comparing acrylamide gel patterns of restriction enzyme digests of plasmid pbr322 with those of pbr322/fpv2-22, the plasmid

Transcription

1 Proc. Nati. Acad. Sci. USA Vol. 77, No. 9, pp , September 1980 Biochemistry Nucleotide sequence of influenza virus RNA segment 8 indicates that coding regions for NS1 and NS2 proteins overlap (fowl plague virus/dna cloning/nucleotide sequence determination/nonstructural protein genes) ALAN G. PORTER, JOHN C. SMITH, AND J. SPENCER EMTAGE Searle Research and Development, Division of G. D. Searle & Co. Ltd., Lane End Road, High Wycombe, Buckinghamshire, England Communicated by Purnell W. Choppin, May 27,1980 ABSTRACT The smallest RNA segment of influenza A viruses (vrna segment 8) has recently been shown to code for two unrelated nonstructural proteins (NS1 and NS2) translated from separate mrnas. Molecular weight considerations indicated that there might not be enough space on vrna segment 8 for the two coding regions unless they overlap. We have recently cloned in bacteria plasmids several genes of an avian influenza A virus, fowl plague virus (FPV), and now present the complete nucleotide sequence of FPV RNA segment 8 largely determined from the cloned DNA. The DNA sequence predicts two open protein synthesis reading frames that can be translated into polypeptides of sizes similar to those of NS1 and NS2. The coding regions for these polypeptides overlap by the equivalent of amino acids, the exact amount depending on which of several possible methionines initiates the synthesis of NS2. Influenza virus has a segmented genome, consisting of eight separate and unique single-stranded RNA segments ranging from 890 to about 2500 nucleotides (1-4), each of which is transcribed into mrna that codes for a different viral polypeptide (5-7). These polypeptides are: the three polymeraseassociated proteins (P1, P2, and P3), the hemagglutinin (HA), the nucleocapsid protein (NP), the neuraminidase (NA), the membrane or matrix protein (M), and the nonstructural protein (NS1). Several groups also observed a ninth virus-coded polypeptide (NS2) in infected cells (8-12). NS2 is not a breakdown product of any other viral polypeptide (12) and is translated from a separate species of mrna (13-15). These findings, together with the absence of a ninth virus RNA segment, led to the discovery (14, 15) that the smallest genome RNA species (segment 8), unlike the other segments, codes for two polypeptides (NS1 and NS2). Although it was not possible to map the positions of the NS1 and NS2 coding regions within segment 8, it was suggested that they might overlap, from a consideration of the combined molecular weights of the proteins and the size of the RNA (14, 15). Here we present the complete nucleotide sequence of fowl plague virus (FPV) RNA segment 8 determined largely from cloned DNA. The sequence predicts that the coding regions for NS, and NS2 do indeed overlap, and allows the complete primary structure of the NS1 protein to be deduced from the gerretic code. MATERIALS AND METHODS Synthesis of Influenza cdna and Molecular Cloning. In vitro synthesis of cdna copies of FPV (Rostock strain) RNA was as described (16). A bacterial clone carrying a plasmid with a segment 8 insert was isolated in the cloning experiment described in ref. 17, where details of plasmid construction, The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C solely to indicate this fact. transformation, and identification of clones may be found. All cloning experiments were performed in a Category III laboratory inspected and approved by the Genetic Manipulation Advisory Group (GMAG), London. Restriction Enzyme Mapping. The number and positions of sites for each enzyme shown in Fig. 1 were determined by comparing acrylamide gel patterns of restriction enzyme digests of plasmid pbr322 with those of pbr322/fpv2-22, the plasmid with the 759-base-pair segment 8 cdna inserted at the single HindIll site of pbr322. More detailed map positions were determined from restriction enzyme digestions of the purified segment 8 insert labeled with 32P at the 5' ends. In both cases, the sizes of the fragments were estimated by reference to the mobilities of fragments of known length derived from a Hae III digest of phage 4X174 DNA (New England BioLabs). Gel Electrophoresis and Elution. Agarose and sequencing gels were prepared as described (17, 18). Separation of restriction fragments of nucleotides was by electrophoresis in 40-cm-long polyacrylamide gels (6% acrylamide/0.2% bisacrylamide in 50 mm Tris/borate, ph 8.3/1 mm EDTA) at V for 16 hr. Elution of DNA was by crushing and soaking (19). The eluate was freed of acrylamide monomer by gravity filtration through a 0.5-ml column of DEAE-cellulose (Whatman DE52) (20). RNA fragments were separated in 40-cm gels made up in 3% acrylamide/0.15% bisacrylamide/25 mm citric acid, ph 3.5/2 mm EDTA/7 M urea. Electrophoresis was at 150 V for 16 hr. DNA Sequence Determination Methods. The chemical sequencing procedure (19) was used on restriction enzyme fragments labeled with 32p only at one 5' or 3' end (17, 20). Sequences of ': nucleotides from the HindIII sites were confirmed by the dideoxynucleotide method adapted for 5'- end-labeled restriction fragments (21), with the modification that a "chase" with 0.25 mm concentration of the four deoxynucleoside triphosphates was done at the end of the incubation (20 C for 20 min). The chemical sequencing procedure for RNA was performed essentially as described (22) on 3'-endlabeled RNA (23). Chemicals and Enzymes. Radioactive compounds were from the Radiochemical Centre, Amersham, terminal transferase from P-L Biochemicals, restriction enzymes from New England BioLabs or Bethesda Research Laboratories (Rockville, MD), polynucleotide kinase from New England BioLabs, bacterial alkaline phosphatase (grade F) from Worthington, dideoxynucleoside triphosphates from Collaborative Research, and acrylamide and bisacrylamide "specially pure for electrophoresis" from British Drug House (Poole, England). Abbreviations: NS, nonstructural protein; FPV, fowl plague virus; vrna, virion RNA; CRNA, complementary RNA. 5074

2 RESULTS Identification of Segment 8 Sequences Cloned in pbh322. We have previously described the reverse transcription of the FPV genome (16) and the cloning and sequence analysis of the hemagglutinin gene (17). In the same experiment (17), we also isolated a bacterial clone containing a plasmid, named pbr322/fpv2-22, that hybridized specifically to a 32P-labeled RNA probe derived from segment 8. That the DNA was from segment 8 was confirmed during sequence analysis (see later), which showed that T1 RNase oligonucleotides predicted from the DNA sequence were present in the catalogue of segment 8 T1 oligonucleotides determined from the RNA (24). Restriction Enzyme Mapping and Sequence Determination. There are two HindIII sites in pbr322/fpv2-22, and these mark the extremities of the cloned segment 8 insert (Fig. 1), which was calculated to be t750 base pairs from its mobility in polyacrylamide gels compared to known markers. This is about 150 nucleotides shorter than the full-length RNA segment 8 (1, 16). Within the segment 8 insert, the cut sites for several restriction enzymes were mapped prior to sequence analysis (Fig. 1). Mapping clearly indicated two Ava II sites, but a third was discovered during nucleotide sequence analysis (G-G-A- C-C at nucleotides ; Fig. 2). This site is shown in parentheses in Fig. 1 to indicate that it is completely resistant to Ava II, presumably because the C at nucleotide 387 is methylated as part of She overlapping EcoRII (BstNI) site at nucleotides (Fig. 2). Most of the restriction sites, except those for Mbo II, Sau3A, and Alu I, were labeled for sequence analysis. As an example, the plasmid pbr322/fpv2-22 was digested to completion with Ava II, and the 5' ends were labeled with 32P (17, 19). The two Ava II fragments containing the HindIII sites were cut with HindIII, and the 481- and 189-base-pair fragments, derived from the left and right end of the insert, respectively, were resolved in a 6% polyacrylamide gel. Sequences of nucleotides were determined for both fragments by the partial chemical degradation procedure (19). The arrows beneath the restriction map in Fig. 1 show that the nucleotide sequence of the insert was determined completely on both DNA strands. This was considered necessary owing to the absence of any amino acid sequence information to support the interpretation of sequencing gels. A sequence 100 bp nt C b. )I 3' -818 Biochemistry: I~ Porter et al. w z - C~ FIG. 1. Restriction enzyme map of FPV segment 8 insert in pbr322. The line between the map and the arrows shows the position of the full-length segment 8 RNA relative to the DNA insert with the number of nucleotides missing from the clone indicated at each end. The arrows show the distances in which nucleotide sequences were determined; the one with an asterisk is the single case in which 3' end labeling was performed. bp, Base pairs. g Proc. Natl. Acad. Sc. USA 77 (1980) 5075 determined on both DNA strands is very unlikely to contain errors. The right-hand end of the insert (the right-hand HindIII site; Fig. 1) overlaps the published sequence (26) at the 5' end of segment 8 by 19 nucleotides, showing that 48 nucleotides are missing from this end of the clone (Fig. 2). In the region of the overlap, the DNA and RNA sequences are in agreement. The left-hand end of the insert (the left-hand HindIII site; Fig. 1), on the other hand, does not overlap the published sequence of 72 nucleotides at the 3' end of segment 8 (26), so we determined a sequence of 105 nucleotides directly from the 3' end of segment 8 by 3'-end labeling of FPV RNA with [5'-32P]pCp in the presence of phage T4 RNA ligase (23) and subjecting the gelpurified segment 8 to the chemical sequencing procedure for RNA (22). This sequence now overlaps the left end of the insert by 17 nucleotides, showing that 88 nucleotides corresponding to the 3' end of segment 8 are absent from the insert (Fig. 2). Although the RNA sequence was determined only on one strand, it was found to be in complete agreement with the published 3'-terminal sequence of 72 nucleotides deduced by enzymatic methods (26). In addition, the stretch of RNA sequence we determined (nucleotides 73-88) linking the published 3'-terminal sequence with the left end of the cloned DNA was found to be identical to the corresponding region in cloned DNA derived from segment 8 of X-47, a human influenza recombinant (unpublished data). The nucleotide sequence of segment 8 is shown in Fig. 2. Segment 8 is 890 nucleotides in length, only slightly longer than recent estimates based on migration rate of RNA in polyacrylamide gels containing formamide or urea (16). DISCUSSION The molecular weights of the two proteins (NS1 and NS2) coded by FPV RNA segment 8 have been determined directly and are r23,000 and -11,000, respectively (8-10, 27). In the RNA sequence shown in Fig. 2 there are two open reading frames in the crna (coding strand), which would give proteins of molecular weights 25,980 and 11,019-12,872, presumably NS1 and NSZ, respectively. The value of 25,980 is a slight underestimate, because it does not take into account the fact that NS, is a phosphoprotein (28). NS, is initiated by the AUG at nucleotides and is terminated by the UGA at On the basis of a molecular weight of 4t1,000 for NS2 (8-10), there are four possible methionines that could initiate NS2, at amino acids 1, 3, 6, or 18, shown boxed in Fig. 2. The respective molecular weights of NS2 would be 12,872, 12,585, 12,238, or 11,019. Assuming that NS2 does not undergo posttranslational cleavage, the amino acid sequences in Fig. 2 predict that FPV NS2 has between four and six methionine-containing tryptic peptides, depending on which methionine is the actual initiator. Unfortunately, the published [ass]methionine-labeled tryptic peptide maps of FPV NS2 are not helpful because of contaminating spots (15). In contrast, NS2 of a human influenza A strain (WSN) gives four major methionine-containing tryptic peptides (12). However, it would be dangerous to draw any conclusions about the position of the initiation site of FPV NS2 on the basis of the WSN maps, because the extent of homology between segment 8 of human and avian influenza A strains is less than 100% (29). Whichever methionine initiates NS2, there is extensive overlap of coding regions for NS1 and NS2 of between 127 and 179 nucleotides, corresponding to amino acids. These figures for the overlap are similar to those derived from a combination of SI nuclease mapping and hybrid-arrest translation using restriction fragments of cloned segment 8 from

3 5076 Biochemistry: Porter et al. Proc. Natl. Acad. Sci. USA 77 (1980) vrna 3'-UCGUUUUCGUCCCACUGUUUUUGUAU UAC CA AGG UUG UGA CAC AGU UCG AAA GUC CAU JUG ACG AAA GAA ACC GUA CAG crna 5'-AGCAAAAGCAGGGUGACAAAAACAUA AUG GAU UCC AAC ACU GUG UCA AGC UUU CAG GUA GAC UGC UUU CUU UGG CAU GUC N-terminus NS 1 Met-Asp-Ssr-Asn-Thr-Vel-Ser-Ser-Pho-Gln-Vsl-Aep-Cys-Pho-Lou-Trp-Hie-Val-(1i) GCG UUU G&I AAA CGU CUG GUU CUU UAC CCA CUA CGG GGU AAG GAA CUG GCU GAA GCG GCU CUA GUC UUC AGG GAC 'JCC CCU I 4O,, CGC AAA CGA UUU GCA GAC CAA GAA AUG GGU GAU GCC CCA UUC CUU GAC CGA CUU CGC CGA GAU CAG AAG UCC CUG AGG GGA Arg-Lys-A*-Pho-Al4-Asp-Gln-Glu-Mot-Gly-Asp-Ale-Pro-Pho-Lou-Asp-Arg-Lu-Arg-Arg-Asp-Gln-Lys-Ser-Lou-Arg-Gly-(45) UCU CCG UCG UGA GAA CCA GAC CUG UAG CUG UGU CGA UGA GCA CAA CCU UUC GUC UAU CAC CUC GCC UAA GAC CUC CUG CUU AGA GGC AGC ACU CUU GGU CUG GAC AUC GAC ACA GCU ACU CGU GUU GGA AAG CAG AUA GUG GAG CGG AUU CUG GAG GAC GAA Arg-Gly-Ser-Thr-Lou-Gly-Lou-Asp-Ile-Asp-Thr-Ala-Thr-Ar'g-Val-Gly-Lys-Gln-Ilo-Val-Glu-Arg-Ile-Lou-Glu-Asp-Glu-(72) AGG CUA CUC CGU GAA UUU UAC UGG UAA CGG AGA CAU GGA CGA UGU GCG AUG GAU UGA CUG UAC UGA GAA CUU CUC UAC AGU UCC GAU GAG GCA CUU AAA AUG ACC AUU GCC UCU GUA CCU GCU ACA CGC UAC CUA ACU GAC AUG ACU CUU GAA GAG AUG UCA Ser-Asp-Glu-Ala-Leu-Lys-Mat-Thr-Ile-Ala-Sor-Val-Pro-Ala-Thr-Arg-Tyr-Lou-Thr-Asp-Mat-Thr-Lou-Glu-Glu-Me't-Ser-(99) UCC CUG ACC AAG UAC GAG UAC GGG UUU GUC UUU CAC CGU CCG AGG GAA ACG UAG UCU UAC CUG GUC CGC UAG UAC CCC UUC AGG GAC UGG UUC AUG CUC AUG CCC AAA CAG AAA GUG GCA GGC UCC CUU UGC AUC AGA AUG GAC CAG GCG AUC AUG GGG AAG Arg-Asp-Trp-Phe-Mot-Lou-Mot-Pro-Lys-Gln-Lys-Val-Ale-Gly-Ser-Lou-Cyi-Ile-Arg-Mat-Asp-Gln-Ala-Ilo-Mot-Gly-Lye-(126) UUG UAG UAU GAC UUU CGU UUG AAG UCA CAC UAA AAG CUA GCC GAC CUU UGA GAG UAU GAU AAU UCC CGA AAU UGG CUA CUC AAC AUC AUA CUG AAA GCA AAC UUC AGU GUG AUU UUC GAU CGG CUG GAA ACU CUC AUA CUA UUA AGG GCU UUA ACC GAU GAG Asn-Ile-Ile-Lou-Lys-Ala-Asn-Pho-Sor-Val-Ilo'-Pho-Asp-Arg-Lou-GIU-Thr-Leu-Ile-Leu-Lou-Arg-Ala-Leu-Thr-Asp-Glu-(153) CCU CGU UAA CAG CCG CUU UAA AGU GGU AAC GGA AGA GAA GGU CCU GUA UGA CUA CUC CUA CAG UUU UUA CGU UAA CCC CAG GGA GCA AUU GUC GGC GAA AUU UCA CCA UUG CUU CCA GGA CAU ACU GAU GAG GAU GUC AAA AAU GCA AUU GGG GUC Gly-Ala-Ile-V'al-Gly-Glu-Ile-Ser-Pro-Lou-Pr-SrLou-Pro-Gly-His-Thr-A~so-Glu-Aoo-Val.-Lys-Asn- la-11o-gly-vlal-( 180) N-terminus NS2? 9 Arg Ser-Lys Gin-Lou-Gly-Ser-(10) GAG UAG CCU CCU GAA CUU ACC UUA CUA UUG UGU CAA GCU CAG AGA CUU UGA UAU GUC UCU AAG CGA ACC UCU UCG UCA UUA CUC AUC GGA GGA CUU GAA UGG AAU GAU AAC ACA GUU CGA GUC UCU GAA ACU AUA CAG AGA UUC GCU UGG AGA AGC AGU AAU Leu-119-Gly-Gly-Leu-Glu-Trp-Asn-Asp-Asn-Thr-Val-Arg-Val-Ser-Glu-Thr-Ile-Gln-Arg-Phe-Ala-Trp-Arg-Sor-Ser-Asn-(207) -Ser-Ser-Glu-Asp-'Leu-Ash-Gly I~ll-Thr-Gln-Phe-Glu-Sor-Lou-Lys-Leu-Tyr-Arg-Asp-Ser-Lou-Gly-Glu-Ala-Val-Met-(37) CUC UUA CCC CCU GGA GGU GAG GGA GGU UUC GUC UUU GCC UUU UAC CGC UCU UGU UAA CUC AGU CUU UAA ACU UCU UUA UUC GAG AAU GGG GGA CCU CCA CUC CCU CCA AAG CAG AAA CGG AAA AUG GCG AGA ACA AUU GAG UCA GAA AUU UGA AGA AAU AAG Glu-Asn-Gly-GlyPro-P'ro-Leu-Pro-Pro-Lys-Gln-Lys-Arg-Lys-Mot-Ala-Arg-thr-Ile-Glu-Ser-tlu-Ile. terminue NS1 (230) -Arg-Met-Gly-Asp-Leu-His-Ser-Leu-Gln-Ser-Arg-Asn-Gly-Lys-Trp-Arg-Glu-Glh-Lou-Ser-Gln-Lys-Phe-Glu'Glu-Ile-Arg-(64) U ACC AAU UAA CUU CUU CAC UCU GUA UCU AAC UUC UAU UGU CUC UUA UCA AAA CUU GUU UAU UGU AAA UAC GUU CGG AAU A UGG UUA AUU GAA GAA GUG AGA CAU AGA UUG AAG AUA ACA GAG AAU AGU UUU GAA CAA AUA ACA UUU AUG CAA GCC UUA - Trp-L(u-9l0-Glu-Glu-)al-Arg-His-Arg-L(u-Lys-Ile-Thr-Glu-Asn-S(r-Ph9-Glu-Gln-Ile-Thr-Ph9-M)t-Gln-Ala-Leu- (90) GUU GAU GGC GAA CUU CAC CUC GUU CUC UAU UCU UGA AAG AGC AAA GUC GAA UAA AUU ACUAUUUUUUGUGGGAACAAAGAUGA-5' vrna CAA CUA CCG CUU GAA GUG GAG CAA GAG AUA AGA ACU UUC UCG UUU CAG CUU AUU UAA UGAUAAAAAACACCCUUGUUUCUACU-3' crna Gln-Lsu-Pro-Leu-Glu-Val-Glu-Gln-Glu-Ile-Arg-Thr-Phe-Ser-Phe-Gln-Leu-Ile.(108) C-terminus NS2 FIG. 2. (Legend appears at. bottom of the next page.)

4 Biochemistry: Porter et al A 3' vrna UCGUUUUCGUCC U CG U UAAC~ AGL CC B 3' UCC AC uguuuuu UC UACA UUUUU FIG. 3. Homologies between the 3' end and interior of segment 8. The boxed-in nucleotides are homologous, indicated by lines in Fig. 2. The numbering refers to the distance from the 3' end of the vrna. HoN1 and H3N2 strains of influenza (30). Here the overlap was calculated to be amino acids. In the same experiments the body of the NS2-coding region (-t340 nucleotides) was found to map at the 5' end of the vrna-i.e., in a position corresponding to the 3'-terminal part of the NS, mrna. The nucleotide sequence in Fig. 2 indicates that the same is true in FPV segment 8, in which the termination codon of the NS2 mrna (UAA) occurs within a few nucleotides of where transcription of NS, mrna is known to terminate (31). This strongly suggests that the NS, and NS2 mrnas have a common poly(a) addition site (30). There are basically two mechanisms by which the NS2 mrna could be generated (14, 15). One is by a specific nuclease cleavage of NS, mrna late in infection and the other is by initiation of transcription of NS2 mrna in the interior of segment 8. The first mechanism could involve either a single nuclease cut or excision of unwanted nucleotides (32, 33). The nucleotide sequence does not allow us to distinguish between these possibilities, although the mapping experiments described above (30) support an interpretation that, if splicing did occur, it would not involve cuts in the region coding for NS2. We may speculate that if internal initiation of transcription is the mechanism that generates NS2 mrna, the transcriptase may have to recognize an internal sequence similar to the conserved sequence at the 3' ends of all eight influenza virus RNA segments (26) in order to initiate transcription in the correct place (34). There are two internal areas with a sequence resembling that at the 3' end; the first beginning at nucleotide 419 with just the 3'-terminal hexanucleotide (3')U-C-G-U- U-U(5'), and the second occurring between nucleotides 488 and 499, in which 8 out of 12 nucleotides are a repeat of the 3' end, including the 3'-terminal (3')U-C-G-U-U(5') (Fig. 3A). In the absence of splicing and excluding the 3'-terminal poly(a), the length of the NS2 mrna would be 472 nucleotides if the first and 403 nucleotides if the second sequence is the start point for the viral transcriptase. The value of 403 nucleotides is much closer to 340 nucleotides for the NS2 mrnas of HoN1 and H3N2 strains (30). There is another, more extensive, homology between the 3'-terminal region of the vrna and a sequence close to where Phe UUU 3 UUC 5 Leu UUA 2 UUG 1 Leu CUU 9 CUC 4 CUA 2 CUG 5 Ile AUU 8 AUC 5 AUA 4 Met AUG 10 Val GUU 2 GUC 5 GUA 2 GUG 4 Table 1. Proc. Natl. Acad. Sci. USA 77 (1980) 5077 Use of codons in NS1 coding region Ser UCU 3 Tyr UAU 0 Cys UGU 0 UCC 4 UAC 1 UGC 2 UCA 4 Ochre UAA 0 Opal UGA 1 UCG 0 Amber UAG 0 Trp UGG 4 Pro CCU 4 CCC 1 CCA 5 CCG 0 Thr ACU 8 ACC 2 ACA 4 ACG 0 Ala GCU 4 GCC 2 GCA 6 GCG 2 His CAU 2 CAC 0 Gln CAA 1 CAG 7 Asn AAU 4 AAC 4 Lys AAA 8 AAG 4 Asp GAU 9 GAC 9 Glu GAA 8 GAG 8 Arg CGU 1 CGC 2 CGA 4 CGG 4 Ser AGU 2 AGC 3 Arg AGA 5 AGG 3 Gly GGU 2 GGC 3 GGA 7 GGG 3 In all, 22.6% of codons end in C; 27.0% end in A; 23.9% end in G; and 26.5% end in U. the NS2 mrna may be initiated in which 11 out of 13 nucleotides are the same (Fig. 3B). However, because this repeated sequence is not conserved near the 3' end of any other influenza virus RNA segment (26), it is unlikely to be part of a general transcriptase binding signal (34). The amino acid sequences of NS1 and NS2 show that neither protein has a short NH2-terminal hydrophobic "signal" peptide that is cleaved off, leaving the mature protein (17, 35-37). This indicates that the nonstructural proteins do not extensively interact with or bind to cell membranes. The longest stretch of hydrophobic amino acids in NS1 is from residues (Fig. 2). In NS1 there are several short regions that are noteworthy because one amino acid occurs frequently. These are: methionine 6 times in 32 residues (93-124), arginine 5 times in 12 residues (35-46), and proline 4 times in 5 residues ( ). The latter sequence (Pro-Pro-Leu-Pro-Pro) is uncommon in proteins. Examination of the amino acid composition of NS2 shows that, whichever is the NH2-terminal methionine, at least 25% of the protein is composed of leucine and glutamic acid. As with other genes (38), the codons for some amino acids in NS1 are apparently not selected at random (Table 1). It is interesting that codons ending in CG for serine, proline, and threonine (underlined in Table 1) are not used at all (17). In future studies the exact position of the 5' end of the NS2 mrna coded by segment 8 can be determined from molecularly cloned NS2 mrna. This is also likely to reveal whether or not any part of the 5' terminus of virally coded NS2 mrna is derived from noncontiguous regions of crna by a splicing mechanism (32, 33). Future studies can also be aimed at elucidating the NH2-terminal amino acid sequence of purified. FPV NS2 to determine if one of the four methionines shown in Fig. 2 is the initiator of this protein. We are grateful to Dr. R. Lamb and Dr. S. Inglis for communicating their results before publication, Dr. P. Choppin and Dr. Ching-Ju Lai for interesting discussions, Mr. E. Farmer for performing some of the FIG. 2 (on preceding page). Nucleotide sequence of FPV segment 8. The complementary RNA (crna) is shown beneath the virion RNA (vrna), with every 30th nucleotide numbered. If the codons for NS1 are in the +1 frame, then the codons for NS2 in the overlap region are in the +2 frame. The boundaries of the insert of plasmid pbr322/fpv2-22 are shown by vertical lines, close to which are black triangular arrows above the vrna to indicate the distance from the 5' or 3' end whose sequence was determined directly from the RNA. Lines above and below the vrna indicate homologies (see text) and termination codons. The four possible initiator methionines of NS2 are boxed. The sequence C-C-U-U-C is also boxed to indicate a possible ribosome binding site area (25). The dots beneath amino acids refer to a potential site of glycosylation.

5 5078 Biochemistry: Porter et al. restriction mapping, and Mr. B. Jenkins for excellent assistance. Our thanks also go to Dr. N. H. Carey and Dr. P. Fellner for their advice and encouragement. We are indebted to Dr. A. J. Hale for providing excellent facilities for research. 1. McGeoch, D., Fellner, P. & Newton, C. (1976) Proc. Natl. Acad. Sci. USA 73, Scholtissek, C. (1978) Curr. Topics Microbiol. Immunol. 80, Pons, M. W. (1976) Virology 69, Palese, P. & Schulman, J. L. (1976) J. Virol. 17, Palese, P. (1977) Cell 10, Scholtissek, C., Harms, E., Rohde, W., Orlich, M. & Rott, R. (1976) Virology 74, Inglis, S. C., McGeoch, D. & Mahy, B. W. J. (1977) Virology 78, Skehel, J. J. (1972) Virology 49, Follet, E. A. C., Pringle, C. R., Wunner, W. H. & Skehel, J. J. (1974) J. Virol. 13, Minor, P. D. & Dimmock, N. J. (1975) Virology 67, Etkind, P. R. & Krug, R. M. (1975) J. Virol. 16, Lamb, R. A., Etkind, P. R. & Choppin, P. W. (1978) Virology 91, Stephenson, J. R., Hay, A. J. & Skehel, J. J. (1977) J. Gen. Virol. 36, Lamb, R. A. & Choppin, P. W. (1979) Proc. Natl. Acad. Sci. USA 76, Inglis, S. C., Barrett, T., Brown, C. M. & Almond, J. W. (1979) Proc. Natl. Acad. Sci. USA 76, Emtage, J. S., Catlin, G. H. & Carey, N. H. (1979) Nucleic Acids Res. 6, Porter, A. G., Barber, C., Carey, N. H., Hallewell, R. A., Threlfall, G. & Emtage, J. S. (1979) Nature (London) 282, Sanger, F. & Coulson, A. R. (1978) FEBS Lett. 87, Maxam, A. & Gilbert, W. (1977) Proc. Nati. Acad. Sci. USA 74, Proc. Nati. Acad. Sci. USA 77 (1980) 20. Hindley, J. & Phear, G. A. (1979) Nucleic Acids Res. 7, Maat, J. & Smith, A. J. H. (1978) Nucleic Acids Res. 5, Peattie, D. A. (1979) Proc. Natl. Acad. Sci. USA 76, Keene, J. D., Schubert, M., Lazzarini, R. A. & Rosenberg, M. (1978) Proc. Natl. Acad. Sci. USA 75, Smith, J. C., Carey, N. H., Fellner, P., McGeoch, D. & Barry, R. D. (1978) in Negative Strand Viruses and the Host Cell, eds. Mahy, B. W. J. & Barry, R. D. (Academic, London), pp Kozak, M. (1978) Cell 15, Robertson, J. S. (1979) Nucleic Acids Res. 6, Inglis, S. C., Carroll, A. R., Lamb, R. A. & Mahy, B. W. J. (1976) Virology 74, Privalsky, M. L. & Penhoet, E. E. (1978) Proc. Natl. Acad. Sci. USA 75, Scholtissek, C. & Von Hoyningen-Huele, V. (1980) Virology 102, Lamb, R. A., Choppin, P. W., Chanock, R. M. & Lai, Ching-Juh. (1980) Proc. Natl. Acad. Sci. USA 77, Hay, A. J., Abraham, G., Skehel, J. J. & Smith, J. C. (1977) Nucleic Acids Res. 4, Crick, F. (1979) Science 204, Kolata, G. B. (1980) Science 207, Robertson, H. D., Dickson, E., Plotch, S. J. & Krug, R. M. (1980) Nucleic Acids Res. 8, Strauss, A. W., Bennett, C. D., Donohue, A. M., Rodkey, J. A. & Alberts, A. W. (1977) J. Blol. Chem. 252, Blobel, G. & Dobberstein, B. (1975) J. Cell Biol. 67, Davis, B. D. & Tai, P.-C. (1980) Nature (London) 283, Grantham, R., Gautier, C., Gouy, M., Mercier, R. & Pave, A. (1980) Nucleic Acids Res. 8, r49-r62.