Sequence Analysis and Organization of the Neodiprion abietis Nucleopolyhedrovirus Genome

Abstract

Of 30 baculovirus genomes that have been sequenced to date, the only nonlepidopteran baculoviruses include the dipteran Culex nigripalpus nucleopolyhedrovirus and two hymenopteran nucleopolyhedroviruses that infect the sawflies Neodiprion lecontei (NeleNPV) and Neodiprion sertifer (NeseNPV). This study provides a complete sequence and genome analysis of the nucleopolyhedrovirus that infects the balsam fir sawfly Neodiprion abietis (Hymenoptera, Symphyta, Diprionidae). The N. abietis nucleopolyhedrovirus (NeabNPV) is 84,264 bp in size, with a G+C content of 33.5%, and contains 93 predicted open reading frames (ORFs). Eleven predicted ORFs are unique to this baculovirus, 10 ORFs have a putative sequence homologue in the NeleNPV genome but not the NeseNPV genome, and 1 ORF (neab53) has a putative sequence homologue in the NeseNPV genome but not the NeleNPV genome. Specific repeat sequences are coincident with major genome rearrangements that distinguish NeabNPV and NeleNPV. Genes associated with these repeat regions encode a common amino acid motif, suggesting that they are a family of repeated contiguous gene clusters. Lepidopteran baculoviruses, similarly, have a family of repeated genes called the bro gene family. However, there is no significant sequence similarity between the NeabNPV and bro genes. Homologues of early-expressed genes such as ie-1 and lef-3 were absent in NeabNPV, as they are in the previously sequenced hymenopteran baculoviruses. Analyses of ORF upstream sequences identified potential temporally distinct genes on the basis of putative promoter elements.

The balsam fir sawfly (Neodiprion abietis, Hymenoptera, Symphyta, Diprionidae) is a native sawfly species that occurs throughout Canada and the United States. The balsam fir sawfly is a defoliating species that feeds primarily on the balsam fir Abies balsamea (L.) Mill. and spruce Picea spp. (96). An outbreak of the balsam fir sawfly in Newfoundland has persisted since 1991. This 15-year outbreak is in contrast to the typically observed outbreaks of 4- to 5-year durations that occur every 5 to 15 years. The current infestation, spanning more than 40,000 ha of balsam fir in western Newfoundland, threatens the significant silvicultural investment in that region, as severe defoliation greatly reduces tree growth and may cause tree mortality (81). The application of a virus naturally pathogenic to the balsam fir sawfly, the N. abietis nucleopolyhedrovirus (NeabNPV), may offer great potential in controlling sawfly populations (76). Recently, aerial application of NeabNPV has successfully initiated collapse of growing and peaking balsam fir sawfly populations (71).

NeabNPV is a member of the family Baculoviridae, a well-studied family of invertebrate viruses that have large, double-stranded, circular DNA genomes within rod-shaped nucleocapsids that are enveloped and occluded. The two genera of baculoviruses, Nucleopolyhedrovirus (NPV) and Granulovirus (GV), are typically distinguished by the size and localization of their occlusion bodies (OBs). NPVs have larger nucleus-localized OBs, ranging in diameter from 1 to 5 μm, with singly or multiply enveloped virions. GVs have smaller OBs with diameters of 0.3 to 0.6 μm in which single enveloped virions are normally occluded, and these OBs are distributed throughout the cell following nuclear disintegration. GVs have only been isolated from lepidopteran hosts, while NPVs have been isolated from lepidopterans, hymenopterans, and dipterans. NeabNPV is a singly enveloped NPV that infects the balsam fir sawfly.

Thirty baculovirus genomes are available in GenBank. Of these, the lepidopteran baculoviruses are the most highly represented (27 sequenced genomes). Two NPVs that infect the diprionid sawflies Neodiprion sertifer (NeseNPV) (26) and Neodiprion lecontei (NeleNPV) (62) and one that infects the mosquito Culex nigripalpus (Diptera, Culicidae), CuniNPV (2), have been completely sequenced. Baculovirus genomes range in size from 81.8 kb to 161 kb and have G+C contents ranging from 32% to 56%. The sawfly NPVs have both the smallest genomes and the lowest G+C contents of any baculoviruses sequenced to date (26, 62).

Baculovirus phylogeny was first resolved with the conserved polyhedrin gene sequence. This analysis identified two distinct taxa within the lepidopteran NPVs, termed groups I and II (103). More recently, full-genome phylogenetic analyses suggested that single-gene phylogenies might not accurately resolve evolutionary relationships among baculoviruses (44). Instead, seven genes are considered to share the topology of whole-genome phylogenies (44). Complete-genome phylogeny, along with gene content and gene order phylogenies, suggests that the lepidopteran-infecting baculoviruses form three distinct clades, i.e., GV, group I NPV, and group II NPV (43, 44). The dipteran-infecting baculovirus CuniNPV is taxonomically distinct from the lepidopteran-infecting baculoviruses (44). Recent genome sequencing of the hymenopteran-infecting baculoviruses NeseNPV and NeleNPV has indicated that they represent a separate clade that is distinct from both lepidopteran-infecting and dipteran-infecting baculoviruses (26, 44, 62).

Here we report the genome sequence of NeabNPV. Our data provide further insight into the evolution of sawfly baculoviruses. Furthermore, a detailed analysis of the upstream sequences of predicted NeabNPV open reading frames (ORFs) identified promoter motifs that potentially discriminate temporally distinct genes. Twenty-nine genes are believed to be conserved throughout all of the fully sequenced baculovirus species. This set of genes, however, does not include a number of essential early-expressed genes (ie-1, lef-3, pp31). Because of the essential role that these genes play in the lepidopteran baculovirus life cycle, we investigated promoter characteristics that distinguish early- and late-expressed genes. A subset of predicted early-expressed genes may include functional analogues for these genes.

MATERIALS AND METHODS

Source of NeabNPV.

NeabNPV OBs were produced and purified as described by Moreau et al. (71) This procedure involved thawing frozen (−20°C) larvae in 5 volumes of 0.3% sodium dodecyl sulfate (SDS) solution, followed by homogenization. The sample was twice treated with a process of filtration through a 1-mm² plastic mesh, followed by resuspension of solid matter in 0.3% SDS and rehomogenization and refiltration until the filtrate was clear. Filtrates were then pooled, filtered twice through eight layers of cheesecloth, and centrifuged for 15 min at 9,000 × g. The supernatant was discarded, and the NeabNPV OB pellet was resuspended in 0.3% SDS. Centrifugation and resuspension were repeated until a clear supernatant was obtained. The NeabNPV OB pellet was then resuspended in water and concentrated to 10⁹ OBs/ml.

Virus DNA purification.

NeabNPV OB suspensions were washed with sterile double-distilled H₂O and centrifuged (15,000 × g, 5 min at 20°C) three times and then treated with SDS (0.4%) under agitation for 1 h. The resultant suspension was washed and centrifuged (15,000 × g, 10 min at 20°C) three more times and incubated with dissociation buffer (0.1 M Na₂CO₃, 0.04 M sodium thioglycolate) for 20 min. Debris was pelleted twice by centrifugation (2,000 × g, 5 min at 4°C), and the supernatant was collected. The supernatant was washed, pelleted (15,000 × g, 30 min at 4°C), and resuspended in Tris-HCl. This was then incubated with 0.01 mg/ml proteinase K (Invitrogen) and 2% sarcosyl for 16 h at 50°C. The solution was then treated with DNAzol (Invitrogen) following the manufacturer's standard protocol, and purified NeabNPV DNA was resolved by 0.8% agarose pulse-field gel electrophoresis in 0.5× Tris-borate-EDTA buffer at 200 V with a switch time of 60 s for 15 h, followed by a switch time of 90 s for 8 h, at 14°C. NeabNPV DNA was stained with 0.5 μg/ml ethidium bromide, viewed under long-wavelength UV illumination (365 nm), and excised. Excised DNA was purified with the Fermentas DNA extraction kit (Fermentas) according to the manufacturer's protocol.

DNA cloning and sequencing.

NeabNPV DNA was fragmented by restriction enzyme digestion and by hydrodynamic shearing (HydroShear; GenMachine, San Carlos, CA). Partial restriction libraries were generated by digestion of NeabNPV DNA with HindIII (New England Biolabs) followed by ligation into pBluescriptII KS+ (Stratagene) or by digestion with EcoRI (New England Biolabs) followed by ligation into pT7/T3a18. Random libraries fragmented by hydrodynamic shearing were generated and ligated into pSMART (Lucigen Corporation). PCR was used to generate gap-spanning fragments. The result was an average of 15-fold sequence redundancy. DNA sequencing was conducted with an ABI3700 automated DNA sequencer (Applied Biosystems, Inc.), and the reactions were carried out with an ABI Prism BigDye Terminator Cycle Sequencing Ready Reaction kit (Perkin-Elmer Applied Biosystems).

Genome sequence analysis.

The complete NeabNPV genome was automatically assembled by PHRED/PHRAP/CONSED with a PHRAP repeat stringency of 0.8 (22, 23). The contiguous sequence was manually edited and confirmed. ORFs were identified with the NCBI ORF Finder program (http://ncbi.nih.nlm.gov/gorf/gorf.html), and those encoding more than 50 amino acids with minimal overlap were considered putative genes. Characterization of the ORFs involved BLAST analysis with BLASTP and PSI-BLAST against the NCBI nonredundant protein database (3), as well as RPS-BLAST against the CDD database (69). Baculovirus putative homologues were accepted on the basis of a BLASTP (default parameters) E value of ≤0.001. Predicted ORFs were then analyzed with InterProScan (104), which integrates queries against the PROSITE (45), PRINTS (5), Pfam (6), ProDom (19), SMART (88), TIGRFAMs (35), PIR superfamily (101), SUPERFAMILY (29), Gene3D (13), and PANTHER (70) databases.

Genome parity and repeat analysis.

The repeat regions were identified by REPuter (57). The 5′ and 3′ limits were defined as the outermost repeat with no other repeats within 1 kb on the external side. The minimal repeat element was queried against the NeleNPV and NeseNPV genomes with BLASTN 2.2.6. Repeat regions were also illustrated by generating parity plots of the NeabNPV genome sequence against itself with PipMaker (89). Additionally, genome parity between NeseNPV and NeabNPV and between NeleNPV and NeabNPV was illustrated with PipMaker (89).

Promoter sequence analysis.

Sequences 160 bp upstream of the predicted ORF start codons were analyzed for potential promoter motifs. These sequences were compared to experimentally verified promoter elements with SIGNAL SCAN v4.05 (82) to query the TRANSFAC database (99). Potential promoter elements were identified by aligning all upstream sequences with AlignACE v3.0 (48). The sequences were also analyzed for common known baculovirus promoter elements, including the TATA sequence (TATA, TATAW), CAKT, and DTAAG elements. These were also sought in the −40-bp to −20-bp region of the upstream sequences. Only putative predicted promoter elements with a P value of <0.05 were reported, except for TATA elements that were overrepresented in the −160-bp or −40-bp to −20-bp upstream sequences and a putative CAKT element, because of its established function as a baculovirus promoter element (84).

Calculation of P values.

To generate a P value for a sequence of a given length, 1,000 random loci of equal length (160 or 20 bp) were sampled. The expected frequency of occurrence was calculated for each motif in this data set. The P value for observing x motif-containing putative promoter elements out of N ORFs given an expected frequency F_e and a cumulative binomial distribution was calculated as follows: P(occ ≤ x) = [N!/x! × (N − x)!] × (F_e)^x × (1 − F_e)^{N − x}.

Nucleotide sequence accession number.

The NeabNPV genome sequence has been deposited in GenBank under accession number DQ317692.

RESULTS AND DISCUSSION

Nucleotide sequence analysis.

The NeabNPV genome was 84,264 bp in size, making it among the smallest baculovirus genomes, along with NeleNPV at 81,755 bp (62), NeseNPV at 86,462 bp (26), and Adoxophyes orana GV at 99,657 bp (100). The G+C content was equivalent to that of NeleNPV and NeseNPV at 33.5%. By comparison, the G+C contents of lepidopteran baculovirus genomes range from 32.4% for Cryptophlebia leucotreta GV (60) to 57.5% for Xestia c-nigrum GV (40).

The NeabNPV genome contained 93 potential methionine-initiated ORFs (Fig. 1; Table 1). This is greater than the number of ORFs predicted to be encoded by the NeleNPV (89 ORFs) or NeseNPV (90 ORFs) genome (26, 62). Similar to the genome of NeseNPV, the NeabNPV ORFs are biased in orientation such that 60.9% are oriented clockwise and 39.1% are oriented in the opposite direction.

FIG. 1.

Linear sequence map and HindIII physical map of the circular NeabNPV genome. The transcriptional direction of each ORF is represented by an arrow labeled with the ORF number. Known baculovirus predicted homologues are presented below the ORF numbers. Repeat regions (rep) are shown as triangles. Repeat regions depicted in Fig. 2 are outlined with dotted lines.

TABLE 1.

NeabNPVORFs

ORF	Name	Position	Length (amino acids)	Predicted sequence homologuea					Feature of coding sequenced	Putative promoter elementsc
				NeleNPV	NeseNPV	AcMNPV	PxGVb	CuniNPV
1	polyhedrin (polh)	0 > 741	247	1 (99) 247	1 (93) 246	8 (46) 245	4 (36) 248			TATA box, DTAAG, M7, M8, M9
2		906 < 1953	349	5 (84) 52	3 (14) 388				Cyclin-like F-box domain, RNA-binding motif	M1, M2
	rep1	1881 > 2755
3		2093 < 2273	60							M1, M2
4		2380 < 2551	57							TATA box, M1, M2, M4
5		2797 > 3469	224	13 (62) 247	20 (21) 270				Similar to AcMNPV P33, signal peptide	TATA box, Zeste, DTAAG, M1, M6, M7, M8, M9
6	dbp	3557 < 4268	237	14 (88) 236	22 (45) 251	25 (19) 316				TATA box, M2, M5
7	lef-11	4248 < 4557	103	15 (84) 101	23 (43) 100					TATA box, CAKT, M9
8	p33	4570 < 5341	257	16 (87) 256	24 (51) 252	92 (19) 259	76 (18) 250	14 (14) 370		DTAAG, M3, M6
9		5342 > 5855	171	17 (91) 170	25 (47) 170					TATA box, DTAAG
10		5872 > 6523	217	18 (77) 216	26 (44) 218				1 predicted TM domain, signal peptide	TATA box, DTAAG, M7
	rep2	6570 > 6954
11		6635 > 6809	58						2 predicted TM domains	TATA box, M3
12	dnapol	6795 < 9555	920	20 (84) 923	28 (64) 913	65 (23) 984	94 (23) 979	91 (17) 1138	DNA polymerase type B family	Zeste, CREB/ATF-47, CAKT, M6
13		9553 > 11638	695	21 (66) 728	29 (28) 792		93 (9) 651	67 (13) 1357	Putative ATPase, putative IMP dehydrogenase domain; prefoldin structure	CREB/ATF-47, M3, M6, M7
14		11742 > 12963	407	22 (63) 380	3 (21) 388
15		13175 > 13937	254	25 (35) 261					Leucine-rich repeat	TATA box, DTAAG, M3
16	odv-c56	14060 > 15071	337	23 (92) 336	38 (71) 339	148 (34) 376	16 (38) 351	101 (20) 361	1 predicted TM domain, IMP dehydrogenase/GMP reductase domain	TATA box, DTAAG, M5, M7, M9
17		15159 > 15738	193	24 (92) 192	39 (50) 192					TATA box, DTAAG, M3
18		15870 < 16698	276	25 (52) 261						M3, M4, M9
19		16912 < 17596	228	26 (67) 226						TATA box, Zeste, M1, M6, M9
20		17757 < 18267	170	2 (34) 108						TATA box
	rep3	17804 > 18262
21	iap	18332 > 18941	203	11 (75) 260	17 (24) 181	27 (23) 286	98 (21) 281		Baculovirus IAP repeat domain	TATA box, Zeste, M1, M7
22		18912 < 19074	54							CAKT, M7
23		19610 < 22079	823	9 (60) 794	16 (27) 637					TATA box, M3
24		22749 > 24213	488	8 (50) 526	11 (16) 461				Ring finger domain, zinc finger, LMP repeat	TATA box, M3
25		24466 > 24868	134	7 (71) 131						TATA box, M3, M6, M7
26	trypsin-like protein	25091 > 25871	260	6 (96) 259	7 (73) 258				Trypsin-like serine protease	DTAAG, M3, M6
27		25989 < 27069	360	4 (73) 332	9 (20) 416					TATA box, Zeste, M4
	rep4	26312 > 26527
28		27373 < 27559	62						Signal peptide	Zeste, CAKT, M1, M6, M9
	rep5	27427 > 29532
29		28291 > 29335	348	5 (84) 52	3 (14) 388				F-box domain	TATA box, CREB/ATF-47
30		29972 > 30539	189	27 (51) 190	37 (29) 133					TATA box, M1, M2
31		30612 < 30869	86							TATA box, CAKT, M5
32	p40	30950 < 32051	367	29 (91) 366	35 (64) 368					TATA box, Zeste
33		32105 < 32423	106	30 (75) 116	34 (47) 117					TATA box, DTAAG, M7
34	p48	32419 < 33628	403	31 (85) 390	33 (54) 402	103 (10) 387			IMP dehydrogenase/GMP reductase domain	TATA box, M6
35		33646 < 34348	234	32 (88) 238	32 (63) 235		40 (17) 206			Zeste, M6
36	alk-exo	34451 > 35663	404	33 (82) 402	31 (48) 399	133 (22) 419	106 (25) 378	54 (15) 367	IAP repeat restriction endonuclease like
37		35659 > 36133	158	34 (84) 157		152 (14) 408			1 predicted TM domain, signal peptide	TATA box, CAKT, DTAAG, M6, M9
38		36417 < 37371	318	36 (60) 319					1 predicted TM domain, signal peptide	TATA box, M6
39	lef-9	37363 < 38953	530	37 (87) 503	40 (62) 507	62 (29) 516	99 (31) 494	59 (15) 590		TATA box, M3, M7, M9
40		38981 < 39404	141	38 (87) 141	41 (48) 147	68 (20) 192	96 (15) 128	58 (18) 137	1 predicted TM domain	TATA box, M7
41		39533 < 39722	73							TATA box, DTAAG, M7, M8, M9
42		39720 > 40707	329	39 (70) 353	42 (24) 444					Zeste, M3, M7
43		40703 < 41081	126	40 (87) 125	43 (39) 125					TATA box, DTAAG, M7
44		41077 < 41308	74	41 (90) 76	44 (63) 82				1 predicted TM domain, signal peptide	CREB/ATF-47, M6
45	vlf-1	41310 < 42375	355	42 (86) 354	45 (65) 353	77 (24) 379	89 (29) 346	18 (16) 358	DNA breaking-rejoining enzyme, catalytic core, IMP dehydrogenase/GMP reductase domain	TATA box, CAKT, DTAAG
46		42458 > 42617	53						1 predicted TM domain, signal peptide	TATA box
47	gp41	42649 < 43315	222	44 (94) 270	47 (60) 312	80 (30) 409	87 (19) 283			TATA box, DTAAG, M4, M8
48		43484 > 44012	176	45 (90) 175	48 (75) 176	81 (36) 233	86 (34) 191	106 (14) 186	2 predicted TM domains	Zeste, DTAAG, M3
49	p47	44255 < 45434	393	46 (87) 389	49 (61) 386	40 (22) 401	51 (20) 386	73 (10) 439		Zeste, CAKT, M7
50	p74	45435 < 47340	635	47 (86) 633	50 (63) 634	138 (38) 645	49 (31) 578	74 (32) 681	3 predicted TM domains, IMP dehydrogenase/GMP reductase domain	TATA box, M3
51		47317 < 47854	179	48 (69) 178	51 (28) 142				Double-stranded RNA-binding motif	TATA box, M8, M9
52		47916 > 48585	223	49 (76) 218	52 (40) 220				Zinc finger	TATA box, Zeste
53		48577 > 49933	452		53 (48) 386				1 predicted TM domain IMP dehydrogenase/GMP reductase domain	TATA box, M4, M5, M6
54		49929 > 50508	193	51 (83) 192	54 (47) 184					TATA box, CREB/ATF-47, DTAAG, M6, M9
55	pif-2	50537 > 51692	385	52 (93) 384	55 (73) 384	22 (43) 382	37 (46) 368	38 (43) 403	1 predicted TM domain, signal peptide	TATA box, DTAAG, M3, M6, M7
56		51761 > 52094	111	53 (88) 104	56 (57) 115					TATA box, DTAAG, M7
57	lef-2	52095 > 52683	196	54 (80) 195	57 (37) 200	6 (16) 210		25 (11) 225		TATA box, M4
58	lef-5	52709 < 53417	236	55 (88) 235	58 (57) 230	99 (27) 265	69 (31) 247		Zinc finger	TATA box, M9
59	38K	53407 > 54325	306	56 (88) 305	59 (59) 304	98 (31) 320	70 (31) 340	87 (25) 303	IMP dehydrogenase/GMP reductase domain, putative phosphatase	Zeste, DTAAG
60		54310 < 54817	169	57 (92) 168	60 (63) 170	96 (25) 173	71 (24) 161	90 (25) 202	1 predicted TM domain, signal peptide	TATA box, M5
61	helicase	54803 > 58214	1,137	58 (89) 1134	61 (59) 1143	95 (18) 1221	72 (18) 1124	89 (12) 1332	IMP dehydrogenase/GMP reductase domain, P-loop containing nucleoside triphosphate hydrolases	M7
62	lef-4	58210 < 59617	469	59 (85) 468	62 (52) 477	90 (22) 464	78 (25) 432	96 (16) 497		TATA box, M4, M6
63		59633 > 60962	443	60 (89) 442	63 (52) 441	142 (20) 477				TATA box, DTAAG, M3, M6
64		60970 > 61210	80	61 (89) 78	64 (43) 78				1 predicted TM domain	TATA box, DTAAG, M4, M5
65	odv-e18	61285 > 61489	68	62 (95) 85	65 (83) 83				1 predicted TM domain, signal peptide, aspartate and glutamate racemases signature 1	TATA box, DTAAG
66	odv-e27	61503 > 62292	263	63 (94) 262	66 (71) 260	144 (21) 290	80 (12) 287			CAKT, DTAAG
67		62319 > 62652	111	64 (87) 110	67 (70) 108	145 (20) 77	12 (27) 98		1 predicted TM domain, signal peptide, putative chitin-binding motif	TATA box, DTAAG
68	lef-1	62659 < 63295	212	65 (85) 211	68 (54) 190	14 (20) 266	55 (27) 251	45 (26) 235	DNA primase	TATA box, CREB/ATF-47, M5
69		63352 > 63940	196	66 (89) 193	69 (63) 190	115 (24) 204	29 (28) 181	46 (27) 203	1 predicted TM domain, signal peptide	TATA box, Zeste, M9
70		63939 > 64119	60						1 predicted TM domain, signal peptide	TATA box, Zeste, DTAAG, M6, M7
71		64093 > 65158	355	67 (86) 354	70 (57) 357	109 (21) 390	43 (20) 414	69 (17) 409		TATA box, M3, M7, M9
72		65157 > 65367	70	68 (89) 69	71 (53) 71				1 predicted TM domain	TATA box, DTAAG
73	rcc1	65353 > 65851	166	69 (80) 136	72 (60) 118				RCC1 domain, 1 predicted TM domain	TATA box, M3, M5, M6
74	rcc1	66214 > 66565	117	71 (71) 99	72 (7) 118				RCC1 domain, 1 predicted TM domain, signal peptide	TATA box, M7
75		66996 > 67545	183	72 (64) 283	75 (39) 273					TATA box, M7
76		67599 > 67755	52							TATA box
77		67787 > 68585	266	74 (81) 265	77 (36) 284					TATA box, M4, M7
78		68962 > 69592	210	75 (79) 209	78 (27) 143					TATA box, CREB/ATF-47, M9
79	pif	69573 < 71187	538	76 (91) 530	79 (63) 519	119 (29) 530	7 (28) 536	29 (30) 523	1 predicted TM domain, signal peptide, zinc finger	Zeste, CREB/ATF-47, DTAAG, M1, M6, M9
	rep6	71014 > 72092
80		71218 > 71485	89	35 (25) 97						TATA box, CAKT, M7
81		71728 < 71899	57	2 (51) 108						TATA box, M3, M6
82		71923 < 72379	152	77 (70) 149	80 (45) 156			28 (13) 284		TATA box, CREB/ATF-47
83	lef-8	72377 > 74909	844	78 (87) 843	81 (69) 846	50 (30) 876	109 (32) 838	26 (20) 922	Subunits of DNA dependent RNA polymerase	TATA box, Zeste, CAKT, M5
	rep7	74887 > 75140
84		74926 > 75151	75	2 (28) 108						TATA box
85		75391 < 75556	55							TATA box, DTAAG, M4, M8
86		75878 > 76430	184	81 (89) 148	83 (66) 145				Capsid structural protein	TATA box, M1, M5, M8
87	vp91	76438 > 78844	802	82 (82) 803	84 (51) 824	83 (23) 847	84 (21) 533	35 (22) 741	1 predicted TM domain, signal peptide, IMP dehydrogenase/GMP reductase domain, chitin-binding peritrophin A domain	TATA box, Zeste, DTAAG, M4, M7, M9
88	vp1054	78845 < 79787	314	83 (93) 313	85 (60) 310	54 (20) 365	115 (16) 311	8 (13) 329	IMP dehydrogenase/GMP reductase domain, DNA polymerase beta, N terminals like
89		79912 > 80125	71	84 (81) 69	86 (58) 73
90		80132 > 81056	308	85 (83) 305	87 (38) 311					TATA box, DTAAG, M6
91		81100 > 82480	460	87 (54) 476	18 (11) 301					TATA box, M7
92	vp39	82640 < 83588	316	88 (98) 315	89 (71) 312	89 (18) 347	79 (14) 320	24 (10) 289	Putative RNA 2′-phosphotransferase	CAKT, M2, M5, M6
93		83597 > 84140	181	89 (80) 183	90 (43) 192				Putative RNA 2′-phosphotransferase	TATA box, Zeste, DTAAG

^aThe ORF number of a putative homologue is shown with the percent amino acid identity in parentheses, followed by the length of the homologue.

^bPxGV: Platella xylostella GV.

^cPromoter elements and abbreviations are defined in Table 2. A TATA box was defined as a TATA sequence within the −40 to −20-bp upstream region of the ORF or a TATAW sequence within 160 bp upstream of the ORF. CAKT was reported only when it occurred within the −40 to −20-bp upstream region of the ORF. All other elements were reported if they occurred within 160 bp upstream of the ORF.

^dSummary of BLAST, PSI-BLAST, RPS-BLAST, and INTERPRO SignalScan analyses.

NeabNPV gene content.

Twenty-nine genes are shared among all of the baculovirus genomes sequenced to date (26, 62). Putative homologues of all 29 of these conserved genes are found in NeabNPV. Of the 93 predicted NeabNPV ORFs, 72 have corresponding homologues in NeseNPV and 81 have corresponding homologues in NeleNPV. Eleven NeabNPV predicted ORFs are unique to NeabNPV (neab3, neab4, neab11, neab22, neab28, neab31, neab41, neab46, neab70, neab76, and neab85). Ten NeabNPV predicted ORFs (neab15, neab18, neab19, neab20, neab25, neab37, neab38, neab80, neab81, and neab84) have a putative sequence homologue in the NeleNPV genome but not in the NeseNPV genome, while one NeabNPV predicted ORF (neab53) has a putative sequence homologue in the NeseNPV genome but not in the NeleNPV genome.

Potential envelope fusion proteins in sawfly baculoviruses.

In Autographa californica multiple NPV (AcMNPV), the GP64 protein appears to be a critical protein for cell-to-cell transmission of the baculovirus budded virus, as gp64 deletion mutants were localized to insect midguts (10). In the Lymantria dispar NPV genome, where a gp64 homologue was not identified (58), the gene encoding a potential gp64 analog, ld130, was proposed and its Spodoptera exigua NPV sequence homologue was later shown to mediate pH-dependent membrane fusion between cells (50). An orthologue for gp64 was not identified in NeabNPV, NeleNPV, or NeseNPV (26, 62) by BLAST analysis. Sequence homology searches, however, may not be the most effective means of identifying analogs for envelope fusion proteins, as L. dispar NPV Ld130 does not have significant sequence similarity to AcMNPV GP64. According to the original criteria on the basis of which Ld130 was proposed as a potential envelope fusion protein (58), candidate protein analogs should possess both an N-terminal signal and a transmembrane (TM) domain.

Two NeabNPV ORFs, neab10 and neab44, possess these characteristics and may be candidate envelope fusion genes. InterProScan analysis (see Materials and Methods) of the predicted gene products indicated the presence of both an N-terminal signal and a TM domain. The putative sequence homologues of these two gene products in NeleNPV and NeseNPV also encode a potential N-terminal signal and a potential TM domain (26, 62). Alternatively, the sawfly baculoviruses may not encode an envelope fusion protein or produce a budded virus since for almost all known nonlepidopteran NPVs virus replication occurs only in the midgut epithelium (24).

NeabNPV repeats.

Two repeat regions (rep2 and rep4) each contain a cluster of localized repeat elements that are not interspersed throughout the genome (Fig. 2A and B). The remaining repeat clusters (rep1, rep3, rep5, rep6, and rep7), however, have a repetitive element in common with the consensus sequence 5′-CAACTTGTCAAATGTGTTGGACCTCGAGCCCAACAAACGCGACATCT-3′ that is interspersed at multiple loci throughout the genome (Fig. 2C and 3A). This core sequence is present in NeleNPV direct repeat 5 but does not occur in the NeseNPV genome. The loci known as homologous regions, hrs, of the lepidopteran-infecting NPVs have been found to function as enhancers of early gene transcription (30, 33) and origins of DNA replication (56, 80). The hrs of the lepidopteran-infecting NPVs have also been implicated as sites of DNA recombination (49). In NeabNPV, the location of this repeat element appears to coincide with major rearrangement events between the NeleNPV and NeabNPV genomes (Fig. 3B).

FIG. 2.

Loci of the NeabNPV genome with repeating elements. The triangles represent the repeated core element in the direction indicated by the point. The name and genome location of the repeat region are indicated on the right, and the repeat element consensus sequences are as follows: A, ATAAAAAACAGTAAATATT(C/T)CAATACGATGC AAACGCACGTGATTAATGT; B, TCAG(A/C)ATTGTCGTTGTTGTTTT(G/C)AG TGTTTTCTGT(G/A)TTATTTC; C, AGATGTCGCGTTTGTTGGGCTCGAGGT CCAACACATTTGACAAGTTG. White arrows indicate the orientation and location of NeabNPV ORFs depicted in Fig. 4.

FIG. 3.

Genome parity plots. Genome plots were derived by comparing (A) NeabNPV against itself, (B) NeabNPV against NeleNPV, (C) NeabNPV against NeseNPV, and (D) NeleNPV against NeseNPV. The axes are labeled with the genome name and the distance (in base pairs). The interspersed repeat sites of NeabNPV are labeled on plots A and B. Triangles above plots A and B represent the locations of interspersed repeats 1 to 7.

Repeated genes.

Lepidopteran baculovirus genomes, except Spodoptera exigua NPV (51), Plutella xylostella GV (39), and A. orana GV (100), commonly encode members of a family of repeated genes known as bro genes. These genes are typically repeated elements of the baculovirus genomes, and it has been proposed that bro gene products are involved in DNA replication and transcription (105), as well as in maximizing polyhedron formation in specific host species (8). Outside of the family Baculoviridae, bro and bro-like genes have been found in ascoviruses, insect iridoviruses, entomopoxviruses (ALI gene family), the phycodnavirus ectocarpus siliculosus virus, bacteriophages, and prophages integrated into bacterial genomes (2, 7, 8, 53). Despite the widespread occurrence of bro genes, no homologues of these genes were found in NeabNPV or the other sawfly baculoviruses. Five ORFs unique to NeabNPV and NeleNPV (neab3, neab4, neab80, neab81, and neab84), however, have in common an amino acid motif duplicated in two NeleNPV ORFs (Fig. 4) and are proximal to interspersed genome repeats (rep1 and rep7). These repeated amino acid motifs are shared in two genes in NeleNPV but were not previously reported (62). Duplicated genes flanked by repeated regions were reported in NeseNPV (26), but they do not share the amino acid motifs of the NeabNPV and NeleNPV repeated genes.

FIG. 4.

Alignment of conserved motifs repeated in multiple predicted ORFs of the NeabNPV and NeleNPV genomes. Gaps in the alignment are indicated by dashes, and conserved amino acids are outlined.

The G+C composition of the putative NeabNPV ORFs within the repeat regions contrasts significantly with the genome average. While the predicted coding sequences of the NeabNPV genome had a mean G+C content of 34% ± 4%, eight putative ORFs had a G+C content of greater than 40%, including four ORFs within the repeated segments, neab3 (45% G+C), neab4 (47% G+C), neab80 (44% G+C), and neab81 (46% G+C). In bacteria, the G+C content is homogeneous throughout genomes and heterogeneous between genomes (73). Furthermore, genome G+C content is related to phylogeny (78) and has been used to predict genes that have been acquired by horizontal gene transfer (27, 28, 74). As a result, G+C content is often used as a genome signature and can be used to compare baculovirus genomes. Given that the base composition of the repeat region sequences contrasts so greatly with the average of the sawfly baculovirus genomes and that the repeat regions are found in NeabNPV and NeleNPV but not in NeseNPV, it is possible that a common ancestor of NeabNPV and NeleNPV acquired those elements by horizontal transfer following its divergence from NeseNPV.

NeabNPV evolution.

Sawfly baculoviruses represent a distinct taxon of the virus family Baculoviridae (26, 62). Nucleotide sequence plots revealed greater extended colinearity between NeabNPV and NeleNPV than between NeabNPV and NeseNPV (Fig. 3B and C). Furthermore, NeabNPV and NeleNPV share the same conserved loci compared to NeseNPV (Fig. 3C and D). Although all three host species originally inhabited North America, N. sertifer is believed to have migrated via the Bering Land Bridge to Eurasia. Following this migration, N. sertifer became extinct in the New World only to be reintroduced in ca. 1925 (67). This geographic isolation of host species is supported by the viral DNA sequences and is consistent with the divergence of NeseNPV from NeabNPV and NeleNPV.

The parity plots (Fig. 3) illustrate that the region of the genome from 37 to 84 kb (neab41 to neab93), relative to the polyhedrin gene, holds more colinearity than the preceding region. Among the first 40 ORFs in NeabNPV, only 67.5% have predicted sequence homologues in NeseNPV. Among ORFs 41 to 93, however, 87% have predicted sequence homologues. Similarly, among the first 40 ORFs, only 6 (0.15%) are conserved in all of the sequenced baculovirus genomes, whereas among ORFs 41 to 93, 23 (43.4%) are conserved in all of the sequenced baculovirus genomes.

Genome recombination is a mechanism for host range expansion in baculoviruses (55). Arends and Jehle (4) noted that inversions of genome loci are often flanked by hrs in the lepidopteran baculoviruses. The repeat element in the NeabNPV and NeleNPV genomes appears to be analogous to the hrs in that its expansion and interspersal may promote recombination between repeats. Given the role of genome recombination in host range expansion, the repeat sequences may have contributed significantly to the evolution of NeabNPV and NeleNPV following their ancestral divergence from NeseNPV.

Promoter analysis.

Gene expression of lepidopteran-infecting baculoviruses is temporally regulated, with four distinct stages, immediate-early, early, late, and very late. Immediate-early and delayed-early genes are expressed before DNA replication and are transcribed by insect host RNA polymerase II (25, 46). These early genes may encode promoter elements such as the TATA element that acts as an assembly site for the host RNA polymerase II transcription complex (32, 83) and the CAKT sequence element that enhances gene expression and serves as a transcription start site in early-expressed baculovirus genes (9, 83).

Putative TATA elements were identified upstream of 69 NeabNPV ORFs, and CAKT elements were identified in the −40- to −20-bp region upstream of 12 ORFs. On the basis of the lepidopteran baculovirus model, ORFs with both elements should represent early-expressed genes. Seven ORFs with putative TATA box motifs also have putative CAKT motifs. Among these are suspected homologues of the viral RNA polymerase subunit gene lef-8 (31), lef-11, and vlf-1 and four uncharacterized ORFs. CAKT was also identified in six predicted TATA-less promoters, including the early-expressed dnapol (16, 47, 65, 93) and p47 (15, 61) genes, as well as the late-expressed odv-e27 (12) and vp39 genes (11, 41, 92), and two uncharacterized ORFs. The occurrence of TATA elements was significantly more frequent in the regions upstream of the predicted ORFs than in the whole genome, suggesting that they are general transcription motifs. The CAKT sequence was not significantly overrepresented either 160 bp upstream or in the −40- to −20-bp upstream regions, relative to the predicted translation start sites of NeabNPV ORFs. This may be due to the fact that the early genes would represent a small subset of viral genes and this element would not be a general transcription element. Alternatively, the absence of lepidopteran baculovirus immediate-early gene homologues in NeabNPV might suggest that sawfly baculoviruses utilize an alternative mechanism for early gene regulation.

Late gene expression in lepidopteran baculoviruses is associated with a DTAAG sequence motif (72). The DTAAG promoter element acts as the initiation site for transcription mediated by the virus-encoded RNA polymerase, and the strength of expression from this site depends on the context of the sequence between it and the translational initiation site (68, 77, 82, 83, 97). The DTAAG motif was a common element in NeabNPV and was overrepresented in the regions 160 bp upstream of the predicted ORF translation initiation sites.

Among the predicted NeabNPV ORFs that have well-characterized putative sequence homologues in other baculoviruses, there is little evidence that the presence of these three promoter elements is indicative of the temporal expression of the genes. From a genome analysis perspective, the DTAAG sequence occurred in NeabNPV early genes as often as in late genes. However, it is likely that this site acts as an alternate start site for later transcripts. Whether the viral RNA polymerase of sawfly baculoviruses utilizes a DTAAG motif sequence as a transcription start site remains to be experimentally verified.

In order to explore alternative approaches to baculovirus promoter analysis, we examined two features. The first approach involved determining whether there was a general overrepresentation of host transcription factor binding sites upstream of NeabNPV ORFs and whether the occurrence of these binding sites is biased toward genes within a similar temporal group. We hypothesized that early baculovirus promoters would rely on host transcription factors to a greater degree than later genes, as they utilize the host transcription mechanism.

The second involved using AlignACE (48) to perform multiple alignments to identify conserved motifs. This approach successfully identified core promoter elements in Drosophila (75), and we hypothesized that it would identify overrepresented conserved promoter motifs in the NeabNPV genome. In particular, we were interested in whether the multiple-alignment method would identify core promoter elements likely to be employed by the insect host cell transcriptional apparatus or common baculovirus elements.

To address the first question regarding the utilization of known host transcription factor binding sites, the upstream sequences were queried against the TRANSFAC database. Three experimentally confirmed insect promoter elements were overrepresented in the upstream sequences of predicted NeabNPV ORFs. One was the TATA box, a core promoter element common in insect and baculovirus genomes. Another was the motif TGAGHB, which corresponds to the DNA binding motif for the Zeste protein.

Zeste is a transcription factor that facilitates transvection, the mechanism by which a gene can trans regulate the expression of a homologue (21, 54, 102). It has been observed that AcMNPV hr1 can act as an enhancer in trans in mammalian cells (95). Although the hr1 region did not permit transvection of genes encoded by host insect genomes, it is possible that hymenopteran baculovirus genes could utilize Zeste-mediated trans regulation. The zeste promoter motif has been identified upstream of seven genes with a bias toward early-expressed genes, including dnapol (16, 47, 65, 93), iap-3 (14, 52), p40 (66), p47 (15, 61), and 38K(pp31) (91). Two genes with this sequence upstream, pif and vp91, are expressed late in lepidopteran NPVs (34, 87, 90). zeste promoter motifs were identified in the upstream sequences of a further 13 predicted NeabNPV ORFs.

A CREB/ATF47 binding site (known as the cis-acting replication element) mediates transcription in response to members of the CREB/ATF protein family. This is a large family of leucine zipper transcription factors (38) that, despite their diverse activities, have in common the abilities to respond to environmental signals and maintain cellular homeostasis. Some ATF proteins (ATF2, ATF3, and ATF6) play a role in mediating stress response (37), CREB and ATF1 regulate transcription in response to intracellular cyclic AMP concentrations (36), and ATF4 acts as a negative regulator of cis-acting replication element-dependent transcription (20). ATF proteins are important elements in regulating early transcription of members of the family Adenoviridae. The adenovirus E1A protein interacts with host-encoded ATF-2 to activate transcription of several adenovirus early genes (63). In the NeabNPV genome, the CREB/ATF binding sites were observed upstream of both the putative early-expressed genes dnapol (16, 47, 65, 93) and lef-1 (79), as well as the putatively late-expressed gene pif (34, 90). The remaining six ORFs with upstream CREB/ATF binding sites had no characterized homologues.

By the second approach, alignment of upstream sequences identified nine sequences that we designated M1 to M9 (Table 1 and Table 2). Contrary to our prediction, promoter alignment did not report the core Drosophila promoter elements or the known baculovirus promoter elements. Within these overrepresented conserved motifs, however, certain spatial and temporal patterns were observed.

TABLE 2.

Predicted putative promoter elements for NeabNPV

^aTranscription factors derived from SignalScan analysis of the eukaryote promoter database (see Materials and Methods). Only those that occur significantly more often in the upstream sequence than in the whole genome were reported.

^bAlignAce analysis of upstream sequences identified a number of putative common elements. Only those that occur significantly more often in the upstream sequence than in the whole genome were reported.

^cThe P value represents a statistical test that the element occurs more often −160 bp upstream of predicted ORFs than in the whole NeabNPV genome. The value in parentheses represents a statistical test that the element occurs more often in the −40 to −20-bp region upstream of predicted ORFs than in the whole NeabNPV genome.

The M1 motif was not overrepresented in genes known to be temporally distinct but rather was clustered spatially. Of the 10 occurrences of this motif, 8 are within a 30-kb region (747 to 29,973 bp) within the more variable half of the circular NeabNPV genome. Similarly, five of six ORFs with the upstream M2 motif were within an approximately 5-kb region (bp 82,641 to 84,264 and bp 0 to 3,558). The M8 motif was upstream of the late-expressed polh (17, 18, 86, 94) and gp41 (64, 98) genes, as well as five other putative ORFs. Five of the seven ORFs with an upstream M8 motif also possess putative TATA and DTAAG motifs. Motifs M2 and M8 occurred in only six and seven upstream sequences, respectively, but these still occur there significantly more often than in the whole genome. Motif M3 occurred upstream of 10 predicted ORFs, including p33, trypsin-like, lef-9, p74, pif-2, and rcc1. There is no available temporal characterization for all of these genes except lef-9 and p74, which are both expressed late (1, 31, 59, 85). The other motifs (M4, M5, M6, and M7) did not appear to have characteristic clustering or temporal properties.

Our hypothesis that host transcription factor binding sites might be enriched in the promoters of early-expressed baculovirus genes appears to hold for the Zeste protein binding site but not for the CREB/ATF47 binding site. Conversely, motifs characterized by promoter alignment with AlignACE did not correspond to temporally grouped genes; however, the spatial clustering of genes with these motifs is intriguing. These patterns and hypotheses require experimental confirmation and emphasize the need for more sophisticated analyses of baculovirus promoters.

Concluding remarks.

Although this study represents the third sawfly baculovirus genome to be completely sequenced, our analysis conforms to previous observations that the sawfly baculoviruses represent a distinct taxon (26, 62); our study suggests a possible mechanism for sawfly baculovirus evolution through genome rearrangement events between interspersed genome repeats. Rearrangement between these repeats may be similar to the genome arrangements described between the repeated lepidopteran hrs. Although not as well conserved as the lepidopteran baculovirus bro genes, predicted genes encoded within the NeabNPV repeats have an amino acid motif in common.

Our comparison of gene content and genome rearrangement agrees with the hypothesis proposed by Herniou et al. (42) that baculoviruses coevolve with their host. We observed that the geographically isolated Old World baculovirus NeseNPV has diverged more from the New World sawfly baculoviruses NeabNPV and NeleNPV.

The intriguing absence of immediate-early gene homologues and membrane fusion proteins in NeabNPV is consistent with previous studies of nonlepidopteran baculoviruses (2, 26, 62) and raises interesting questions regarding how early gene expression is mediated and how the sawfly baculovirus infection process differs from that of the lepidopteran baculoviruses. These questions, along with the renewed interest in sawfly baculoviruses in pest management, provide us with an opportunity to examine the unique and exciting characteristics of the baculoviruses outside the order Lepidoptera.