A high-resolution oligonucleotide map generated by restriction of poliovirus type I genomic RNA by ribonuclease III

A high-resolution oligonucleotide map generated by restriction of poliovirus type I genomic RNA by ribonuclease III

VIROLOGY 104, 3?5-3% (1980) A High-Resolution Poliovirus Oligonucleotide Type I Genomic Map Generated by Restriction RNA by Ribonuclease III MAR...

10MB Sizes 0 Downloads 2 Views

VIROLOGY

104, 3?5-3%

(1980)

A High-Resolution Poliovirus

Oligonucleotide Type I Genomic

Map Generated by Restriction RNA by Ribonuclease III

MARGARET L. STEWART,’ ROBERT J. CROUCH, AND J. V. MAIZEL, Laboratoyy

of

JR.

Genetics, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland 20205

of Molecular

Accepted March 24, 1980

Processing of poliovirus Type I (Mahoney) genomic RNA with Escherichia coli ribonuclease III produces a large number of discrete fragments. Analysis of these fragments (and isolated oligonucleotides) using a high-resolution sodium dodecyl sulfateagaroie-acrylamide preparative gel and a two-dimensional fingerprinting gel system has allowed us to map ‘76 T, ribonuclease-resistant oligonucleotides within the poliovirus genome at 200- to 300-base intervals from the poly(A) end. Strikingly, the most preferred cleavage sites on the poliovirus genomic RNA are located close to each terminus, approximately 75 bases from the 5’ end, and 616 bases from the 3’ end. Furthermore, these and other internal sites appear to punctuate the translation products of the poliovirus mRNA intercistronically. Thus cleavage of poliovirus genomic RNA by ribonuclease III greatly resembles processing of bacteriophage T, early mRNA by this enzyme.

Leis et al., 1978) which may regulate gene expression. Two-dimensional gel analyses of T,We have been additionally concerned resistant oligonucleotides have been most with the possible effects that secondary useful in diagnostically characterizing the structure may have upon expression of genomes of many RNA viruses (Joho et al., poliovirus RNA whether defined by the 1975; Frisby et al., 19’76; Kennedy, 1976; primary structure of the RNA alone or Lee and Wimmer, 1976; Inglis et al., 1979; mediated by the specific affinity that certain Davis and Nyak, 1979). Most unique oligoregulatory proteins may have for it. nucleotides in fingerprints of messenger The electron microscope has been most RNA can be accurately mapped within the useful in defining large regions of poliogenome by fragmentation of the RNA under virus RNA molecules locally compleweak alkaline conditions, followed by end mentary, and even distally related to each selection of 3’ poly(A)-containing fragments other (Lundquist, Peterlin, Meyer, Sullivan, by affinity chromatography on appropriate columns (Coffin and Billeter, 1976; Wang and Maize& unpublished data). Resolution, however, is presently limited to sequences et al., 1975; Nomoto et al., 1979). In addithat are 100 bases or more in length, tion to defining segments of genomic, and and this of course excludes local secondary messenger RNAs that have been deleted, structures which could, nevertheless, be or otherwise rearranged, such maps are biologically relevant. expected to be useful in identifying segRibonuclease III is an endoribonuclease ments of RNA captured in biologically from Escherichia coli which recognizes and active complexes with ribosomes (Atkins processes specific structures in RNAs from et al., 1979; Pinck et al., 1979; McClain varied sources (Dunn and Studier, 1975a, et al., 1980a), polymerase molecules (Mills b; Rosenberg and Kramer, 1977; Robertet al., 1977), and other proteins (Golden- son, 1977)including eucaryotic viruses (Leis berg et al., 1979; Sonenberg et al., 1979; et al., 1978; Westphal and Crouch, 1977; Wertz and Davis, 1979). Apparently similar ’ To whom reprint requests should be addressed. INTRODUCTION

375

0042-6822/80/1003X-23$02.00/O Copyright All rights

0 1980 by Academic Press, Inc. of reproduction in any form reserved.

376

STEWART,

CROUCH,

structures are present within the poliovirus genome at specific sites, since ribonuclease III specifically processes this RNA, too (Nomoto et ccl., 1979; Harris et al., 19’78;and this paper). Such specificity is proving most useful in providing unique, subgenomic classes of poliovirus RNA for further functional as well as structural analyses described above. This study describes how specific fragmentation of polio RNA with ribonuclease III and analysis of the fragments by polyacrylamide gel electrophoresis has allowed us to map 76 T,-ribonuclease-resistant oligonucleotides within the poliovirus genome. The location of discrete ribonuclease III sites within the genome suggests that specific secondary structures in poliovirus genomic RNA may play a relevant role in genomic expression of the virus.

AND MAIZEL

liters TSE at 4”, changed once, for 3 hr. RNA was purified by extraction in phenolm-cresol-chloroform as described by Kacian and Myers (1977) in the presence of 0.1% P-mercaptoethanol (MSH), and stored in small aliquots of l/10 TSE in liquid nitrogen. Specific activity was invariably 3-6 x loj cpm/pg RNA. Marker single-stranded RNA was prepared by phenol extraction (Kacian and Myers, 1977) of poliovirusinfected cell extracts prepared with NP40 and deoxycholate (Penman, 1966). Digestion of polio&ion RNA by ribonuclease III and isolation of fragments.

Poliovirion RNA, 125 pg (26 x 10” cpm), was preincubated at 37” in 1.5 ml of 0.02 M Tris-HCI, pH 7.9, containing 45 mlM NaCI, 10 mM MgC&, and 10% glycerol for 5 min. Two micrograms of ribonuclease III (Crouch, 1974) was added, and the reaction mixture was incubated for 15 min MATERIALS AND METHODS at 37”. SDS (sodium dodecyl sulfate) Virus and cells. Stock poliovirus, Type and EDTA were added to give a final I (Mahoney), was grown in HeLa S3 cells concentration, respectively, of 1% and as previously described (Phillips et al., 0.002 M. The solution was then heated 1968). Defective interfering virus particles to 55” for 3 min to stop the reaction and were grown and purified essentially also as dissociate the ezyme from the RNA. The salt concentration was adjusted to 0.4 &’ described previously (Lundquist et al., in a final volume of 8.0 ml and the RNA 1979). was phenol extracted as described above Radiolabeling of viral RNA. Four liters of cells (4 x lo5 cells/ml) were washed once except that fragments were precipitated in ice-cold Earle’s salt solution (Gibco), and by addition of 2 vol of ethanol and centrifresuspended in warm Eagle’s minimum ugation was overnight in the Spinco SW medium (400 ml), without sodium phos- 40 rotor at -5”. Poly(U)-Sepharose selection of RNase phate (Gibco), and buffered with 10 n&f poliovirus RNA fragments. each HEPES, BES, and TES, pH 7.2 III-restricted (Eagle, 1971). Thirty minutes after in- Poly(A+) RNA fragments were separated fection with 60 PFU/cell dialyzed fetal from poly(A-) RNA fragments on poly(U j calf serum (Gibco) was added to give a Sepharose columns essentially as described final concentration of 5%. Ninety minutes by Ihle et al., (1974). postinfection actinomycin D (5 pg/ml) and Ethanol-precipitated fragments were re25-30 mCi [32P]orthophosphoric acid (ICN) suspended in 50 ~1 of freshly made RNA were added and at 4.5 hr postinfection. gel sample buffer (6 J4 urea, 0.007% phenol Wild-type virus was harvested and purified red, 20% glycerol, 1.0% SDS, and 0.1% as described (Phillips et al., 1968) except MSH in 0.002 M EDTA and 0.01 J4 that virus was banded in CsCl twice Tris-HCl, pH 8.3), and heated for 2 min overnight in the SW 40 rotor of the Spinco at 55” before application to a preparative ultracentrifuge to eliminate all traces of SDS-agarose -acrylamide gel. radiolabeled free RNA. SDS -agarose -acrylamide slab gels. PreIsolation of Viral RNA. The virus band parative slab gels, 30 cm long x 15 cm wide in CsCl was diluted to 3.0 ml in TSE X 1.5 mm thick, containing 0.32% agarose, (0.01 M Tris-HCI, pH 8.3, 0.1 M NaCI, 2% acrylamide in 1% SDS in TSE were 0.002 M EDTA) and dialyzed against 2 prepared as described (Stewart and Crouch,

OLIGONUCLEOTIDE

MAP OF POLIOVIRUS

1980a). Electrophoresis was at 70 V for 20 hr. Extraction of fragments from SDS gels. Radiolabeled bands detected by radioautography for varying times up to 25 min were placed in 1.5 ml TSE pH 7.4 containing 0.5% SDS and 0.1% MSH and 20 ~g carrier E. coli tRNA as required. Slices were counted by Cerenkov Radiation in a Beckman 350 spectrometer, and then frozen at -20”. For extraction, slices were thawed at room temperature, and homogenized in a tissue grinder with a Teflon pestle. After chilling at 4” for 30 min, to precipitate SDS, the homogenates were centrifuged in the Spinco SW 60.1 (or SW 50) rotor at 30,000 rpm for 15 min at 4” to remove SDS and most of the gel material. The solutions were made 0.4 M in NaCl and 2 vol of ethanol (at -20”) were then added, and precipitated RNA was isolated by centrifugation at 50,000 rpm in the Spinco SW 50.1 or 60.1 rotor at -5” overnight. Occasionally, to optimize yields, the gel homogenates containing the largest fragments (5.6--7.7 kb) were reextracted and supernatants combined before ethanol precipitation. Recovery from gel slices is typically 85% of the largest RNAs and close to 100% for smaller molecules. Ribawlease T, digestion of RNA isolated jkorrz gel slices. RNA to be fingerprinted was digested with 2 units of T1 ribonuclease (Calbiochem) per 20 pg RNA (including carrier where necessary; Barrell, 1971) in a volume of 5 ~1 Tris-EDTA and this was diluted 1:l with RNA sample buffer (Kennedy, 1976) for application to gels. Digestion of fragments purified in gels was in 10 ~1 of a solution of 0.02 M Tris-HCl, pH 7.4, 0.004 M EDTA containing 100 units of ribonuelease T,, and 50% DMSO. Ten microliters of fingerprinting sample buffer (Kennedy, 1976), containing 6 M urea and 50% sucrose in 0.02 M Tris-HCI, pH 7.4, was then added for electrophoresis. T~uo-d~n~,~q~s~onal ~~~,ge~~~~tanaEysis of T, ribonmeleuse-resista~~t oligonucleotides. T,-resistant oligonucleotides were separated by eleetrophoresis in two-dimensional

TYPE

I GENOMIC

RNA

377

fingerprinting gels, 13 em wide x 10 cm long x 1.5 mm thick, and the oligonucleotides were transferred under vacuum in buffer (0.02 M Tris-Cl, pH 7.4, 0.002 M EDTA) for 16hr to DEAE paper ~atman~ as described (Stewart and Crouch, 1980b). Under these conditions, little poly (A), 90% of the largest oligonucleotides, numbered 1 and 2 (see Fig. 4B), and all of the smaller oligonucleotides are transferred to the DEAE paper. Normally, 8.0 cpm “-‘P per nu~leotide is sufficient for detection of a spot (on the dried DEAE paper) after exposure for 16 hr to Kodak X-Omat film using a DuPont Lightening Plus Intensifying screen at -70”. Secondary analysis of o~igon~e~eotides: Desig?za‘t~o~of spots as ~~i~~e OTco~~2~le~. oligonucleotides transferred to DEAE paper were eluted with triethylamine carbonate (Barrel& 1971), evaporated to dryness three times, digested with pancreatic ribonuclease, and subjected to highvoltage electrophoresis on polyethyleneimine as described (Stewart and Crouch, 1980b). To compare the identity of spots in ribonuclease III-generated fragments used for mapping oligonueleotides, patterns of pancreatic digests of spots isolated from the ribonuelease III fragments were compared to the same spots derived from the whole RNA fingerprint. All spots numbered 22 or lower had patterns identical to those taken from whole RNA. By this criterion, they were considered to be unique and were given a simple number in the map. If a spot from the fragment appeared to be less complex than the same spot in the whole RNA fingerprint, a subscript, a, was added to the spot number defining the most 3’ component, and b, etc., defining the more 5’ component constituting the pattern for that spot. Not all spots were analyzed in this way. Thus unless otherwise stated, for each spot numbered 23 or greater, the numerical designation without a subscript refers only to a position in the whole RNA fingerprint. Digestion of RNA lath ~~ote~~ase K. Stock proteinase K (Boehringer) was dissolved in l/10 TSE, pH 7.4, at a concentration of 10 mg/ml and incubated at 3’7”for 3 hr and stored at 4”.

STEWART, CROUCH, AND MAIZEL

378 A+

A-

5E I-

*-

34-

FIG. 1. Autoradio~am of poly(A-) and poly(A-) fragments generated by digestion of poliovirus genomic RNA by ribonuclease III. Poliovirus RNA, 122 pg (26 x 10” cpm), was digested with RNase III. Poly(A-) and poly(A--) fra~ment.s were prepared by affinity chromatography on poly(U)-Sepharose, separated in a preparative gel, and auto~dio~aphed as described under Materials and Methods. Poly(A+) fragments applied to gel; 6 x IO” cpm; poly(A-) fragments: 18.6 x 10” cpm. Arrows mark position of uv-absorbing marker RNAs (7 pg each), of singlestranded RNA from poliovirus-infected cells (Bishop and Koch, 1969), plus E. coti 4 S RNA.

Fragments in supernatants (0.75 ml) of gel homogenates containing carrier tRNA (where necessary) were incubated with 200 pg proteinase K for 3 hr at 37” to digest VPg (Nomoto et al., 1977; Flanegan et al., 19’77), phenol extracted, ethanol precipitated, digested with T, ribonuclease, and dissolved in fingerprinting sample buffer for analysis in gels as described above. RESULTS

~~ectro~~oretic ~e~arutio?~ t$’ RNase IIIReskicted Poliovirus Genomic RNA

Phosphorus-32-labeled poliovirus genomic RNA was digested with ribonuclease

III and poly(A>containing fragments were isolated from the digest by affinity chromatography on poly(U)-sepharose as described under Materials and Methods. Fragments binding to the column, (poly(A+) fragments) and those not binding (poly(A-) fragments) were separated by size in a preparative, high-salt SDS-agarose-acrylamide gel (Fig. 1). The pattern of poly(A+) is rather clearly defined in the gel with at least five major bands marked Al, Bl, Cl, Dl, and El, easily visible. In other experiments, not shown here, short exposures (1-2 min) demonstrated that fragments marked Al, A2, and A3 indeed represented three separate bands resolvable in these gels. That E2 corresponds to a real band is shown in a longer exposure (panel left) of t.he A+ lane. Bands at E3 and E4 are more easily visible in other experiments using twofold more ribonuclease III (data not shown). As could be expected, the A- lane contains a more heterodisperse array of radioactivity with very few large bands visible. It is notable that there is almost no radioactivity corresponding to full-length RNA molecules at the top of the A- lane, though five smaller bands 1,2,3,4, and 5 are visible amid the background at shorter exposures (right panel). At the bottom of the A- lane, two bands, marked VPg 75 and VPg 115 are visible. Sixes and Yields of PoEy( A+) and Polyf A-) Fragments Produced by RNase III Digestion

~Iobility of RNA in these gels is directly related to the log molecular weight of the moleeules (Stewart and Crouch, 1980a). The size of the RNA at each position in the gel can be determined from a standard line defined by the position of the appropriate markers (Fig. 2). Slices (Lo-l.5 mm) were cut from the gel at the positions marked in Fig. 1 and the radioactivity in each was determined by Cerenkov radiation. The slices were then frozen at -20” for further analysis in fingerprinting gels. Table 1 shows a summary of the sizes of poly(A+) and poly(A-) fragments in the gel. From these sizes, the

OLIGONUCLEOTIDE

I 20

1 40

I 60

MAP OF POLIOVIRUS

I 60

1 100 DISTANCE

I 120

TYPE I GENOMIC RNA

I 140

I l60

I 160

I I 200 220

379

I

FROM ORIGIN (mm)

FIG. 2. Size deter~~tion: poly(A+) and poly(A-) fragments. Arrows mark positions of uvabsorbing marker RNAs. (A) A+ fragments; (A) A- fragments, numbered as in Fig. 1.

position on the intact genome of the 5’ end of each poly(A+) fragment is listed and the amount of radioactivity in each slice is also shown. The total amount of radioactivity recovered in slices from the poly(A+) lane (5.8 x lo6 epm) accounts for the 76% of the radioactivity applied to the sample well of the preparative gel. Ribonuclease III has been shown to specifically cleave a loo-base fragment containing VPg from the 5’ end of poliovirus Type I genomic RNA (Harris et al., 1978). An octanucleotide, protected from ribon&ease digestion by VPg and detectable by its anomalous electrophoretic migration due to its net positive charge (Nomoto et al., 1977; Flanegan et al., 19’77), can also be detected in first-dimension tiger-printing gels in which the polarity of the electrodes has been reversed (Stewart and Crouch, unpublished data). Such analysis of fragments identical to those shown here demonstrated that VPg can be found only in fragments

VPg 75 and VPg 115indicating that all of the fragments in the gel in Fig. 1 have been cut at their 5’ termini by ribonuclease III (data not shown). From the amount of radioactivity present in the starting digest (26 x lo6 cpm), a maximum of 3.0 x lo5 cpm could be expected to be in a fragment 115 bases in length, and 2.5 x lo5 cpm in a fragment 75 bases long. After digestion, phenol extraction, affinity chromato~aphy, and gel electrophoresis, the amount of radioactivity in VPg 115 and VPg 75, respectively, is 17.2 and 28.6% (a total of 45.8% of the original amount). Thus sites 75 and 115 bases away from the 5’ VPg appear to be highly preferred by RNase III. Since the molar yield of VPg 75 is by far the largest, the yields of other fragments, relative to 100 mol of VPg 75, were calculated. Thus, VPg 115 is present in the next largest yield (38.98 mol/lOO mol VPg 75). Next is the 3’ poly(A+) fragment, El

380

STEWART,

CROUCH, TABLE

AND MAIZEL 1

RELATIVE AMOUNTS OF FRAGMENTS GENERATED BY RNase III DIGESTION OF POLIOVIRUS TYPE I GENOMIC RNA Size Otb)

Map position of 5’ end

cpm x lo-$

2 3 4 5 6 7

7.60 7.40 7.00 6.60 6.05 5.80 5.50

0.01 0.04 0.09 0.14 0.21 0.25 0.29

497.6 666.4 547.5 331.5 157.5 339.8 246.0

5.65 7.85 6.78 4.35 2.26 5.14 3.90

Bl 2 3 4 5 6 7

5.00 4.60 4.20 4.00 3.60 3.30 3.20

0.35 0.40 0.45 0.48 0.53 0.57 0.58

603.4 277.5 253.5 184.5 196.1 153.4 210.0

10.45 5.20 5.20 4.01 4.69 4.01 5.65

Cl 2 3

2.80 2.60 2.40

0.64 0.66 0.69

252.8 121.5 93.4

7.90 3.95 3.44

Dl 2 3 4 5

2.05 1.90 1.50 1.30 1.08

0.72 0.75 0.80 0.83 0.86

180.0 84.4 105.0 52.5 69.4

7.51 3.72 6.21 3.50 5.49

El 2 3 4

0.62 0.46 0.31 0.15

0.92 0.94 0.96 0.98

144.3 34.0 10.9 5.0

20.46 6.30 4.88 3.57

Fragment Al

Moles fragment/ (moles VPg 75 X 10’)

5800.0

Total 7.00 5.70 4.80 3.80 2.60

poIy(A-1 2 3 4 5 VPg 115 VPg 75

-

0.115 0.075

270.0 450.0 559.1 1041.4 542.1 51.62 85.90

8.16 6.89 9.55 23.33 16.78 38.98 100.00

Note. The size of each fraction was determined from the plot in Fig. 2. The location of each 5’ end relative to the poly(A) tract was positioned from the size. Radioactivity in each fragment was determined by Cerenkov radiation, and corrected for quench and decay.

(20.46 moU100 mol VPg ‘75), showing that another highly preferential site is located approximately 616 bases from the 3’ end of the RNA. If ribonuclease III were recognizing each cleavage site equally, the shortest poly(A+) fragments would be present in the greatest

molar yields. Yet, a different order of abundance for the poly(A+) fragments is established from Table 1 as Bl > (Cl > Dl > A2) > A3 > Al. Furthermore, secondary cuts are found at the 5’ ends of fragments A6, B2, B3, B5, B7 and D3, D5, and E2, respectively. So that oligonucleotides

OLIGONUCLEOTIDE

MAP OF POLIOVIRUS

could be mapped at intermediate positions, some slices were cut from the gel at positions which at this enzyme concentration did not contain detectable bands. These correspond to fragments A4, A5, A7, B4, B5, B6, C2, C3, D2, D4, E3, and E4. T,-Resistant Oligonucleotides ii Poly( A+) RNase III-Generated Fragments

Figure 3 shows the fingerprints of the poly(A+) fragments isolated from gel slices and digested with T, ribonuclease as described under Materials and Methods. The fragments, serially increasing in size, are described beginning from the shortest at the 3’ end. Fragment E. The fingerprint of the simplest A+ fragment (154 bases including the tract of poly(A) (60-90 bases, Yogo and Wimmer, 1972)), contains five dominant spots starting with the largest: 16,40a, 66a, 56a, 68, 48, and 36. The complexity of the pattern of secondary digests of spots 40,56, and 66 in whole RNA suggested they were composed of mixtures of corn&rating spots (Stewart and Crouch, 1980b) and compositional analyses on PEI of spots 40a, 56a, and 66a from Fragment El showed they indeed are components of the compositions obtained for spots 40, 56, and 66 in whole RNA (data not shown). To indicate this, as described under Materials and Methods, the subscript a is assigned to the spots in these fragments. The next fragment (E3), only 50 bases longer than E2, clearly contains a new oligonucleotide, 60, of approximately 18-19 bases in length. Secondary analysis of this spot was identical to the analysis of spot 60 from the whole RNA, therefore, it is presumed to be a truly unique spot, characteristic of the 50-base region just 5’ adjacent to the sequence in E4. A new spot (arrow in fragment E3) which also appears is not numbered. Fragment E2 contains four more oligonucleotides: spots 9, 45a, 46, and 58. Completion of the 3’ terminal fragment, El, positions spots 10,23,32, and 53 at the 5’ end of the E fragment. Thus some of the largest (U rich) oligomers in the fingerprint, 9, 10, and 1.6,are diagnostic of the presence of the fragment El or the 616 bases derived from the 3’ end (including the poly(A)) of the poliovirus genome.

TYPE

I GENOMIC

RNA

381

Fragment D. Two spots, 4 and 41 (located between 40a and 50), appear in fragment D5, and fragment D4 uniquely contains spot number 50 within the 5’-adjacent 200 bases. (The 3’ triplet, marking fragment El, occasionally behaves somewhat erratically with spot No. 16 often appearing streaked in the gel, see D3, arrow.) The spot 65’ in D2 refers to an unmarked spot moving just ahead of 65 in the whole RNA fingerprint. Oligonucleotide 18 is located in the 5’, 200-base sequence of fragment D3. This fragment also contains spot 38, which by secondary analysis is a subcomponent of spots 37-38 isolated from whole RNA fingerprints. Spot 39 in this fragment fills in the 38-39-40-41-53 configuration in whole RNA and it is interesting that all these spots (A rich by secondary analysis) with multiple runs of 4, 5, and 6 bases (Stewart and Couch, 1980b) are clustered close to each other around positions 0.8 and between 0.92 and 0.94. Fragment C. In fragment C3, a large unique oligonucleotide 5 is located, and 7b appears within the next 200 bases. (Spots 7a and 7b are separate spots in the whole RNA fingerprint. 7a appears more 5’ in fragment B4, see below.) Filling in between spots 50 and 60 is spot 59 in fragment Cl and spot 35 together with the large oligonucleotide 11. Fragment B. Additional spots 42 and 33 are located in B7. No easily discernible spots appear in B6 but the first spot of the triplet 72-73-74 appears in fragment B5. The fingerprint of fragment B3, though generally poorly resolved, is nevertheless included here since it clearly shows the acquisition of the large unique spots 8, 15, and 3 and spot 69. In fragment B4 spot 45b, the partner of 45a in fragment E2 appears. Spots 51 and 75 appear in fragment B2 and spot 19 appears in Bl completing the fingerprints for the B fragment. Fragment A. Fragment A contains two unique spots 13 and 14, spot 70, and also spot 66b. The intensity of spot 68 increases in fragment A4, suggesting that this spot, also, is a complex of two or more spots. Unfortunately, its small size, U-rich character, and scarcity of adenosine make this difficult to ascertain by secondary analysis. Though

382

STEWART,CROUCH,ANDMAIZEL

a t; 0

-9: 0

0 -6

- N 0 I +T 0 I i_

0.4

0.5

0.6

Cl2

0.7

nl POLY A 3’

6

FIG. 3, Two-tlimensional Cngerprints of poly(A’) fragments. Digests containing 8.0 cpmlnucleotideifragment were analyzed by two-dimensional fingerprinting gels, transferred to DEAF, paper, and exposed overnight as described under Materials and Methods. Arrows with numbers mark oiigonucleotides present in the 5’ entl of the poly(A’ ) fragment, but not in the next smaller (3’) fragment fingerprinted. For convention adopted for designation of spots, see Materials and LMethotls.The position of the bromphenol blue marker used in these gels coincides with spot 56, a,b and is occasionally marked by (:. ) in some gels. Region rtf the genome represented by each fingerprint is shown at the bottom of each panel. (A) Site producing a major fragment; (Q) site producing a minor fragment; (.!S)5’ position of fragment cut from gel in absence of detectable band for use in mapping intermediate regions.

5’

67

C,, 2600 BASES

C,, 2800 BASES

B,, 3200 BASES

Be, 3300 BASES

VPcl

B5, 3600 BASES

B,, 4000 BASES

B3, 4200 BASES

B,, 4600 BASES

5’

D,, 1100 BASES

Da, 1300 BASES

VI%

D,, 2100 BASES

C,, 2400 BASES

r\lrA 0.7

C3D12

E,, 780 BASES

26

D,, 1900 BASES

3

4

4 //ii 0.8

5

El 2 ,bi 0.9

-4

38’

q POLYA

E,, 530 BASES

,+8

D,, 1500 BASES

x

4

“d bl

t

OLIGONUCLEOTIDE

P

MAP OF POLIOVIRUS

TYPE

I GENOMIC

RNA

385

386

STEWART,

CROUCH,

AND MAIZEL

2 Wild Type

1

POLYA

DI VQ passage 19

4

FIG. 4. Two-dimensional fingerprints of poliovirus Mahoney Type I genomic RNA and its deletion mutant(s), DI(A), passage 19. Ten micrograms of radiolabeled RNA (1 x lo5 cpm) with 10 pg carrier tRNA were digested with ribonuclease T, as described under Materials and Methods and the digest was subjected to two-dimensional electrophoresis. Whole gels were frozen in a sandwich made of thin mylar sheets and exposed to film at -70” as described under Materials and Methods. (B and D) Diagrammatic representations of A and C, respectively. (Spot 17 which is apparently susceptible to digestion under conditions used in Fig. 3 was not mapped.)

difficult to see, in fragment A3, spot number 37 appears. The fingerprint of fragment A5 most closely resembles the fingerprint of DI RNA (Fig. 4C below). Spots 22, 71, 6, 20, 68, and 55 are located in fragment A4. Spots

63b, 37, 34, and 65 are located in A3. Spots 57 and 74, which are present in the DI fingerprint, are located very near the 5’ end of the RNA, in fragment A2. There are no easily discernible differences between fragments Al and A2.

OLIGONUCLEOTIDE

MAP OF POLIOVIRUS

Ribonuclease T,-Resistant Oligonucleotide Fingerprints of Poliovirus Type I, and the DI(A), Passage 19 Deletion Mutants Derived from It A new isolate of defective interfering particles, DI(A), passage 19, has been recently characterized (Lundquist et al., 1979). For comparison, fingerprints of the parental wild-type RNA and the RNA from extensively purified DI particles are shown in Fig. 4. Diagrammatic representations are shown for each in the right panels. As many as 50 oligonucleotides in wildtype RNA shown in Fig. 4A have been separately characterized by digestion with pancreatic ribonuclease and secondary analysis on PEI (Stewart and Crouch, 1980b). In some cases, resolution of two closely associated spots in these hngerprints was obtained because they were distally located in separate RNase III-generated fragments shown in Fig. 3: for example, two spots, marked 7a, and 7b, Fig. 4A, though not well resolved in that gel are clearly separable in fragments C2 and B4 (Fig. 3). Similarly resolvable are spots 72, 73, 74 in fragments B4, A6, and A2, respectively. Electron microscopic analyses of heteroduplexes formed between RNA extracted from purified DI(A) particles and wildtype RF (Lundquist et al., 1979) showed that the DI(A) RNAs are derived from mixtures of particles belonging to roughly three populations defined by the position of the deletion(s) on the genome. These deletions (amounting to a maximum of 15% of the RNA) are located roughly in two regions between map units 0.08 and 0.20; and 0.25 and 0.50. The boundaries of the 5’ deletable region appear sharply delineated, and those of the 5’ region are more variable in position. Overall, the deletions, though in the same left half of the genome, span a broader region than those in DI particles isolated by Cole et al. (1971), and fingerprinted by Lee and Wimmer (1976). Eleven oligonucleotides (2, 6, 20, 22, 34, 37,55,63,71,65, and 68)are missing, or are greatly reduced in the fingerprint of DI(A) RNA (Fig. 4C). In fact, this fingerprint is almost identical to that of fragment A5

TYPE

I GENOMIC

RNA

387

in Fig. 3, verifying that all of these deletions are located in the 5’ half of the genome. Spots 6,71,65, and 22 of the DI(A)deletions correspond to the spots missing in the fingerprints of Lee and Wimmer. The fingerprints in Fig. 3 show that all of these spots map close to or between 0.14 and 0.22 on the genome. Spots 2 and 37 are present in the fingerprints of Lee and Wimmer, but they are missing in the fingerprints of DI(A) RNA (Fig. 4C). The latter spots are mapped in Fig. 3 between 0.08 and 0.14 map units showing that they are associated with the more 5’ deletion in the new DI (A) particles. T,-Resistant Poly(A-) VPg 115

Oligonucleotides in the Fragments, VPg 75 and

The smallest poly(A-) fragments, VPg 75 and VPg 115, produced in the greatest molar amounts of RNase III digestion of polio RNA (Table 1) were extracted from the gel slices, divided in two parts, and one-half of each sample was digested with proteinase K as described under Materials and Methods. After phenol extraction, and digestion with ribonuclease T1, oligonucleotides were separated in two-dimensional gels (Fig. 5) and then analyzed by secondary digestion and electrophoresis on PEI for comparative analyses (data not shown). Because none of the large, unique spots in the whole RNA fingerprint (Fig. 4A) are present in these 5’ fragments, visual examination of a fingerprint containing intact RNA would not reveal the presence or absence of the 5’ terminal loo-base sequence of the RNA. Overall, the fingerprint pattern of VPg 75 resembles that published for a similar fragment of 100 bases produced by processing of genomic RNA under different salt conditions (Harris et al., 1978). Clear differences in the fingerprints, however, suggest that the 5’ sequence of the RNA in our virus differs from that of the viral RNA used by Harris et al. (1978). For example, the position of the largest oligonucleotide (1) in our fingerprints indicates that though probably similar in size, its composition (relatively U rich), and there-

STEWART, CROUCH, AND MAIZEL

388

- Protease

-k Protease

5

FIG. 5, Two-dimensionaf fingerprints of VPg 75 and VPg 115. Super~ata~lts of gel homogenates of fragments VPg 75 and VPg 115 were divided in two. One-half was treated with proteinase K at 37” as described under Materials and Methods and the other was similarly incubated. (An unincubated control for each gave identical fingerprints to A and C.) “X” marks the VPg-associated oligonucleotide in VPg 75, arrows mark tracks of partial digestion products. Numbers are not related to those in the whole RNA fingerprint in Fig. 4A. Identical numbers in A and C mark spots with identical compositions by secondary a&y&s as described under Mat,erials and Methods. Small letters denote minor spots.

fore its sequence, is different in our virus. By secondary analysis (Stewart and Crouch, 1979b), the composition of this oligonucleotide greatly resembles that determined for spot 62 in the whole RNA fingerprint. Extra spots (2 and 3) are present in our VPg 75, and no small spot corresponding to the smallest one seen in the fingerprints of Harris et al. (1978) is detectable, indicating that there are ather differences in our 5’ fragments. Treatment of the fragment VPg 75 with proteinase K prior to digestion with ribonuclease T, produces a new spot (X) in the ~gerp~nt, and also tracks of incomplete digestion products (arrows}. A secondary

analysis of the composition of X is compatible with that published by Harris et al. (1978) (data not shown). This is good evidence that VPg is indeed present in VPg 75, and that spot X in our fingerprint is very similar, if not identical to, the VPg-adjacent octanucleotide at the 5’ end of the genome. The fingerprint of VPg 115 (Fig. 5C) contains many oligonucleotides in common with VPg 75 (spots 1, 2, 3, 5, 6, 9, 10, and 13). However, they are apparently present in widely differing amounts. furthermore, under the digestion conditions used here, tracks of partial digestion products are visible in (D), but no oligonucleotide eorresponding “X” is produced. This suggests

OLIGONUCLEOTIDE

MAP OF POLIOVIRUS

TYPE

I GENOMIC

RNA

389

A comparison of all the A- fragments shows that they are uniformly deficient in the unique oligonucleotides 9, 10, and 16 which mark the presence of the 3’ terminal fragment, El (Fig. 3). Since they are also lacking the 5’ fragment VPg 75 and 115 as described above, all the A- fragments Oligonucleotide Map of Poliovirus Genomic must be derived from the region between 0.02 and 0.92 on the map. RNA Fragment A-l, 7000 bases in size and Seventy-six spots mapped in the frag- present in very low yields (Table l), reprements in Fig. 4 are shown positioned on the sents the product of the primary, terminal poliovirus genomic RNA (Fig. 6). Until cuts on the genome. Fragment A-2, 5700 more data are collected, spots numbered 23 bases long (0.74 genome), is missing oligoand greater and not subscripted refer nucleotide 4, setting the 3’ terminus on the simply to positions in the whole RNA 5’ side of 0.86. These fragments could fingerprint (see Materials and Methods be positioned variably between 0.02 and and Fig. 3, Results). 0.12 (5’) and 0.74 and 0.86 (3’). Since spot All of the spots numbered from 1 to 22 (the 2 appears reduced, the majority of the largest unique spots) are rather evenly fragments are probably derived from the distributed along the genome. The identifi- region 0.12-0.86. Fragment A-3, 4800 cation and distribution of A-rich oligo- bases (0.62 genome), like A-2, has its 3’ end nucleotides is of interest because they may limited to 0.86. Terminal spots in this bind specific proteins (Goldenberg et al., fingerprint appear to be evenly reduced 1979) or in phase, may code for lysine-rich (see spots 2 and 6; and spots 5 and 7b), proteins. These are less evenly distributed indicating that the majority of the fragalong the RNA with the greatest concen- ments overlap about evenly in the middle tration of them located in fragments D3 - D4 of the genome. and the 5’ end of El. Fragment A-4 (not shown) gave essenInterestingly, all of the spots missing in tially a complete fingerprint with El DI(A) RNA fingerprints map between posi- missing positioning the majority of the fragtions 0.08 and 0.22, even though by EM ments (3800 bases, 0.49 genome), equally mapping (Lundquist et al., 1979) the 3’ with little overlap along the genome. The deletable region can extend rightward as far majority of fragments in A-5 (2600 bases, as 0.5 on the genome. That none of the 0.34 genome), less El, could be centered spots mapping between 0.25 and 0.5 ap- around 0.65 on the map since spot 11 is pear to be detectably reduced in the DI accented in the fingerprint. fingerprints (Fig. 4C), is probably because of the relatively heterogeneous distribution of the 3’ deletions in these mixed popuDISCUSSION lations of particles. Processing of the poliovirus type I genomic RNA by E. coli ribonuclease III Fingerprint Analysis of Large Poly(A-) under controlled conditions produces a large Fragments: Positioning on the Genome number of discrete fragments (Fig. 1). StepSince the poly(A-) fragments are not wise analysis of these fragments at 200- to end-selected, they are composed of a 300-base intervals along the RNA has proheterogeneous mixture of fragments with vided a high-resolution, T, ribonuclease5’ and 3’ ends at variable positions on the resistant oligonucleotide map (Fig. 6). Clear differences however in both the genome. However, by analysis of the unique oligonucleotides mapped close to each map and the fingerprints described here terminus (in Fig. 7), large poly(A-) frag- suggest that there may be true genetic ments can be roughly positioned within the differences between our laboratory strain of virus and various other strains mainfull length of genomic RNA. that the protein in VPg 115 may be more resistant to digestion with proteinase K than that associated with VPg 75. A similar fragment thought to be a partial RNase III digestion product has been described by Harris et al. (1978).

390

STEWART,

CROUCH,

AND MAIZEL

0.4 5

83 I

3, 4.8Kb

0.64

0.60

0.86

0.92

1

t

4

3

2

A- fragments I

6 FE. 7. T~,~-~jim~nsional fmgerprints of large poly(A-1 fragments+ Missing spots are marked by arrows. Numbers refer to spots described in text. Salid horizontal lines define region of ~enome containing majority of f~a~~nts from ~gerp~int (see text).

a.hs

A6

6

A? VWA joz

m2 f .09

2, 5.7Kb

A-,Fragment 1, 7.OKb

STEWART, CROUCH, AND MAIZEL

392

/=4 -_---c

y---.---------

_---140 . . . . . . . GAGUCUG

150 . . . . . . . . . . CUAAAAUCAG _____e-----

130 . . GAAGAG

1 CCAAUUG

----I 120 . . . ., CUUUAUUG

m 10

100

AG

UACUCAACAUUG

8

0

70

ACUCAUUUUAG

CCGUUGGCUUG

60

SO UCGAAUUG

UAACCCDACCUCAC

66a

“iCCC,C,

56a

68

36 90

. CUCCCAG

I

30

40 UCAI’ACUG

GA u u GGG

40a

C U G I!A G C C G

48

10

20 “AAA”UU”“CUUUAAU”C~

GAGG

-

poly

A

-

3’(60-90b)

:

16

FIG. 8. Positioning of T,-resistant oligonucleotides in fragment E4 within the 3’ 150-base sequence of poliovirus Type I strains. Sequence from Porter ela2. (19’78).T,-Resistant oligonucleotides greater than 5 bases in length are grouped together. Fingerprint insert from fragment E4, Fig. 3. Solid underlining defines sequence corresponding to oligonucleotides. (*) Defines possible modified base (Stewart and Crouch, 1980b). Position of low-affinity RNase III site producing visible band at high enzyme concentration shown by arrow in sequence and structure. Dashed line defines ragged 5’ end caused by variable length, 60-90 bases (Yogo and Wimmer, 1972); 40-80 bases (Ahlquist et al., 1979) of poly (A) tract.

tained and characterized in separate laboratories. For example, the RNA fingerprint of Type I virus grown in this laboratory for the last 10 years differs from the fingerprint of Lee and Wimmer (19X), and also Flanegan et al. (1977)-particularly in the eonfigurationofspots 13-14-15and 72-7374. Furthermore, the position of spot number 2 on our map, verified by the DI(A) fingerprints and electron microscopic, mapping as being very near the 5’ end of the RNA is in agreement with the suggestion of Flanegan et al. (1977) who fingerprinted the shortest plus strands of RNA syn-

thesized by replicative intermediate complexes. It does not agree with the position assigned it by Nomoto et al. (1979) near the middle of the genome. In the latter case, these, and other differences, could also be due to the fact that the latter authors generated their fragments by weak alkaline hydrolysis, and purified them in sucrose gradients. Evidence jbr ~~n~e~~t~on in Terminal Sequences

and

V~~~t~on

The apparent similarity of the VPgassociated octanucleotide in the 5’ frag-

OLIGONUCLEOTIDE

MAP OF POLIOVIRUS

ment, VPg 75 (Fig. 5) to that published by Karris et al. (1978) suggests that this short sequence may be conserved in different strains of poliovirus type I (Mahoney) maintained in separate laboratories. However, the differences in the fingerprints of the 5’-adjacent loo-base sequence of our strain compared to those of Harris et al. (1978) suggest that this region varies among different laboratory strains of the virus. The absence of the VPg-associated octanueleotide (X) in the VPg 115 fragment and the fact that many spots shared by the VPg 75 and VPg 115 fra~ents are present in different relative intensities in the latter fragment suggest that the protein associated with VPg 115 may be in a more proteinase K-resistant configuration than that in VPg 75 or that possibly it differs in other ways. It is also possible that VPg 75 is not wholly contained within VPg 115 on the same strand of RNA-a result that would be obtained if our virus stocks contain a mixed population of particles with different 5’ terminal sequences. The 3’ termini of our laboratory strain, however, may be very similar, if not identical to that of other strains of Mahoney Type I poliovirus which have been sequenced in other laboratories (Porter et al., 1978), suggesting that the 3’ termini of different laboratory strains may be highly conserved. To demonstrate this, we can show that the oligonucleotides of our fragment E4 can be located within the published sequence (Fig. 8). Five dominant Tl-resistant oligonucleotides 16, 40a, 66a, 56a, and 48 (from their position and size in two-dimensional fingerprints and from compositions in secondary digests, Stewart and Crouch, 1980b) are marked in the sequence with heavy underlining. Four other spots, 23, 36, 68, and an unmarked one to the upper left of spot 48 are additionally present in the tingerprint though in reduced amounts. This is likely due to the heterogeneity of size of the poly(A) tract (Yogo and Wimmer, 1979; Ahlquist et al., 1979), causing a 45-base, 5’ “ragged” end on the fragment varying between bases 110 and 155 where these spots and some possibly ambiguous bases are located. As mentioned under Results, Fig. 1, this region at higher enzyme concentrations con-

TYPE

I GENOMIC

RNA

393

tains a weak ribonuclease III site. By computer analysis of this sequence in Fig. 8, a structure contai~ng a possible ribonuclease III site at base 134 in the sequence may explain why no spot corresponding to the sequence around this base is found. Coincidence of RNase III Sites on Poliqvirus Genomic RNA with Primary and Stable Translation Products of the Message E. coli ribonuclease III recognizes specific structures in many RNA molecules of viral and cellular origin. The system best studied, however, is that of T, early mRNA in which monocistronic mRNA can be generated by cleavage of the precursor RNA by ribonuclease III at intergenic sites to produce the separate classes of messenger RNAs (Dunn and Studier, 1975). It is interesting that for the 0.3 gene, at least, processing is essential for efficient translation of the gene, and the initiator codon is just 35 bases 3’ of the cutting site (Steitz and Bryan, 1977). A refined map of the gene products of poliovi~s RNA incorporating several methods of analysis has been recently described (Rueckert et al., 1979). Superimposition of the primary and stable gene products upon the major ribonuclease III sites on the poliovirus genome shows that as for T, early mRNAs, the sites appear to punctuate internal cistrons of the message (Fig. 9). The only apparent misfit is within the capsid region of the genome. The precursor la (95kdaltons, Rueekert, 1979) requiring 2.6kb coding capacity in the message fits very well into the region between Al and 131, containing 263kb. This sequence contains an excess of 30 bases over that needed to encode la. However, the true size of la is thought to be greater than 95kdaltons (Rueckert et al., 1979) and the sum of the capsid products, VP 4, VP 2, VP 3, and VP 1 (8,31,26, and 35kdaltons, respectively), requires 2.73-kb coding capacity, or 100 more bases than there are in the Al-B1 region. Since present evidence does not support the idea of overlapping genes in the capsid region (Rueckert, personal communication), it is

t

5

!

.Ol

I .23

A ,074

Al A2

(l.17)

VP0

,_,’

A .25

2.4

A6

3cc1.7,

,P’ L_

AI .09

t

VP3 t.71 I

t1.91)

A3

VP2 1.85)

11.17)

VP0

3a

la ( 2.61

CAPSID

I

VP1t.951

VPlC.95)

A 35 I

Bl 40

A

82

2.2

x LSSJ

5b t 1.~)

3 b (I.~I

I 53

87 AI

.62



32 6

A

A .6-l t

D’l

Cl

9ar 381_;.

NON-CAPS10

(2.3)

I

1.5

NCVP2

lb (2.1)

: : :

.I5 .54

.s2 34 1 f

I AA

El,ti

i---i

: : i

I

I

\

3’

W-

\

stobre Proians

and

t~~m4dlate,

Rimory,

-RNA.% III SITES fractional length - lntrogenic Siie (Kbf



I

FIG. 9. Punctuation of translational products by ribonuclease III sites on the poliovirus genome. Gene products are mapped as described by Rueekert et al. (1979). Alternate cleavage patterns for capsid proteins are shown. Diagonal dashed line (in 3a) shows nonfitting overlap of cleavage products. Lines are proportional to coding sequence needed. Numbers in parentheses are kilobases needed to encode each product (using average of 110 daltons per amino acid, and 340 daltons per nucleotide).

VP91

122) ;vp4

,

F

5 G

% u

3

OLIGONUCLEOTID~

MAP OF POLIOVIRUS

possible that the position of the 5’ end of the B fragment is misplaced by 100 bases needed to completely encode the stable gene products. Otherwise, the “fit” of all the other polypeptides, notably 9a, 7c, and NCVP 4, within the RNase III-generated fragments is surprisingly good, and it closely resembles a similar relationship between ribonuclease III sites and gene products observed with T, early mRNAs (Dunn and Studier, 1975). For example, the precursor to X, polypeptide 3b (65kdaltons) requires a coding sequence of 1.8kb, and 9a, another primary cleavage product (Rueckert et al., 1979), requires an additional sequence of 0.38 kb (a total of 2.18 kb). These gene products fit very well between Bl and Cl leaving only 20 extra bases in this region. The polymerase precursor, lb, requires 2.3kdaltons coding capacity. Polypeptide 7c, requiring 0.57 kb to encode it, fits nicely into the C-D region with 50 bases left over. Between D and E there are 1.5 kb, and this is exactly the needed coding sequence for NCVP4, 56kdaltons. Apparently the tryptic peptides for the genomic protein, VPg, are present in lb, but not in NCVP2 or 7c (Rueckert personal communication). This basic protein of 660 daltons (Golini et al., 1979) would require 0.18 kb of coding sequence. It is interesting to speculate that a small polypeptide could be encoded by the E fragment, extending from map position 0.92 to 0.95 (0.18 kb) on the genome. The 5’ end of this region (0.92-0.94) contains tracts of A-rich oligonueleotides (Fig. 6) that could encode lysine residues in a basic protein. Overall, this positioning of the coding sequences within the framework of the ribonuclease III sites suggets that an initiating AUG would be positioned at, or close to, the ribonuclease III cutting site at Al (approximately 75-100 bases from the 5’ end) (Fig. 7 and Harris et al., 1978). The 3’ end would contain termination signals approximately 430 bases from the 5’ end of the poly(A) tract in fragment E2. Analyses of the translatability of RNase III-generated fragments are proving most interesting. Recent studies (~cClain et al., 1980b) show that ribonuclease III processing

TYPE

I GENOMIC

RNA

395

of poliovirus genomic RNA enhances synthesis by reticulocyte lysates of a 56,000da&on protein. Furthermore, the isolated D fragment indeed encodes a 56,000-dalton polypeptide that comigrates with NCVP4 in SDS-acrylamide gels. In contrast, the translatability of adenovirus mRNAs (which by electron microscopy are known to have more secondary structure; Maizel, unpublished data) is obliterated after processing by ribonuclease III (Westphal and Crouch, 1977). Taken together, these data strongly suggest that, like T7 early mRNA, poliovirus genomic RNA possesses unique intergenic structures specifically recognized by E. coli ribonuclease III. Because RNase III sites on polio RNA may be described as “secondary” compared to the sites on T, mRNA (Nomoto et al. 1979), they could differ slightly in structure from procaryotic sites. The idea of a possible role for specific secondary st~ctures exerting negative translational control, and as sites for processing the polioviral RNA to generate specific subgenomic RNA fragments in infected cells, has been suggested by studies with other RNA-containingviruses (Hunter et al., 19’76).In this context, it is interesting that enzymes having ribonuclease III-like activity have been found in eucaryotic cells (Hall and Crouch, 1977). Whether or not specific restriction of poliovirus RNA can be easily demonstrated in t:ivo, these fragments, having provided an oligonucleotide map for the poliovirus genomic RNA, should be most useful in further structural and functional studies currently being pursued in this laboratory. ACKNOWLEDGMENTS We are grateful to Dr. Ronald Lundquist for providing us with inocula of DI particles and also for many encouraging and helpful discussions throughout this work. The contribution of Ms. Terri Broderick in preparation of this manuscript is greatly appreciated. REFERENCES P., and KAESBERG, P. (1979). Determination of the length of poly(A) at the 3’terminus of the virion RXAs of EMC virus, poliovirus, rhinovirus, RAV-61 and CPMV and of mouse globin mRNA. Nucleic Acids Res. 7, 1195-1204.

AIILQUIST,

396

STEWART, CROUCH, AND MAIZEL

ATKINS, J. F., STEITZ, J. A. ANDERSON,C. W., and MODEL, P. (1979). Binding of mammalian ribosomes to MS2 phage RNA reveals an overlapping gene encoding a lysis function. Cell 18, 24’7-256. BARRELL, B. G. (19’71). Fractionation and sequence analysis of radioactive nucleotides. In “Procedures in Nucleic Acid Research” (G. L. Cantoni and D. R. Davies, eds.), Vol. 2, pp. 751-779. Harper & Row, New York/London. COFFIN, J., and BILLETER, M. (1976). A physical map of the Rous sarcoma virus genome. J. Mol. Biol. 43,293-318.

COLE, C. N., and BALTIMORE, D. (1973). Defective interfering particles of poliovirus. J. Viral. 12, 1414-1426. CROUCH, R. J. (1974). Ribonuclease III does not degrade deoxyribonucleic-ribonucleic acid hybrids. J. Biol. Chem. 249, 1314-1316. DUNN, J. J. (1976). RNase III Cleavage of singlestranded RNA. J. Biol. Chem. 251,3807-3814. DUNN, J. J., and STUDIER, W. (19’75).Effect of RNase III cleavage on translation of bacteriophage T7 messenger RNA’s. J. Mol. Biol. 99, 487-499. DUNN, J. J., and STUDIER, F. W. (1975). Processing transc~ption and translation of bacteriophage T7 messenger RNAs. In “Processing of RNA” (J. J. Dunn, ed.). U. S. Dept of Commerce, Springfield, Va. EAGLE, H. (1971). Buffer combinations for mammalian cell culture. Science 174, 500-503. FLANEGAN, J. B., PETTERSSON,R. F., AMBROS,V., HEWLETT, M. J., and BALTIMORE, D. (1977). Covalent linkage of a protein to a defined nucleotide sequenee at the 5’ terminus of the virion and replieative intermediate RNAs of poliovirus. Proc. Nut. Acad. Sci. USA 74, 961-965. FRISBY, D. P., NEWTON, C., CAREY, N. H., FELLNER, P., NEWMAN, J. F. E., HARRIS, T. J. R., and BROWN, F. (1976). Oligonucleotide mapping of picornavirus RNAs by two-dimensional eleetrophoresis. Virology 71, 379-388. GOLDENBERG,S., VINCENT, A., and SCHERRER,K. (1979). Evidence for the protection of specific RNA sequences in globin messenger ~bonueleoprotein particles. Nucleti Acids Res. 6, 2787-2797. GOLINI, F., NOMOTO,A., and WIMMER, E. (1978). The genome-linked protein of picornaviruses. IV. Difference in the VPg’s of encephalomyelitis virus and poliovirus as evidence that the genomelinked proteins are virus-coded. Virology 89, 112-118.

HALL, S. and CROUCH, R. J. (1977). Isolation and characte~~ation of two enzymatic activities from chicken embryo which degrade double-stranded RNA. J. Biol. Chem. 252,4092-4097. HARRIS,T. J. R., DUNN, J. J.,and WIMMER, E. (1978). Identification of specific fragments containing the 5’ end of poliovirus RNA after ribonuclease III digestion. Nucleic Acids Res. 5, 4039-4054.

HUNTER, T. R., HUNT, R., KNOWLAND, J., and ZIMMERN, D. (1976). Messenger RNA for the coat protein of tobacco mosaic virus. Nature (London) 260, 759-764.

IHLE, J. N., LEE, K-L., and KENNEY, F. T. (1974). Fraetiona~on of 34s ribonucleic acid subunits from oncornaviruses on pol~~dylat~seph~ose columns. J. Riol. Chem. 249,38-42.

JOHO, R. H., BILLETER, M. A., and WEISSMAN, C. (19’75).Mapping of biological functions on RNA of avian tumor virus: location of regions required for transformation and determination of host range. Proc. Nat. Acad. Sci. USA 72,4772-4776. KACJAN, D. L., and MYERS, J. C. (1977). Synthesis of extensive, possibly complete DNA copies of poliovirus RNA in high yields and at high specific activities. Proc. Nat. Acad. Sci. USA 74,2840-2843. KENNEDY, S. I. T. (1976). Sequence relationships between the genome and intracellular RNA species of standard and defective-interfering Semliki Forest virus. J. Mol. Biol. 108, 491-511. LEE, Y. F., and WIMMER, E. (1976). Fingerprinting high molecular weight RNA by two-dimensional gel electropho~sis: Application to poliovirus RNA Nucleic Acids Res. 3, 1647-1658. LEIS, J. P., MCGINNIS, J., and GREEN, R. W. (1978). Rous sarcoma virus PI9 binds to specific doublestranded regions of viral RNA: Effect of P19 on cleavage of viral RNA by RNase III. Virology 84,87-98.

LUNDQUIST,R. E., SULLIVAN, M., and MAIZEL, J. V., JR. (1979). Characterization of a new isolate of poliovirus defective interfering particles. Cell 18, ‘759-769. MCCLAIN, K., STEWART, M. L., and MAIZEL, J. V., JR. (1980a). Three ribosome binding sites on poliovirus type I genomic RNA. Virology, in press. MCCLAIN, K., MAIZEL, J. V., JR., and STEWART, M. L. (1980b). Differential translation of a ribonuclease III-generated subgenomic messenger RNA of poliovirus. Virotogy, in press. MILLIS, D. R., NISHIHARD, T., DOBKIN, C., KRAMER, F. R., COLE, P. E., and SPIEGELMAN, D. (1977). The role of template structure in the recognition mechanism of 2B replicase. In “Nucleic AcidProtein Recognition” (H. J. Vogel, ed.). Academic Press, New York. NOMOTO,A., DETJIN, B., POZZATTI,R., and WIMMER, E. (1977). The location of the polio genome protein in viral RNAs and its implication for RNA synthesis. Nature ~L~~~ Z&270-272. NOMOTO,A., JACOBSON,A., LEE, Y. F., DUNN, J., and WIMMER, E. (1979a). Defective interfering particles of poliovirus: Mapping of the deletion and evidence that the deletions in the genomes of DI(l), (2) and (3) are located in the same region, J. Mol. Biol.

128, 179-196.

NOMOTO,A., LEE, Y. F., BARICH, A., JACOBSON,A.,

OLIGONUCLEOTIDE

MAP OF POLIOVIRUS

DUNN, J. J., and WIMMER, E. (1979b). Restricted fragmentation of poliovirus type 1, 2 and 3 RNAs by ribonuclease III. J. Mol. Biol. 128, 165-1’77. PENMAN, S. (1976). R&j,A metabolism in the HeLa cell nucleus. J. Mol. Biol. 17, 117-130. PINCK, L., FRANCK, A., and FRITSCHE, C. (1979). Formation of ribosome RNA-initiation complexes with alfalfa mosaic virus RNA 4 and RNA 3. Nucleic Acids Res. 7, 151-166. PORTER, A. G., FELLNER, P., BLACK, D. N., ROWLANDS,D. J., HARRIS, T. J. R., and BROWN,F. (1978). 3’Terminal nucleotide sequences in the genome RNA of picornaviruses. Nature (London) 276,298301. ROBERTSON,H. D., DICKSON, E., and DUNN, J. J. (1977). A nucleotide sequence from a ribonuclease III processing site in bacteriophage T7 RNA. Proc. Nat. Acad. Sci. USA 74, 822-826. ROSENBERG,M., and KRAMER, R. (1977). Nucleotide sequence surrounding a ribonuclease III processing site in bacteriophage T7 RNA. Proc. Nat. Acad.

TYPE I GENOMIC RNA

397

COLONNO,R. J., and SHATKIN, A. (1979). Interaction of a limited set of proteins with different mRNAs and protection of 5’-caps against pyrophosphatase digestion in initiation complexes. Nucleic Acids Res. 7, 15-29. STEITZ, J. A., and BRYAN, R. A. (1977).TWOribosome binding sites from the gene 0.3 messenger RNA of bacteriophage T7. J. Idol. Biol. 114, 527-543. STEWART, M. L., and CROUCH, R. (1980a). A high resolution SDS-acrylamide-agarose, high salt gel for analysis and preparation of ribonucleic acids, In preparation. STEWART, M. L., and CROUCH, R. J. (1980b). A sensitive, two-dimensional RNA fingerprinting gel method. Anal. Biochem., in press. WANG, L.-H., DUESBERG, P., BEEMON, K., and VOGT, P. K. (1975). Mapping Tl RNase-resistant oligonucleotides of avian tumor virus RNAs: Sarcoma specific oligonucleotides are near the poly A end and oligonucleotides common to sarcoma and transformation-defective viruses are at the Sci. USA 74, 984-988. poly A end. J. Viral. 16, 1051-1070. RUECKERT, R. R., MATHEWS, T. J., KERR, 0. M., WERTZ, G. W., and DAVIS, N. L. (1979). RNase III PALLANSCH,M., MCLEAN, C., and OMMILIANOWSKI, cleaves VSV genome-length RNAs but fails to cleave D. (1979). Synthesis and processing of picornaviral viral mRNAs. J. Viral. 30, 108-115. polyprotein. In “The Molecular Biology of PicornaYOGO, Y., and WIMMER, E. (1972) Polyadenylic acid viruses” (R. Perez-Bercoff, ed.). Plenum, New York. at the 3’-terminus of poliovirus RNA. Proc. Nat. SONENBERG, N., MORGAN, M. A., TESTA, D., Acad. Sci. USA 69, 1877-1882.