ml) in Hank's balanced salt solution. After the incubation, the cells were spun out at 2000 × g for 2 min and incorporation stopped by the addition of 10 volumes of ice-cold NaC1/KC1/MgClz solution. (b) Isolation of nuclei and nuclear RNA. Two different methods have been used to prepared nuclear RNA: Method 1 is a modification of the m e t h o d developed by Penman [18] for HeLa cells. Cells were lysed and the nuclei washed three times with 10 volumes of 1% Triton X-100 (Rohm and Haas) or 1% NP40 (Shell) in NaC1/KCI/MgC12 solution. The nuclei were then washed with five volumes of a solution of 1% Tween 40 and 0.5% deoxycholate in 0.010 M NaC1, 0.0015 M MgC12, 0.010 M Tris. HC1, pH 7.4. The resulting sticky pellet was resuspended in five volumes of a high salt buffer, containing 0.5 M NaC1, 0.050 M MgC12, 0.010 Tris • HC1, pH 7.4, and deoxyribonuclease (Worthington, ribonuclease-free grade) was added to a final concentration of 0.5 mg/ml. This suspension was warmed to 37°C and mixed vigorously until the nuclear pellet was completely dissolved. Two volumes of ethanol at --20 ° C were then added to stop the reaction. The resulting precipitate was then phenol extracted at 55°C, as described by Penman [18]. Method 2 derived from that of Holmes and Bonner [19], in which whole cells or nuclei, prepared by washing cells three times with 0.01 M CaC12, are lysed in a buffer containing 8 M urea, 2% sodium dodecyl sulfate (SDS), 0.002 M EDTA, 0.35 M NaC1 and 0.01 M Tris • HC1, pH 7.4. This lysate was then extracted with phenol and chloroform at room temperature and treated with deoxyribonuclease as described, but deoxyribonuclease digestion was continued until well after the DNA pellet had disappeared. The Sepharose 2B colu m n was omitted, since subsequent centrifugation accomplished the same end. (c) Velocity gradient centrifugation. Non-denaturing aqueous linear sucrose gradients consisted of linear 15--40% (w/v) sucrose (Schwartz-Mann, RNAase free) in 0.1 M NaCl, 0.001 M EDTA, 0.5% SDS, 0.010 M Tris • HC1, pH 7.4 {SDS buffer). Samples, in 0.25--0.50 ml SDS buffer, were layered on 12.5-ml gradients, spun for 3 h at 40 000 rev./min (mean centrifugal force of 200 000 × g) at 23°C in a Beckman type SW40 swinging bucket rotor. Each gradient was collected from the bottom. To achieve separations under denaturing conditions, two types of density gradients in dimethyl sulfoxide (Me2SO) were used. Initial work was done with

369 linear 5--20% (w/v) sucrose gradients in 99% Me2SO containing 0.010 M LiC1, 0.001 M EDTA, pH 7.0. These gradients suffer from the disadvantage that solutions of sucrose in Me2SO become very viscous with increasing sucrose concentrations [20] causing large RNAs to slow down as they sediment, with a loss of resolution and an ambiguity in estimation of molecular weight. To overcome these difficulties, we have employed exponential fructose gradients in Me2SO, which are approximately isokinetic (Baltusis, L., Williamson, P., Crouse, G. and Doty, P., unpublished). These gradients were prepared by the method of Noll [21], using a mixture volume of 10.4 ml, a gradient volume of 12.2 ml, and initial and reservoir concentrations of 3.3 and 41.6% (w/v) fructose in 99% Me2SO, 0.010 M LiC1, 0.001 M EDTA, pH 7.0. Samples for both kinds of gradients were prepared by dissolving the RNA in 25--50 pl of water and adding 99% Me2SO, 0.010 M LiCl, 0.001 M EDTA, pH 7.0, to a minimum final concentration of 80%. Gradients were spun for 22--24 h at 40 000 rev./min in polyallomer tubes in the Beckman SW40 rotor at 25°C. Fractions were collected from the bottom. RNA was recovered after precipitation with 3 volumes of ethanol or 2 volumes of ethanol/2 volumes ether at --20 ° C. (d) Globin messenger R N A and complementary DNA (cDNA). Globin messenger RNA was isolated by SDS-sucrose gradient centrifugation of polysomal RNA from the red cells of anemic roosters. Single-stranded DNA complementary to globin messenger RNA was also prepared as described [12], using the method Verma et al. [11]. The cDNA used in these experiments had a specific activity of 5 • 106 cpm/pg and a number average molecular weight of 106 000, as determined by analysis of sedimentation velocity experiments. (Williamson, P.L., unpublished). (e) Hybridization. Hybridization was done as previously described [12], using the method of Housman et al. [22]. Hybridizations were carried out in solution in hybridization buffer consisting of 0.5% SDS, 0.005 M EDTA, 100 pg/ml oligouridine (Calbiochem), and 0.2 M sodium phosphate, pH 6.8, at 68 ° C. Oligouridine was added to suppress adventitious hybridization of the oligo (dT) region of cDNA and poly (A) regions; it did not interfere with the hybridization of cDNA with messenger RNA sequences. Analysis of globin messenger RNA sequences in fractionated hnRNA was performed by dissolving ethanol-precipitated RNA in 10 #l of a mixture of two parts formamide to one part 4 × hybridization buffer. A mixture of 6 pl of this solution and 2 pl of cDNA was sealed in a 10 pl micropet, hybridized at 37°C, and treated as described [12], except that 2 ml of nuclease buffer and 10 pl of nuclease were used. The increased volumes were required to prevent the inactivation of the nuclease by phosphate or EDTA. Hybridizations were carried out for 36--42 h. Since not all of the hybridization reactions were performed under pseudo firstorder conditions, we determined both the second-order rate constant and saturation amounts of hybrid by double reciprocal plots [12]. (f) Determination o f polyadenylic acid sequences in hnRNA. We determined the fraction of the radioactivity in hnRNA that is attached to poly(A) sequence by the method of Greenberg and Perry [23]. In this procedure, oligouridylic acid was incubated with the RNA under conditions which yield triple helices of (AU2). RNAs containing such triple helices are separated from those which do not by means of hydroxyapatite chromatography.

370

(g) Iodination. Iodination of globin messenger RNA was kindly performed by Dr. Argiris Efstradiatis according to a protocol derived from that of Commerford [24]. A 20 pl reaction mixture containing i pg of 10 S RNA in 3 pl of water, 4 pt of 0.1 M HC1, 4 pl (1.3 mCi) l:sI in 0.1 M NaOH (New England Nuclear NFZ-033H, 17 Ci/mg), 7 pl 0.4 M sodium acetate, pH 4.5, and 2 pl T1C13 (2.67 mg/ml) was sealed in a capillary tube and incubated 15 min at 60 ° C. After incubation, the capillary was broken and its contents expelled into 50 pl 0.4 M sodium acetate, pH 4.5, to which was added 10 pl 2-mercaptoethanol, 10 pl glycerol, and 100 pg yeast tRNA in 10 pl water. 12SI-labeled messenger RNA was separated from small molecular weight components on a 9 × 250 mm Sephadex G-100 column in 0.1 M sodium acetate, pH 4.5, with 100 pg/ml heparin added as a ribonuclease inhibitor. Iodinated RNA was c o u n t e d on a Picker ~/-counter and had an initial specific activity of approx. 1.7 • 107 cpm/pg. Results

Size distribution o f erythroblast hnRNA In chicken erythroblasts, as in other vertebrate cells, the first observable incorporation of labeled uridine is into h n R N A (see below). We have used two methods to prepare h n R N A , m e t h o d 1 being derived from t h a t of Penman [18] and m e t h o d 2 from t h a t of Holmes and Bonnet [19]. After 60 min of labeling, h n R N A isolated by either m e t h o d gives the same size distribution after sedimentation in aqueous sucrose gradients containing 0.5% SDS. Fig. 2 shows the size distribution of h n R N A isolated by m e t h o d 2 from three different erythroid cell populations after a 60-min incubation with [14C]uridine. Erythroblasts from anemic adults and from l l - d a y embryos are primarily midand late-polychromatophilic erythroblasts, with 1--2% basophilic erythroblasts also present in the adult cell population [17]. In both these populations, virtually all the label incorporated into nuclear RNA sediments faster than 28 S ribosomal RNA in aqueous sucrose gradients (Fig. 1A and 1B). The labeled RNA is obviously heterogeneous, with no apparent incorporation into ribosomal RNA precursors. In contrast, nuclear RNA isolated from the still less mature erythroblasts of the 5.7-day embryo contains distinct labeled components sedimenting at 45 S and 32 S (Fig. 1C). After exposure to dimethyl sulfoxide, h n R N A prepared by m e t h o d 1 appeared to be much reduced in size, sedimenting markedly more slowly than a 23 S RNA marker. HnRNA prepared by m e t h o d 2, however, was apparently less degraded in the course of preparation and therefore became the focus of closer examination. HnRNA isolated by m e t h o d 2 and sedimenting faster than 28 S RNA in aqueous sucrose gradients was pooled and refractionated in 99% Me2SO; the sedimentation coefficients of this material ranged from 10 S to 50 S. The distribution of RNAs from adult, l l - d a y and 5.7-day erythroblasts are shown in Fig. 2. The RNA populations are clearly heterogeneous, with an apparent modal molecular weight of 1.6 • 106, but with some molecules as large as 7.5 • 106 daltons. About half o f the hnRNA, however, appears to have a molecular weight of less than 1.6 • 106, although all the starting RNA initially sedimented faster than 28 S ribosomal RNA {1.65 • 106) in aqueous sucrose gradi-

371

2O

45S

300

2&£ 16S

oJ

I

_o

~ 20C

x

I0 o o

=

[

IOI

1

?-~0

____

L

- [ 5

_

_ 10

2 0 0 --

15 ~0 45S 23S J,

J,

25 4S t

28S 23S 18 16S

o

5 45

4~s 2~s ?s

I0

15

20

[

'~5S 23S ~

25

4S

'O

30 o

~:= ]5 .... I0 FRACTION

20

1. . . . . . . . 1I0 20

30 FRACTION

40

50

Fig. 1. C o m p a r i s o n o f h n R N A p r o f i l e s f r o m d i f f e r e n t cell p o p u l a t i o n s . R N A w a s isolated b y m e t h o d 2 a n d f r a c t i o n a t e d o n a q u e o u s s u c r o s e g r a d i e n t s as d e s c r i b e d a b o v e . ( A ) R N A p u r i f i e d f r o m e r y t h r o b l a s t s f r o m a r o o s t e r m a d e a n e m i c b y a c e t y l p h e n y l h y d r a z i n e i n j e c t i o n . (B) R N A p u r i f i e d f r o m e r y t h r o b l a s t s in t h e c i r c u l a t i o n o f t h e l l - d a y c h i c k e m b r y o . (C) R N A p u r i f i e d f r o m e r y t h r o b l a s t s in t h e c i r c u l a t i o n of t h e 5.7-day c h i c k e m b r y o . Fig. 2. S e d i m e n t a t i o n o f h n R N A in M e 2 S O . ( A ) h n R N A f r o m a n e m i c a d u l t e r y t h r o b l a s t s s e d i m e n t i n g faster t h a n 28 S r i b o s o m a l R N A was p r e c i p i t a t e d w i t h e t h a n o l , c o l l e c t e d b y c e n t r i f u g a t i o n , a n d dissolved in 50 pl o f distilled w a t e r . T h e s a m p l e w a s t h e n d i l u t e d with 2 0 0 #1 o f 99% M e 2 S O , 0.01 M LiC1, 0 . 0 0 1 M E D T A , p H 7, a n d l a y e r e d o n t o M e 2 S O - f r u c t o s e g r a d i e n t s p r e p a r e d , r u n , a n d a n a l y z e d as d e s c r i b e d . (B) h n R N A f r o m l l - d a y e m b r y o s . (C) h n R N A f r o m 5.7-day e m b r y o s .

ents. Fractionation in Me2SO of rapidly labeled RNA from whole cells gave the same size distribution as rapidly labeled RNA from isolated nuclei (data n o t shown). This result indicates that the reduction in the size distribution of hnRNA when analyzed under denaturing conditions does not result from nicks introduced during the prepurification of nuclei or during the first sedimentation under aqueous conditions. We next wished to determine whether the measured size of the h n R N A was affected by the duration of labeling, as would be the case if large molecules were rapidly processed to smaller and more stable molecules. Cells were labeled for 60 min with [14C]uridine, then pulsed for 5 min with [3H]uridine, and the RNA was extracted by method 2. The size of the labeled RNA was then determined by sedimentation in aqueous and in Me2SO gradients as shown in Fig. 3. In both experiments, the 3H-labeled h n R N A shows an increase in the a m o u n t

372 3

~3

A

23S

414

3H x.x'x~X'x'X'x'x x

A

rx

A

42

I~ '2 0

~E (3.

"il '

I0

L

2'0

B

'

3t0

l,o

q

0%6