JOURNAI OF VIROLOGY, June 1992, p. 3476-3484 0022-538X/92/063476-09$02.00/0 Copyright ©D 1992, American Society for Microbiology

Vol. 66, No. 6

In Vitro Transactivation of Baculovirus Early Genes by Nuclear Extracts from Autographa californica Nuclear Polyhedrosis Virus-Infected Spodoptera frugiperda Cellst BARBARA GLOCKER, RICHARD R. HOOPES, JR., AND GEORGE F. ROHRMANN* Department of Agricultural Chemistry, Oregon State University, Corvallis, Oregon 97331-6502 Received 16 December 1991/Accepted 16 February 1992

Nuclear extracts, prepared from Autographa californica nuclear polyhedrosis virus-infected Spodoptera frugiperda cells during a time course of infection, were analyzed for activation of early gene transcription and for late gene transcription. The templates used in the in vitro transcription assays contained promoters for baculovirus genes that have been classified as immediate early, delayed early, and late. The promoters were derived from the baculovirus 39K, p26, gp64, and DNA polymerase genes. In addition, the adenovirus major late promoter was included in these studies. We found that transcription from promoters classified as immediate early or delayed early was accurately initiated by using extracts from uninfected cells. Furthermore, transcription from all early promoters tested was found to be transactivated by nuclear extracts prepared at 4 and 8 h postinfection. However, baculovirus enhancer-dependent transcriptional activation was not observed in tests with templates containing the hr5 enhancer sequence. Transcription from baculovirus late promoters was also not observed. A decline in transcription by nuclear extracts prepared from cells late in infection was associated with the presence of DNase activity. The Baculoviridae are a family of insect viruses characterized by a complex replication cycle that culminates in the occlusion of virions in a crystalline protein matrix (3). The Autographa califonica multicapsid nuclear polyhedrosis virus (AcMNPV) is the most well-characterized baculovirus and has a double-stranded, circular, supercoiled DNA genome of approximately 128 kbp. Progression through the AcMNPV infection cycle is governed by a cascade of early, late, and very late gene transcription (7). Early transcription begins before the initiation of replication of the viral genome and is inhibited by o-amanitin, consistent with it being RNA polymerase II dependent. Investigations with reporter gene constructs transfected into insect cells suggest that the level of transcription from early gene promoters is modulated by viral transactivating factors (4, 5, 10, 23) and enhancer sequences (12, 18, 22). After initiation of viral DNA replication, late gene expression is initiated and the transcription of some host nuclear genes is repressed (19). Very late in infection, two genes involved in occlusion body formation (polyhedrin and plO) are hyperexpressed (for a review, see reference 3). The transcription of late genes is dependent on the presence of an a-amanitin-resistant RNA polymerase (8, 9, 16) having a unique subunit composition that is differcnt from those of the three host RNA polymerases (26). Although one of the most intriguing aspects of baculovirus replication concerns the control of the viral transcription cascade, few of the host and viral factors participating in the regulation of baculovirus gene expression have been identified. The development of in vitro systems that reflect in vivo gene transcription is crucial for the identification and analysis of these factors. We have recently reported that baculovirus genes classified as immediate early are accurately initiated by nuclear extracts from uninfected insect (Spodoptera frugiperda [Sf9]) and human (Namalwa) cells

(15). In these studies, we concluded that host cell RNA polymerase II and its associated factors were the only components required for basal transcription from baculovirus immediate-early promoters. Wc have now cxtended these initial studies to cxamine in vitro transcription, using nuclear extracts prepared from both uninfected and AcM NPV-infected Sf9 cells. Uninfected-cell nuclear extracts were tested for the ability to support the transcription of genes categorized as immediate early or delayed early, whereas nuclear extracts from infected cells were examined for the transactivation of early gene transcription and for latc promoter-dependent transcription. We also tested for baculovirus hrS enhancer sequence-dependent transcription in vitro. Finally, we compared the relative levels of in vivo and in vitro transcription from a specific promoter at various times after infection. MATERIALS AND METHODS Construction of templates. The templates constructed for this study are described below and are shown in Fig. 1. These templates were derived from the genomcs of AcM

NPV, Orgyia pseludotsutgata multicapsid nuclcar polyhcdrosis virus (OpMNPV), and adenovirus type 2. The sequence of the promoter and mRNA initiation regions of these genes is shown in Fig. 5. (i) AcMNPV 39K gene promoter (p39K). Plasmid p39K contains a 1-kb PstI-SstI fragment from the AcMNPV PstI K fragment (10) cloned into the corresponding restriction sitcs of pBlucscript (pKS-; Stratagcnc Cloning Systcms). (ii) 39K gene promoter plus enhancer hr5 (pHR39K). A constructcd by digcstion plasmid, pBG8, containing hi-5 of the AcMNPV HinidIlI 0 fragmcnt (17) with MluI. The 484-bp fragment containing lit-5 was isolated, the ends wcrc filled with Klcnow polymerase, and the fragmcnt cloncd into the StinaI site of pBluescribe (pBS; Strataigene). To characterize the lr-5 insert, it wals sequcnecd and found to have single nucleotide difference (nuclecotide Intl -525, T was

was

Corresponding author. t Technical report 98t)7 from the Oregon State University Agricultural Experiment Station. *

a

3476

VOL. 66, 1992

BACULOVIRUS EARLY GENE TRANSACTIVATION

A) p39k

214nt -

Ad

ssi

214 nt X >

hr5

pHR39k

i

Kpn.I

HixtII

316 nt B) p64CAT-319

EcoRI

_

EwRI 5%4 nt

C) pAML

143 nt

r

D) pOpp26

-EZ

HindIII

Safl

Tr5 E) pAcHRp26

378nt

-s,gn

Hlnd!!!

W

218 nt F)

pDNAP Ec

I

xioI

200 bp

FIG. 1. Schematic representation of the pilasmids used for in vitro transcription. The arrows indicate the ru noff transcripts produced from the RNA start site when the templalte is truncated by the indicated restriction enzyme. The numbers indi cate predicted runoff sizes in nucleotides (nt). The derivation of tktese plasmids is described in Materials and Methods.

replaced with C) from the hr5 sequence rep)orted by Liu et al. (17). The kIpnl-HindIII fragment of plasnaid pBG8 containing the hr5 enhancer region was cloned inlto the corresponding restriction sites of plasmid p39K to cr*eate pHR39K. (iii) AcMNPV p26 gene promoter with hir5 (pAcHRp26). A plasmid containing the AcAfNPV p26 prormoter and hr5 was constructed by cloning a 1.5-kb HindIII[-SstI fragment of AcMNPV HindIII-Q (17) into pKS-. Pla smid pAcHRp26A contains the 1.3-kb HindIII-SalI fragmen t (Fig. 1E) with a smaller portion of the p26 open readin, g frame and was constructed for use in primer extension a]nalysis. (iv) AcMNPV DNA gene polymerase prromoter (pDNAP). An 800-bp AcMNPV EcoRI V fragmet nt containing the promoter region of the DNA polymera.se gene (24) was subcloned in pKS-. (v) OpMNPV gp64 gene promoter (pl64CAT-319). This plasmid contains 319 bp of the 5' untrans lated sequence of gp64 cloned into pBS- upstream of a chloramphenicol acetyltransferase (CAT) gene and is descri bed elsewhere (4). (vi) OpMNPV p26 gene promoter (pOpp,26). This plasmid contains the p26 gene with 363 bp of the 5' flanking region (1) cloned into pBS- and was provided by C:'hristian Gross. (vii) Adenovirus major late gene promolter (pAML). Plasmid pMLH1, provided by D. Hawley (14), was digested with PstI and SstI, and the fragment containiing the adenovirus major late promoter (260 bp 5' and 536 bp 3V of the RNA start site) was cloned into pBluescript (pKS-) and called pAML. Cells, virus, and nuclear extract prep iaration. Sf9 cells (ATCC CRL 1711), obtained from GIBICO-Bethesda Research Laboratories, were grown in serutm-free Sf900 medium (GIBCO-Bethesda Research Lab( )ratories) supple-

3477

mented with penicillin G (50 U/ml)-streptomycin (50 pg/ml) (Whittaker Bioproducts)-amphotericin B (Fungizone, 375 ng/ml; Flow Laboratories) in shaking flasks (135 rpm) at 27°C. Cells were grown to a density of 2.0 x 106 cells/ml and infected with AcMNPV (E-2 strain, a gift from L. Volkman) at a multiplicity of infection of 10. The time point defined as zero hours postinfection (h p.i.) represents the time immediately before the addition of virus to the cells. During the first hour of infection, the cells were not shaken. Nuclear extracts were prepared as described by Hoopes and Rohrmann (15). In vitro transcription reactions. In vitro transcription reactions employing uninfected- or infected-cell nuclear extracts were carried out as previously described (15), with minor changes. Determination of the protein concentration (BioRad Protein Assay kit) ensured that equivalent amounts of nuclear extract were included in the reactions. Reaction mixtures (final volume, 20 ,ul) contained nuclear extract (3 mg/ml), DNA template (30 ,ug/ml), and MgCI2 (6 mM). Transcription was initiated after 25 min of preincubation (30°C) by the addition of nucleotides (600 ,uM [each] ATP, CTP, and GTP; 25 p.M UTP [Pharmacia]; 5 ,uCi of [a-32P]UTP [DuPont, NEN Research Products]), followed by incubation for 25 to 30 min at 30°C. The reaction was stopped by the addition of 30 p.l of H20-50 p.1 of stop buffer. The RNA was extracted once with water-saturated phenolchloroform and precipitated with ethanol. The labeled RNA products were subjected to gel electrophoresis on 7 M urea-5% acrylamide gels for 35 min on a Bio-Rad minigel apparatus at 175 V. Relative levels of runoff transcription products were determined by laser densitometry (GS300; Hoefer Scientific Instruments) of autoradiographs. Results from these measurements were confirmed by excising the gel bands and directly counting the [32P]UTP incorporated. Isolation of in vivo RNA. During the time course of an infection, cell samples (7.5 ml) were removed from shaking flasks, pelleted, quick frozen in liquid nitrogen, and stored at -80°C. RNA was isolated by a slight modification of the procedure of Sambrook et al. (21). Three hundred microliters of RNA extraction buffer (100 mM NaCl, 10 mM Tris-HCI [pH 8.1], 1 mM EDTA, 5 mM MgCI2, 0.5% Nonidet P-40, 1 mM dithiothreitol; 4°C) was added directly to frozen cell pellets, and the suspension was vortexed for 30 s to resuspend the cells and lyse the membranes. Cells were centrifuged for 2 min at 4°C in a microcentrifuge (12,000 x g). To the supernatant, 17.5 p.1 of a mixture containing 0.5 M EDTA, 8.5 ,ul of 20% sodium dodecyl sulfate, and proteinase K (100 ,ug/,I final concentration) was added and incubated for 30 min at 37°C. The samples were extracted with an equal volume of hot phenol (65°C), followed by extraction with phenol-chloroform and then chloroform-isoamyl alcohol. Finally, the RNA was precipitated with ethanol and resuspended in 30 to 60 p.1 of TE (10 mM Tris-HCI, 1 mM EDTA, pH 8.0). Primer extension analysis and DNA sequencing. For primer extension analysis of in vitro transcription start sites, RNA was prepared as described above, except that no labeled UTP was included in the reaction mixture, the UTP concentration was raised to 200 ,uM, and the pellet was dissolved in TE (10 mM Tris-HCl, 1 mM EDTA, pH 8.0). Primer extension analysis was done by using forward and reverse primers (5'-GTAAAACGACGGCCAGT-3' and 5'-TCACACAGGA AACAGCTATGAC-3', respectively) complementary to sequences of the vector, as previously described (15). To confirm that the in vitro transcription start sites corresponded to the in vivo start sites, RNA isolated from

3478

GLOCKER ET AL.

AcMNPV-infected Sf9 cells was purified as described above and analyzed by primer extension. The primers used for start site determination were 39K, 5'-AACCTTTGAAACAAC CCG-3' (nt -15 to -32 upstream of the ATG [10]); DNA polymerase, 5'-GCATATTCTGCAAAGCGC-3' (nt +50 to +33 downstream of the ATG [24]); and AcMNPV p26, 5'-TGTTGGATCAATTGCATA-3' (nt +39 to +22 downstream of the ATG [17]). Primer extension analysis of in vivo RNA was performed as described above, except that after precipitation with ethanol, the pellets were resuspended in 4 ,l of 0.1 N NaOH-1 mM EDTA and incubated for 25 min at 30°C to hydrolyze the RNA. To these samples, 4 p,l of 98% formamide-2% xylene cyanol-bromophenol blue was added and the samples were analyzed on DNA sequencing gels. DNA sequencing was performed by using a Sequenase sequencing kit (United States Biochemical Corp.), according to the manufacturer's instructions. Tests for the presence of nucleases. To test for RNase activity in the nuclear extracts, radioactively labeled RNA was made in a standard transcription reaction mixture using an uninfected-cell nuclear extract with SstI-digested pHR39K (Fig. 1A) as template DNA. After the normal incubation time, the reaction mixture was divided into four aliquots and treated as follows. One aliquot was stopped immediately by the addition of stop buffer. To the second aliquot, RNase A was added to a final concentration of 1 ,ug/,ul and incubated for a further 10 min at 30°C. To the third aliquot, uninfected extract was added at a concentration of 30% of total reaction volume, and the reaction was incubated for a further 30 min at 30°C. This reaction tested for the presence of RNase activity in the uninfected-cell nuclear extract. The fourth aliquot was used to test for RNase activity in late infected-cell nuclear extracts. This was done by mixing it with a nuclear extract prepared at 34 h p.i. at a final concentration of 30% of the total reaction volume and incubating for 30 min at 30°C. All samples were then processed like standard in vitro transcription reactions and analyzed by polyacrylamide gel electrophoresis and autoradiography. To test for DNase activity, mock transcription reactions were prepared in which SstI-digested pHR39K template DNA (30 ,ug/ml) was incubated with nuclear extracts (3 mg/ml) produced at various times postinfection. Two reactions for each extract sample were prepared, and EDTA was added at a final concentration of 10 mM to one of the samples. After incubation for 60 min at 30°C, the DNA templates were purified by extraction with phenol and precipitation with ethanol. Samples were loaded onto an agarose gel in sample buffer containing RNase A (100 p,g/ml) to remove RNA present in nuclear extracts. After electrophoresis, the DNA was stained with ethidium bromide and photographed on a UV transilluminator. RESULTS Rationale for templates used for in vitro transcription. The purpose of this investigation was to determine whether the transcriptional events observed during the baculovirus infection cycle were mirrored by the in vitro transcription system. The transcriptional processes investigated and the gene promoters included in these studies are listed as follows. (i) Gene class: the genes selected are examples of promoters termed in the literature as immediate early, delayed early, early, and late. (ii) Promoter structure: the AcMNPV DNA polymerase gene was investigated because unlike several other baculovirus early promoters it lacks a conventional

J. VIROL.

TATA promoter element (see Fig. 5). (iii) Transactivation: the AcMNPV 39K and OpMNPV gp64 genes were employed because they both have been shown to be transactivated in vivo (4, 11). (iv) Enhancement by hr sequences: the AcM NPV 39K gene was used as a template because it is enhanced by hr sequences (12). The AcMNPV p26 gene was investigated because it is next to an enhancer sequence (hr5) in its native location (17). The OpMNPV p26 gene (1) has a sequence similar to that of the AcMNPV p26 gene but lacks an adjacent enhancer sequence and would allow examination of the influence of enhancer proximity for comparison to the AcMNPV p26 gene. (v) A non-baculovirus promoter: the adenovirus major late promoter was employed because it is commonly used for investigations of transcription in higher eukaryotes and would allow comparison of events occurring in baculovirus-infected cell extracts to better-characterized systems. In vitro transcription by nuclear extracts from uninfected Sf9 cells. Previously, it was demonstrated that plasmids containing promoters from the OpMNPV gp64, AcAMNPV IE-1 (a gene that encodes a transactivating factor) (13), and the adenovirus major late genes are transcribed by nuclear extracts from uninfected Sf9 cells (15). In these assays they behaved as immediate-early genes. In order to characterize the transcription of early promoters in vitro, the ability of nuclear extracts from uninfected Sf9 cells to transcribe a variety of other baculovirus genes was compared with in vitro transcription of the gp64 and adenovirus major late promoter constructs previously investigated. The constructs used in these investigations included 5' regulatory sequences of genes reported to be (i) immediate early (the gp64 promoter p64CAT-319) (15); (ii) early (the DNA polymerase promoter pDNAP) (24); (iii) delayed early (the 39K promoters p39K and pHR39K) (11); the AcMNPV p26 promoter (16) pAcHRp26 and the OpMNPV p26 promoter pOpp26 (1); and (iv) a commonly used RNA polymerase II promoter (the adenovirus major late promoter pAML). For constructs see Fig. 1. All these templates gave runoff transcripts of the expected sizes when used in the in vitro transcription system with uninfected nuclear extracts (Fig. 2A through F, lanes 1). Primer extension analyses confirmed that the transcripts initiated at the same position both in vitro and in vivo (see below). Whereas the p64CAT-319, pHR39K, and pAML promoter templates produced strong signals, runoff transcripts from pDNAP, pAcHRp26, and pOpp26 were detected at a lower level (Fig. 2A through F, lanes 1). These experiments indicate that nuclear extracts from uninfected Sf9 cells accurately initiate and elongate RNA from the promoters of these genes. This suggests that basal levels of in vitro transcription from these promoters are dependent solely on host enzymes and transcription factors. Transactivation of transcription by nuclear extracts from AcMNPV-infected Sf9 cells. In vitro transcription of templates was examined by using Sf9 nuclear extracts prepared at 0 (uninfected), 4, 8, 12, 16, 24, and 34 h p.i. (Fig. 2A through F). Although the signal produced with uninfectedcell nuclear extracts was relatively weak, there was a marked increase in the intensity of the runoff transcript from most templates employing nuclear extracts prepared at 4 and 8 h p.i. (Fig. 2A through F, compare lanes 2 and 3 with lanes 1). The ratios of the intensity of the runoff signal produced by 8-h p.i. extracts to that produced by uninfected-cell nuclear extracts for each template are as follows: pHR39K, 8; pDNAP, 4.7; pOpp26, 3.3; p64CAT-319, 2.7; pAML, 1.3; and pAcHRp26, 1.2. This activation decreased when nuclear extracts prepared after 12 and 16 h p.i. were employed (Fig.

A

1

2

3

4

5

6

B

7

1

603 -

603 -

310 -

310-

234 194 -

234 -

-

214 nt

194

118 -

2

3

-U"

p _g

4

5

6

7

-

316 nt

-

118 -

6t ill I

U2

0 4-

0

4 8 12 16 24 34 hours post infection

1

2

C 603 310

I

3

4

5

6

0

..raw.

--

1

2

3

4

34

5

603 -

544 nt

-

310

234 194 -

-

234 194 -

118 -

6

4 8 12 16 24 hours post infection

D

7

43

21

-

143 nt

-

218 nt

118 -

21 Iii 0

E 603 -

62

4 8 12 16 24 hours post infection 1

9

2

3

'uilii

3

4

0

34

F

5

qW '. .Fp

1 603-

---

378 nt

310

-

310 -

234 194

-

234 194

-

-

118

-

118

-

¶111.1

4 8 12 16 hours post infection 2

3

4

5

I9

-

64

t

4 8 12 16 0 4 8 12 16 hours post infection hours post infection FIG. 2. In vitro transcription of templates, using AcMNPV-infected Sf9 cell nuclear extracts prepared at different times postinfection. The transcription products shown in the panels were obtained by using the following template DNAs: SstI-digested pHR39K (A), EcoRI-digested p64CAT-319 (B), SstI-digested pAML (C), HindIll-digested pOpp26 (D), SstI-digested pAcHRp26 (E), and XhoI-digested pDNAP (F). The numbers on the left indicate the positions of selected radiolabeled HaeIII-digested +X174 DNA fragments. The positions and sizes of the runoff transcription products are marked with arrows. Below each autoradiogram is a graphic representation of a densitometry scan of the runoff transcript signal, indicating relative levels of transcription. O.D. is the relative absorption of the runoff transcript.

0

3479

3480

J. VIROL.

GLOCKER ET AL. infected

uninfected

214 nt

4 1 2 3 FIG. 3. Lack of in vitro enhancement by the hr5 enhancer sequence. Lanes 1 and 3 (+) contain Sstl-digested pHR39K with intact hr5 as template; lanes 2 and 4 (-) contain the same template as lanes 1 and 3 except it was also digested with EcoRI to destroy hr5. Transcripts in lanes 3 and 4 were produced from uninfected-cell nuclear extracts, and transcripts in lanes 1 and 2 were produced by using nuclear extracts prepared at 8 h p.i. The position and size of the expected runoff product is marked with an arrow. (Note: in these reactions, the DNA concentrations used [45 ,ug/ml] account for levels of activation that are different from those shown in Fig. 2A). For a description of the graph, see the legend for Fig. 2.

2A through F, lanes 4 and 5). In vitro transcription was also examined for IE-1 by using a template described previously (15). We found that IE-1 showed an activation profile similar to those of the other templates (Fig. 2), being activated maximally at 8 h p.i. and showing about 10-fold stimulation (data not shown). No transcription was observed from late promoters. The pHR39K and p64CAT-319 templates contain both early and late promoter elements that are utilized in vivo (2, 10) (see Fig. 7). There was no evidence for specific transcription from late promoters by extracts from late-infected cells (Fig. 2A and B). Presence of the baculovirus enhancer sequence hr5 on templates did not increase levels of in vitro transcription. The AcMNPV genome has several homologous repeated (hr) sequences (6) that contain palindromes centered around EcoRI sites. Fusion of hr sequences to a 39K-CAT construct was shown to result in enhanced levels of CAT activity when the construct was transfected into cells along with a plasmid containing IE-1 (12). Nuclear extracts from Sf9 cells were tested for enhancer sequence-dependent activation of in vitro transcription (Fig. 3) by using several constructs. In the first tests, a plasmid containing the hr5 enhancer element upstream of the AcMNPV 39K gene promoter was used (Fig. 1A, pHR39K). Digestion of pHR39K with SstI, which leaves the hr5 region intact and located in cis to the 39K promoter region, resulted in a runoff transcript of 214 nt, using nuclear extracts prepared from uninfected cells (Fig. 3, lane 3). Double digestion of the same plasmid with SstI and EcoRI, which destroyed the enhancer element but left the promoter intact, failed to reduce the level of transcription (Fig. 3, lane 4). Similar results were obtained when nuclear extracts produced from cells at 8 h p.i. were tested with these templates (Fig. 3, lanes 1 and 2). A repeat of this experiment with two different templates, p39K (without hr5; Fig. 1A) and pHR39K (containing hr5; Fig. 1A), gave the same negative result (data not shown). Further tests with a

plasmid comparable to the CAT construct reported to give maximal levels of enhancement by Guarino and Summers (12) (this plasmid, called p39CAT-Q-, includes the HindlIl Q fragment containing the hr5 enhancer sequence inserted upstream of the 39K promoter region) also showed no in vitro transcriptional activation by hr5 (data not shown). To investigate enhancement by a gene that normally contains an adjacent hr sequence, transcription of the AcMNPV p26 gene (which is located immediately downstream of hr5) was examined. However, this gene could not be tested for hr5-mediated enhancement because we found that the RNA start site was located 17 nt upstream of the published location. As a result, the promoter overlapped the hr sequence (see below and Fig. 4C) and EcoRI digestion to eliminate the hr sequence destroys the AcMNPV p26 gene promoter region. Therefore, the OpMNPV p26 gene, which is highly homologous to the AcMNPV p26 gene (1) but does not contain an adjacent enhancer sequence in its native state, was used. The hr5 sequence was cloned upstream of the promoter region, using the Sall site of the OpMNPV p26 gene indicated in Fig. 1D. No evidence of enhancement was obtained by testing pOpp26 with and without the hr5 enhancer sequence (results not shown). Therefore, this system does not appear to demonstrate enhancer-dependent activation of gene expression. Primer extension analysis of in vitro and in vivo RNA. Primer extension analysis was used to confirm that the in vitro RNAs produced by using nuclear extracts from uninfected cells initiated at the start sites used in vivo. To avoid detection of viral mRNAs in the nuclear extracts from infected cells, we used primers complementary to sites in the vector. To obtain a primer extension product for the AcM NPV p26 gene of suitable length, an additional template was constructed (pAcHRp26A, see Materials and Methods). The primer extension products were resolved on sequencing gels next to a sequencing ladder of the gene, generated by using the same primer. Results from the primer extension analysis of the 39K gene, the OpMNPV p26 gene, and the AcMNPV p26 gene are shown in Fig. 4, whereas the positions of the mRNA start sites in the corresponding DNA sequences are shown in Fig. 5. The in vitro transcription start sites of the 39K gene and the OpMNPV p26 gene (Fig. 4A and B) are identical to the start sites published for in vivo RNA (1, 10). However, the AcMNPV p26 gene primer extension product from in vitro RNA mapped to a transcription start site 17 nt upstream of the one previously suggested (17, 20) (Fig. 4C). This site was confirmed by primer extension analysis of in vivo mRNA (Fig. 6A). In vitro transcription of pDNAP showed a runoff transcript of about 218 nt (Fig. 2F). Our attempts to map the in vitro transcription start sites by primer extension were unsuccessful (probably because the RNA concentration produced by uninfected-cell extracts was so low). However, we did map the in vivo DNA polymerase mRNA start site and found a major start site (Fig. 5 and Fig. 6B) located 122 bp upstream of the DNA polymerase ATG, which corresponds to one of the major start sites previously reported (24). Initiation from this site would produce a runoff transcript similar in size to that observed by in vitro transcription assays. Therefore the in vitro and in vivo DNA polymerase RNA appear to initiate at the same site. Comparison of in vitro transcription by nuclear extracts from a time course of infection with the mRNA levels present in infected cells. To determine whether the increased levels of in vitro transcription seen with 4- and 8-h p.i. extracts were a reflection of in vivo RNA levels, RNA was isolated

BACULOVIRUS EARLY GENE TRANSACTIVATION

VOL. 66, 1992

A

B TGCA 4

\

T

C

W

P

TGCA

P

P

TGCA

a

A A

_

Am

C

T ZA A T T

W

A

*>-

A

G

G C T

-

C

Ar A -

A T T T

W".* 0

i

A

3481

_

_

Is

d A_w _*

;-

- E ->

-

C A G T A

E E

FIG. 7. Primer extension analysis of 39K RNA over a time of infection. Primer extension lanes are designated by the hour postinfection, indicating the time at which mRNA was prepared. A sequencing ladder generated from a plasmid containing the 39K gene and the same primer as used in the primer extension analysis is indicated. E marks the start sites of the early 39K transcripts and L marks the late transcription start site. course

Infected-cell nuclear extracts prepared at 4 and 8 h p.i. activated transcription of most baculovirus genes tested from 3- to 10-fold. Transactivation has been investigated in vivo by cotransfection of AcMNPV 39K or OpMNPV gp64 CAT constructs with the AcMNPV transactivating gene IE-1 into Sf9 cells (4, 11). Although gp64 showed similar levels of transactivation in vitro and in vivo, the 39K gene was transactivated to much higher levels when measured in the in vivo system. Direct comparison of these activated levels is limited by fundamental differences in the two methods of analysis, and identical results are not necessarily expected. The activity measured by in vivo CAT assays may be amplified by multiple rounds of translation of a single mRNA and by recycling of the CAT enzyme. In contrast, in vitro transcription by infected-cell nuclear extracts involves a complex mixture likely to contain not only IE-1 and IE-N (5) but a variety of other as-yet uncharacterized viral transactivators and attenuators and is assayed solely on the basis of the levels of RNA transcribed. The baculovirus promoters tested showed various levels of activation with infected-cell extracts. This activation of transcription could be due to a general, nonspecific improvement in the extracts, possibly as a result of more efficient extraction of transcription factors. Such extracts should cause a higher level of basal transcription from all promoters. Alternatively, the activation could be due to the presence of specific transactivators in the infected-cell nuclear extracts, in which case only targeted promoters would be expected to be activated. Our data show that different promoters were activated to different levels. The 39K and gp64 promoters were activated to the highest extent,

VOL. 66, 1992

BACULOVIRUS EARLY GENE TRANSACTIVATION

3483

B

A 1

2

3

4

1 M C

-

3

2 +

-

+

-

+

4 -

+

5 -

+

6 -

+

7 -

+

214 nt 6108

5090\4072

61

3054 2

2036 1

2

3

4

FIG. 8. Assays for nuclease activity in nuclear extracts. (A) Assay for RNase activity. A standard transcription reaction using an uninfected-cell nuclear extract and SstI-digested pHR39K template DNA was performed. After the normal incubation time, the reaction was divided in four parts and treated as follows. Lane 1, stopped immediately; lane 2, RNase A was added to a final concentration of 1 ,ug/,u, and the reaction was incubated for a further 10 min at 30°C; lane 3, to test for RNase activity in the uninfected-cell nuclear extract, this extract was added at a concentration of 30% of total reaction volume, and the reaction was incubated for a further 30 min at 30°C; and lane 4, to test for RNase activity in late-infected cell nuclear extracts, nuclear extract prepared at 34 h p.i. was added at a concentration of 30% of total reaction volume and incubated for 30 min at 30°C. The position and size of the expected runoff product are marked with an arrow. For a description of the graph, see the legend for Fig. 2. (B) Assay for DNase activity. Extracts were tested for DNase activity by incubating SstI-digested pHR39K template with nuclear extracts in the presence (+) or absence (-) of EDTA. The following indicate the stage of infection at which the nuclear extracts were prepared: lane 1, uninfected; lane 2, 4 h p.i.; lane 3, 8 h p.i.; lane 4, 12 h p.i.; lane 5, 16 h p.i.; lane 6, 24 h p.i.; lane 7, 34 h p.i. Lane C, nuclear extract was not included in the transcription reaction; lane M, 1-kb ladder (Bethesda Research Laboratories) used as size standard.

whereas the adenovirus major late and AcMNPV p26 promoters were not significantly activated. These data argue against nonspecific activation and strongly suggest the involvement of specific transactivation by components of the infected-cell nuclear extracts. Constructs containing the baculovirus enhancer hr5 were examined for elevated levels of transcription, using both uninfected- and infected-cell nuclear extracts. No evidence for enhancement was seen. Conditions or factor(s) essential for enhancement could have been either lacking or inactivated in the extracts. Similar problems could account for the lack of transcription from the late promoters present on the 39K and gp64 constructs. The decline in activity of nuclear extracts prepared late after the initiation of infection suggested that an inhibitor of transcription was present or that nucleases or proteases may be affecting the system. We found that, although there was no evidence of high levels of RNase or protease activity in the nuclear extracts, levels of DNase activity capable of completely hydrolyzing template DNA were detected in the late-infected nuclear extracts. Wilson and Miller (25) have shown that during the course of baculovirus infection, host cell DNA is not degraded. The DNase activity we observed may be the result of viral infection, possibly caused by a turnoff of the synthesis of a host DNase inhibitor or by destabilization of cell compartmentalization (e.g., DNasecontaining lysosomes). The nuclear extraction preparation protocol may cause the release of such nucleases into the extract.

The 39K promoter was used to compare the transcriptional activity of nuclear extracts prepared at different times

-

after the onset of infection with the levels of 39K transcripts present in the infected cells. The results indicated that the nuclear extracts supported transcription of the 39K gene at high levels well before 39K mRNA accumulated to measurable levels in infected cells. The marked contrast between these results may reflect the difference in the two methods of

analysis. In vitro transcription assays measure the ability of and factors present in a nuclear extract to initiate transcription from an abundant exogenous DNA template. In contrast, primer extension analysis of in vivo RNA measures the accumulation of specific RNA transcribed from an endogenous template. This accumulation is governed by the quantity of template, the rate of transcription from the template (influenced by such factors as promoter strength and level of transactivation), the length of time RNA is allowed to accumulate, and the rate of RNA degradation. Comparison of the in vitro and in vivo data suggested that although enzymes and transcription factors necessary for high levels of transcription exist at 4 and 8 h p.i., they were not maximally exploited at this time in vivo. This may be caused by a lack of template DNA in cells infected at a relatively low multiplicity of infection or that levels of early gene mRNA may not accumulate until after DNA replication, when the number of copies of early genes in a cell is amplified, thereby permitting increased levels of transcription. The study described in this report was designed to determine whether this in vitro system mirrored transcriptional events occurring in cells during baculovirus infection. The data we have presented indicate that the use of infected-cell nuclear extracts may prove useful for the identification of enzymes

3484

J. VIROL.

GLOCKER ET AL.

factors involved in transactivation of early genes. However, the presence of higher levels of DNase activity in nuclear extracts from late in infection complicates the use of this approach for the study of late gene regulation. In addition, factors involved in enhancement of early gene expression, through interaction with known baculovirus enhancer sequences, were not evident in these extracts. We are currently investigating the use of different extraction protocols to develop in vitro systems suitable for the investigation of all transcription events occurring during the baculovirus infection cycle. ACKNOWLEDGMENTS The authors thank C. Rasmussen, C. Gross, and G. Blissard for suggestions and criticisms of the manuscript. This work was supported by grant Al 21973 from the NIH. REFERENCES 1. Bicknell, J. N., D. J. Leisy, G. F. Rohrmann, and G. S. Beaudreau. 1987. Comparison of the p26 region of two baculoviruses. Virology 161:589-592. 2. Blissard, G. W., and G. F. Rohrmann. 1989. Location, sequence, transcriptional mapping, and temporal expression of the gp64 envelope glycoprotein gene of the Orgyia pseudotsugata multicapsid nuclear polyhedrosis virus. Virology 170:537555. 3. Blissard, G. W., and G. F. Rohrmann. 1990. Baculovirus diversity and molecular biology. Annu. Rev. Entomol. 35:127155. 4. Blissard, G. W., and G. F. Rohrmann. 1991. Baculovirus gp64 gene expression: analysis of sequences modulating immediateearly transcription and transactivation by IEL. J. Virol. 65:58205827. 5. Carson, D. D., M. D. Summers, and L. A. Guarino. 1991. Transient expression of the Autographa californica nuclear polyhedrosis virus immediate-early gene, IE-N, is regulated by three viral elements. J. Virol. 65:945-951. 6. Cochran, M. A., and P. Faulkner. 1983. Location of homologous DNA sequences interspersed at five regions in the baculovirus AcMNPV genome. J. Virol. 45:961-970. 7. Friesen, P. D., and L. K. Miller. 1986. The regulation of baculovirus gene expression. Curr. Top. Microbiol. Immunol. 131:31-49. 8. Fuchs, Y. L., M. S. Woods, and R. F. Weaver. 1983. Viral transcription during Autographa califomnica nuclear polyhedrosis virus infection: a novel RNA polymerase induced in infected Spodoptera frugiperda cells. J. Virol. 48:641-646. 9. Grula, M. A., P. L. Buller, and R. F. Weaver. 1981. Alpha amanitin-resistant viral RNA synthesis in nuclei isolated from nuclear polyhedrosis virus-infected Heliothis zea larvae and Spodoptera frgiperda cells. J. Virol. 38:916-921. 10. Guarino, L. A., and M. W. Smith. 1990. Nucleotide sequence and characterization of the 39K gene region of the Autographa califomnica nuclear polyhedrosis virus. Virology 179:1-8. 11. Guarino, L. A., and M. D. Summers. 1986. Functional mapping

12.

13. 14.

15.

16.

17.

18. 19. 20.

21. 22.

23. 24. 25. 26.

of a trans-activating gene required for expression of a baculovirus delayed-early gene. J. Virol. 57:563-571. Guarino, L. A., and M. D. Summers. 1986. Interspersed homologous DNA of Autographa califomnica nuclear polyhedrosis virus enhances delayed-early gene expression. J. Virol. 60:215223. Guarino, L. A., and M. D. Summers. 1987. Nucleotide sequence and temporal expression of a baculovirus regulatory gene. J. Virol. 61:2091-2099. Hawley, D. K., and R. G. Roeder. 1987. Functional steps in transcription initiation and reinitiation from the major late promoter in a HeLa nuclear extract. J. Biol. Chem. 262:34523461. Hoopes, R. R., Jr., and G. F. Rohrmann. 1991. In vitro transcription of baculovirus immediate early genes: accurate mRNA initiation by nuclear extracts from both insect and human cells. Proc. Natl. Acad. Sci. USA 88:4513-4517. Huh, N. E., and R. F. Weaver. 1990. Identifying the RNA polymerases that synthesize specific transcripts of the Autographa californica nuclear polyhedrosis virus. J. Gen. Virol. 71:195-201. Liu, A., J. C. Qin, C. Rankin, S. E. Hardin, and R. F. Weaver. 1986. Nucleotide sequence of a portion of the Autographa califomica nuclear polyhedrosis virus genome containing the EcoRI site-rich region (hr5) and an open reading frame just 5' of the plO gene. J. Gen. Virol. 67:2565-2570. Nissen, M. S., and P. D. Friesen. 1989. Molecular analysis of the transcriptional regulatory region of an early baculovirus gene. J. Virol. 63:493-503. Ooi, B., and L. K. Miller. 1988. Regulation of host RNA levels during baculovirus infection. Virology 166:515-523. Rankin, C., B. F. Ladin, and R. F. Weaver. 1986. Physical mapping of temporally regulated, overlapping transcripts in the region of the 10K protein gene in Autographa californica nuclear polyhedrosis virus. J. Virol. 57:18-27. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Theilmann, D. A., and S. Stewart. 1992. Tandemly repeated sequence at the 3' end of the IE-2 gene of the baculovirus Orgyia pseudotsugata multicapsid nuclear polyhedrosis virus is an enhancer element. Virology 187:97-106. Theilmann, D. A., and S. Stewart. 1991. Identification and characterization of the IE-1 gene of Orgyia pseudotsugata multicapsid nuclear polyhedrosis virus. Virology 180:492-508. Tomalski, M. D., J. Wu, and L. K. Miller. 1988. The location, sequence, transcription, and regulation of a baculovirus DNA polymerase gene. Virology 167:591-600. Wilson, M. E., and L. K. Miller. 1986. Changes in the nucleoprotein complexes of a baculovirus DNA during infection. Virology 151:315-328. Yang, C. L., D. A. Stetler, and R. F. Weaver. 1991. Structural comparison of the Autographa californica nuclear polyhedrosis virus-induced RNA polymerase and the three nuclear RNA polymerases from the host, Spodoptera frugiperda. Virus Res. 20:251-264.