- Email: [email protected]
Identification of a Unique Family of F-Box Proteins in Atadenoviruses
Virology 304, 425–433 (2002) doi:10.1006/viro.2002.1734
Identification of a Unique Family of F-Box Proteins in Atadenoviruses Gerald W. Both 1 CSIRO Molecular Science, North Ryde, New South Wales, Australia 2113 Received June 3, 2002; returned to author for revision July 25, 2002; accepted August 5, 2002 Ovine adenovirus isolate 287 (OAdV-7) is the prototype of the atadenoviruses, a genus whose strategy for infection and replication is still being elucidated. A transcription unit at the right end of the genome contains four related genes (ORFs RH1, 2, 4, and 6), at least three of which are nonessential for replication in vitro. Related genes are also present in the genomes of bovine and avian atadenoviruses. To investigate how these apparently redundant genes are decoded, a more detailed transcription map of the right end of the OAdV-7 genome has been deduced. Eight transcripts that were derived from a promoter in the terminal repeat sequence were identified. Five were potentially bicistronic. The transcripts could encode all the potential proteins of the region subject to efficient reinitiation of translation. However, the most interesting and surprising finding in this work was that the related RH proteins carry an F-box motif. This was first identified in OAdV-7 RH1 and subsequently found in RH2, 4, and 6 proteins and the related reading frames from the bovine and avian atadenoviruses. Although very rare among viral proteins, several hundred cellular proteins contain F-box motifs. F-box proteins facilitate the degradation of a variety of important regulatory proteins via SCF ubiquitin ligase complexes. Thus, it appears that atadenoviruses have adopted a strategy to regulate a key cellular pathway(s) that distinguishes them from the other adenovirus genera and from most viruses in general. © 2002 Elsevier Science (USA)
INTRODUCTION
reading frames (ORFs 1, 2, 4, and 6) that were first recognized in OAdV-7 by their high level of amino acid identity (Fig. 1) (Xu et al., 1997). This relatedness is conserved across viruses for ORFs 8 and 9 from DAdV-1 (Hess et al., 1997; Khatri and Both, 1998) and for ORFs RH0, 1, 2, 4, and 6 of BAdV-4 (GenBank Accession No. NC002685) (Fig. 1). This was not previously noticed for ORF RH0 because it lacks an initiating methionine. In OAdV-7, deletion of ORFs RH2–6 is not lethal (Xu et al., 1997), suggesting that some of these ORFs are redundant, at least in vitro. In addition, the short ORF RH3 is not conserved in BAdV-4 and ORF RH5 is not present in DAdV-1. These observations raised questions as to whether each of the ORFs could actually code for a protein. In this work a more detailed transcription map of the region has been elucidated to investigate coding potential. Additional computer searches have also provided an important clue as to the role of these proteins. The indications are that atadenoviruses have a very different strategy for modulating the host cell compared with other adenoviruses, and indeed, most known viruses.
Adenoviruses have recently been classified into at least three genera based on phylogenetic analyses of the hexon and protease proteins and the distinct genome arrangements of members of the mast-, avi-, and atadenoviruses (Benko and Harrach, 1998; Dan et al., 1998; Harrach and Benko, 1999). Members of the atadenoviruses, so named because of the high A/T content of their genomes, include the prototype ovine adenovirus type 7 (OAdV-7), Egg Drop syndrome virus (DAdV-1), and bovine adenovirus type 4 (BAdV-4), all of which have been completely sequenced. Other viruses from cattle (BAdV-7 and -8), black-tailed deer, goats, and possum are also included in the genus but are incompletely characterized (Both, 2001). While the central core of the characterized genomes encodes the DNA replicating enzymes and structural protein, not all capsid proteins are conserved between the mast- and atadenoviruses (Vrati et al., 1996). Emphasizing their distinct nature, the left and right ends of the atadenovirus genomes also lack strong homologues of the mastadenovirus E1A/B and E4 regions and instead carry genes that are unique to the atadenoviruses and sometimes to the individual virus. The functions of these genes have yet to be determined. Of particular interest in this work is a family of four open
RESULTS AND DISCUSSION Transcription at the right end of the OAdV-7 genome The right end of the genome is transcribed in a right to left direction. To identify proteins that could be produced from the right end of the OAdV-7 genome, RT-PCR was used to identify transcripts and their temporal relationship. RNA harvested from infected cells at 2, 6, 12, and
1 To whom correspondence and reprint requests should be addressed at: CSIRO Molecular Science, P.O. Box 184, North Ryde, NSW 1670, Australia. Fax: 61 2 9490 5005. E-mail: [email protected].
425
0042-6822/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved.
426
GERALD W. BOTH
FIG. 1. Comparison of the right ends of the genomes for OAdV-7, BAdV-4, and DAdV-1. Related reading frames within and across virus species are indicated by similar cross-hatching. The relative location and sense of primers GB1-6 and MF6, 7 used for RT-PCR is indicated with arrowheads. E4 1, 2, and 3 refer to the ORFs for the proposed E4 region of OAdV-7.
24 h postinfection (p.i.) and from uninfected cells was reverse transcribed and cDNAs were amplified using appropriate primer pairs (Table 1). Primer pairs GB1/5 (Fig. 2, top) produced two PCR products that were amplified from infected but not uninfected cells, in agreement with an earlier study (Xu et al., 1997) and these were detectable at 6, 12, and 24 h p.i. (lanes 3, 5, and 7), but not at 2 h p.i. (lane 2). Extending the number of PCR cycles from 33 to 40 did not change this profile (data not shown). Cloning and sequencing of these products confirmed their identity and splice junctions (Xu et al., 1997)
(Table 2) and their derivation from transcripts 1 and 2 (Fig. 3). Primers GB3/5 amplified three virus-specific products that were sequenced, the largest of which (⬃1580 bp, Fig. 2, top, lane 13) was shown by sequencing to be unspliced and therefore probably derived from traces of viral DNA that contaminated the mRNA extracted at 24 h p.i. The smaller ⬃1.4-kb product was evident by 6 h p.i. (lanes 9 and 11) and was derived from transcript 3 that was contiguous with the genome, except for an intron that was spliced out between ORFs RH1 and RH2 (Fig. 3). This splicing event was also identified by
TABLE 1 Primers Used for RT-PCR and Their Location in the OAdV Genome Primer GB1 a GB2 GB3 GB4 GB5 a GB6 MF6 MF7 a
Sequence
Location
CGGGATCCATACAGGCCAAGCATCA (⫹) CGGATCCAACTTTGTGAGCAGGCC (⫹) CGGGATCCAACCGCATTCAGGCCAA (⫹) CCGGATCCAACTGATTTAACTGCCC (⫹) GCCGGATCCGGTATGATTGCTTTTCTTCC (⫺) CGGGATCCTCCCAGTGGGTGTGGC (⫺) ACT CAT CAG GAC AAC GGC ATT TAA G (⫹) GCATTTTAAGCTCTCTGCTTCAC (⫹)
RH6 3⬘ end RH5 3⬘ end RH4 3⬘ end RH2 3⬘ end RH1 5⬘ end Adjacent to ITR RH5 3⬘ end RH5 3⬘ end
Primers GB1 and GB5 are identical to PCR1 and PCR2 (Xu et al., 1997).
ATADENOVIRUS F-BOX PROTEINS
427
FIG. 2. Products derived by RT-PCR. RNA prepared from infected (I) and uninfected (U) CSL503 cells at various times postinfection were reverse transcribed and amplified by PCR with the primer pairs indicated. Species indicated with a dot were cloned and sequenced to determine transcript identity and to identify RNA splice junctions.
the products of GB4/GB5 and GB4/GB6 (data not shown). The minor ⬃1-kb species (lane 13) was not evident until 24 h p.i. and resulted from the loss of an additional intron that spliced sequences within ORF RH2 to a site at the 5⬘ end of ORF RH4 (Table 2) (Fig. 3, transcript 5). No splicing event that created an ORF RH2-specific transcript was identified. The promoter that produces right-end transcripts probably lies within the ITR of the genome (Xu et al., 1997). Primer GB6, which was located adjacent to the right ITR at nucleotides 61–46 of the genome, was used in conjunction with downstream primers GB1, 2, or 3 to determine whether additional transcripts could be amplified.
PCR products were detected at 12–24 h postinfection with each combination (Fig. 2, bottom, lanes 5, 7, 9, and 11), confirming that the promoter included the ITR sequence. Nucleotide sequencing of these products revealed a major splice junction that was located two nucleotides upstream of the methionine for ORF RH1. Splicing occurred from this point to different sites within ORFs RH4 (transcript 4), RH5 (transcripts 7 and 8), and RH6 (transcript 6) (Table 2), thus eliminating ORF RH1 coding sequences (Fig. 3). In addition, GB2/GB5 and GB2/GB6 combinations were used to look for transcripts that might encode the ORF RH5 reading frame. However, only short PCR prod-
428
GERALD W. BOTH TABLE 2 Splice Sites in Transcripts from the Right-Hand End Promoter
Transcript 1 2 3 4 5 6 7 8 a
Primer pair GB GB GB GB GB GB GB GB
1/5 1/5 3/5 3/6 3/5 1/6 1/6 2/6
First intron 5⬘ splice site a
First intron 3⬘ splice site a
Second intron 5⬘ splice site a
Second intron 3⬘ splice site a
TTTAAACG⬘GTGAGT TTTAAACG⬘GTGAGT TTTAAACG⬘GTGAGT CTTCTACG⬘GTATGA TTTAAACG⬘GTGAGT CTTCTACG⬘GTATGA CTTCTACG⬘GTATGA CTTCTACG⬘GTATGA
GTTGTTAG⬘TTTCTT TTTTACAG⬘GTCATG TTTAATAG⬘GAAGAA TTTTGTAG⬘GCGCCT TTTAATAG⬘GAAGAA TTTTACAG⬘GTCATG GTTGTTAG⬘TTTCTT GTTTTAAG⬘AGAAGA
na na na na GTGGTCTG⬘GTTGGT na AAAGTTTT⬘GTAAGT na
na na na na TTTTGTAG⬘GCGCCT na TTTTACAG⬘GTCATG na
GenBank Accession No. U40839.
ucts that were derived from transcripts 1, 7, or 8 were obtained. When ORF RH5-specific primers MF6 and MF7 from the intron region were used in conjunction with GB6 (Fig. 3), a larger product of ⬃800 bp that was visible at 6 h but not 2 h p.i. was obtained (Fig. 2, lanes 13–15). Sequencing revealed that this product was derived from transcript 4 (Fig. 3). Thus, eight transcripts from the region were identified (Fig. 3). No ORF RH3-specific transcript was identified and as this ORF is not con-
served in the closely related BAdV-4 genome (Fig. 1) it is unlikely to encode a genuine viral protein. It is likely that all transcripts terminate at the polyadenylation signal that was mapped just beyond the ORF for RH6 (Xu et al., 1997). Transcript four, at least, reads through a potential poly(A) signal that lies within ORF RH5. The failure to identify longer PCR products with primers GB1 and GB5 or 6 was probably due to the preferential amplification of short PCR products under the conditions used.
FIG. 3. Transcription map of the right end of the OAdV-7 genome. The black box to the right of each transcript indicates the promoter within the ITR. Numbers with short tick marks above transcripts refer to primers also shown in Fig. 1. Bold and broken lines indicate sequenced and unsequenced regions of transcripts 1–8, respectively. Thin lines are introns. (*) Polyadenylation site. (M) Signifies the first methionine codon within an ORF. (ˆ) Translation stop codons. M is 11 and 6 nucleotides downstream of ˆ in spliced transcripts 3 and 4, respectively. (#) Methionine codon in ORF RH4 is within the F-box motif. (m) Splicing truncates and frameshifts reading frame for ORF RH2. (Hr pi) Signifies when PCR products were first detected. Product signifies possible protein(s) encoded.
ATADENOVIRUS F-BOX PROTEINS
Proteins from the right end of the OAdV-7 genome The eight transcripts could code for proteins RH1, 2, 4, 5, and 6. Transcripts 1, 2, 3 (detectable at ⬃6 h p.i.), and 5 code for protein RH1 which begins at the first methionine in the genome at that end. RH2 and RH6 proteins could also be produced from transcripts 3 and 2, respectively (Fig. 3), by reinitiation of translation because the initiation codons for these proteins occur only 11 and 7 nucleotides downstream from the terminator for ORF RH1 (Kozak, 1984). An extended form of RH6 could also be produced from transcript 1 using an upstream initiation codon (Fig. 3) but this is not conserved in the BAdV-4 RH6 homologue (Fig. 4). At later times p.i. (⬃24 h), modified splicing produces transcripts 6 and 7 that can only code for ORF RH6. Transcript 4 could produce RH4 protein and RH5 as the second product by reinitiation of translation five nucleotides downstream of the ORF RH4 stop codon (Fig. 3). Transcript 5 could produce RH1 and an aberrant form of RH2 (due to a splice and frameshift) (Fig. 3) but is unlikely to produce RH4 because translation from ORF RH2 does not terminate before the ORF RH4 initiation codon (Kozak, 1984). Thus, the major protein product of the right end is likely to be RH1 because its ORF occupies the 5⬘ location in transcripts 1, 2, 3, and 5. Protein RH6 could be similarly produced from transcripts 6, 7, and 8 and may be more abundant later in the infection. Protein RH4 may be produced from transcript 4 at early times. However, RH2 and RH5 proteins are predicted to arise only as the second product from transcripts 3 and 4, respectively (Fig. 3), and may be less abundant. A unique family of F-box proteins Clues as to the function of this family of atadenovirus proteins were sought from BLAST-P searches of databases using the predicted amino acid sequence of OAdV-7 ORF RH1. Such searches confirmed the relatedness of these sequences as shown in Fig. 1 and uniqueness of these genes as a family. The searches also identified homology between the aminoterminal portion of RH1 protein and an F-box consensus motif (Fig. 4A). OAdV-7 RH2 was also listed as a member of the F-boxcontaining protein family smart00256 (including the pFam00646 subfamily) that was identified by the NCBI Conserved Domain search programme. Only two other proteins of viral origin, from Chlorella virus 1 and from baculovirus, were listed. Thus, the F-box motif is rare among viral proteins but is present in ⬎600 cellular proteins from a diverse range of plants (Arabidopsis), lower eukaryotes (e.g., Caenorhabditis elegans, yeast) (Bai et al., 1996; Patton et al., 1998; Skowyra et al., 1997) and mammalian species including humans (Cenciarelli et al., 1999; Fuchs et al., 1999; Hatakeyama et al., 1999; Winston et al., 1999). The cellular polypeptides also carry a wide range of other structural motifs that reflect their
429
diverse functions (Cenciarelli et al., 1999; Winston et al., 1999). Although BLAST-P searches did not detect F-box homology for other atadenovirus RH proteins, manual inspection of the open reading frame amino acid sequences revealed that similar motifs exist at the Nterminus of OAdV-7 RH4 and RH6, in BAdV-4 RH0, 1, 2, 4, and 6 and in DAdV-1 ORFs 8 and 9 (Figs. 1 and 4B). However, in several ORFs the first methionine falls within the F-box motif (DAdV-1 ORFs 8 and 9, OAdV-7 and BAdV-4 ORF RH4), or is well downstream (BAdV-4 ORFs RH1, 2, 6) or completely absent from the reading frame (BAdV-4 ORF RH0) (Fig. 4B). Consequently, these motifs were not previously recognized. Clearly, in these instances, if the motifs are functional, a splicing event must provide an upstream initiation codon. For BAdV-4, several initiation codons between the ITR and ORF RH0 could fulfill this function. As DAdV-1 has additional genes at this end of the genome (Fig. 1), the promoter for ORFs 8 and 9 could be located some distance away. F-box protein function The importance of F-box proteins is that they interact with key regulatory proteins that are involved in the control of a variety of cellular functions. F-box motifs, which consist of approximately 40–50 amino acids, may be located at the aminoterminus or internally, as in the Skp2 protein. According to structural data (Schulman et al., 2000) the F-box motif in Skp2 mediates an interaction with Skp1, a protein that forms part of the Skp-1, Cul-1, F-box (SCF) complex. Together with Rbx-1 and Cks1 (Spruck et al., 2001) these proteins interact with the E3 class of ubiquitin ligases in the cell (Bai et al., 1996). The F-box protein recruits a phosphorylated partner that becomes polyubiquitinated by the complex (Skowyra et al., 1997) and targeted for degradation via the 26S proteasome of the cell. The SCF system and the families of F-box proteins are thus utilized to modulate various cellular pathways by targeting key proteins for degradation. Examples include signaling cascades involving I kB␣ and -catenin (Fuchs et al., 1999; Hatakeyama et al., 1999) and proteins involved in cell-cycle control (Deshaies, 1999; Koepp et al., 1999; Moberg et al., 2001; Strohmaier et al., 2001; Yew, 2001). The lack of homology between the atadenovirus family of proteins and other F-box proteins (apart from the F-box itself) suggests two possibilities. First, that members of this family interact with a unique target(s) whose identity is presently unknown. If the partner were a key protein such as p53 or I kB␣, for example, the atadenoviruses could potentially modulate host cell metabolic pathways to create conditions favorable for virus replication or survival in vivo. Second, because of their small size relative to most other F-box proteins, it may be that these proteins simply bind to SCF complexes and block ubiquitination, thus affecting multiple metabolic pathways.
430
GERALD W. BOTH
FIG. 4. Identification of an F-box motif in atadenovirus proteins. (A) Homology between a consensus F-box motif and OAdV-7 RH1 (residues 1–45) as detected by the NCBI Conserved Domain search. Identical and similar residues are indicated with X and ( 䡠 ), respectively. Residues conserved in Skp2 (amino acids 109–152), an F-box protein whose structure has been determined in conjunction with Skp1 (Schulman et al., 2000), a component of the SCF ligase complex, are also shown. F-box residues in Skp2 that contact Skp1 in the structure are also indicated (E). (B) Alignment of the related reading frames in the three atadenoviruses using the programme SeqVu (Garvan Institute, Sydney, NSW). The GenBank sequence for BAdV-4 ORF RH6 required a single base insertion to restore the reading frame and this was verified as being present (B. Harrach, personal communication). The F-box motif is overlined and the first Met in each ORF is underscored.
ATADENOVIRUS F-BOX PROTEINS
The aminoterminal F-box motif appears to be present in all the atadenovirus proteins (Fig. 4B), but it is unclear whether they are all functional. Not all motifs were identified by the BLAST-P search. Assuming that an upstream methionine is provided by RNA splicing, there is generally strong conservation of key structural residues in the first part of the F-box motif (Fig. 4A) with more variation among the sequences in the second half of the motif. However, this pattern is consistent with the variation observed among F-box motifs in cellular proteins (see NCBI Conserved Domain search programme). At the C-terminal end of the RH protein family the sequences are very heterogenous (Fig. 4B). ORF RH1 is truncated, relative to the other ORFs in both the bovine and the ovine viruses. The other main feature is the presence of many charged residues, particularly lysines and arginines. As ubiquitin is covalently attached through its C-terminus to lysine residues (Yew, 2001), it is possible that these proteins themselves are targets for turnover by the SCF complex. The existence of these apparently redundant sequences in otherwise compact genomes is puzzling. The number of related ORFs varies from two (DAdV-1) to five in BAdV-4. It would be surprising if the genes were not transcribed and translated after appropriate splicing events. Indeed, the precedent at the transcriptional level is set by the data presented here for OAdV-7, although for ORF RH4 no transcript that provided an initiation codon upstream of the F-box was identified and this putative protein therefore has an incomplete motif. A similar situation could exist for ORF RH4 in BAdV-4 where the same methionine is conserved (Fig. 4B). However, there is clearly redundancy with respect to the function of these genes in vitro because ORFs 2–6 can be deleted without loss of virus viability (Xu et al., 1997). It is conceivable, therefore, that this region might only be of functional importance in vivo as is the mastadenovirus E3 region. Thus, the F-box protein(s) might modulate the immune response either directly or indirectly, as discussed above. The caveat is that we have not yet succeeded in rescuing a virus in which all the RH reading frames were deleted (G. W. Both, unpublished results), although this negative result should be interpreted with caution. An alternative explanation is that the region really is redundant and that during evolution related genes have been acquired multiple times, the one nearest the RH promoter being functional. The existence of spliced RNAs 6–8 that only code for ORF RH6, however, argues against this. Recently it was demonstrated that the degradation of p53 protein in cells infected by a human mastadenovirus occurred via cooperation between the adenovirus E4orf6 and E1B55k proteins and a cullin-containing complex of which Rbx1, elongins B and C, and E2 ubiquitinligase were also components (Querido et al., 2001). This was similar to the proposed von Hippel–Lindau tumor
431
FIG. 5. Schematic drawings of proposed ubiquitination ligase complexes. Top shows the SCF complex with which OAdV-7 F-box proteins may interact. The partner protein (?) is unidentified. The bottom shows how mastadenovirus E1B 55 kd and E4 orf6 proteins are thought to cooperate to degrade p53 (Querido et al., 2001).
suppressor complex. A cysteine-rich motif in E4-orf6 has been implicated in p53 protein binding and ubiquitination (Lorick et al., 1999; Nevels et al., 2000). This motif was first identified in OAdV-7 E4-orf2 and orf3 proteins (Vrati et al., 1996), which raises the question as to whether these OAdV-7 proteins, together with the E1B 55 kDa homologue (Vrati et al., 1996), might be involved in a complex of the type shown in Fig. 5 (bottom). Such complexes may target other regulatory proteins and would be distinct from the SCF E3 ubiquitin–ligase complexes with which atadenovirus F-box proteins may associate (Fig. 5, top). Although the F-box protein partner remains to be identified, these observations highlight the distinct biological strategies that appear to be used by atadenoviruses in their interactions with the host. Last, the role of the RH5 protein is unknown. The ORF is conserved in OAdV-7 and BAdV-4 but not in the avian virus DAdV-1 (Fig. 1). It is intriguing that a comparison of RH5 sequences of OAdV-7, BAdV-4 with the proposed E3 product of frog adenovirus-1 and turkey adenovirus-3, both members of the proposed siadenoviruses (Davison et al., 2000), revealed limited amino acid identity, especially six conserved cysteine residues (Fig. 6) and significant amino acid homology between the proteins (data not shown), suggesting that they may share a similar architecture and function. This relationship together with the nonessential nature of the right end of OAdV-7 would be consistent with the notion that the region is the “E3” region of atadenoviruses. The distant relationship between the sequences (Fig. 6) suggests that the genes
432
GERALD W. BOTH
sis kit (Roche Diagnostics, Mannheim, Germany). PCR primers were synthesized by Geneworks (Adelaide, South Australia). PCR reactions were carried out using 2⫻ RT-PCR mix (Promega, Madison, WI) together with appropriate cDNA samples and primers. PCR products were analyzed by agarose gel electrophoresis. Products of interest were cloned directly from PCR reactions or after fragment purification into pCR2.1-TOPO vector (Invitrogen, Groningen, The Netherlands). Nucleotide sequencing of selected clones was performed using standard primers by SUPAMAC (Camperdown, NSW). ACKNOWLEDGMENTS The author thanks Balazs Harrach, Veterinary and Medical Research Institute, Budapest for clarification in relation to ORF RH6 sequences, and David Boyle and Jason Smythe for comments on the manuscript.
REFERENCES
FIG. 6. Sequence comparison of E3 proteins from frog and turkey adenoviruses with RH5 proteins from BAdV-4 and OAdV-7. Sequences were aligned using the SeqVu programme (see legend to Fig. 4). Identical residues are boxed and shaded and conserved cysteines are highlighted with a dot.
may have been acquired independently from a common ancestor and mutated over time in their respective hosts. MATERIALS AND METHODS Cells and virus The fetal ovine lung fibroblast line CSL503 (CSL Ltd., Parkville, VIC, Australia) was grown in MEM with Earles salts (Gibco) supplemented with nonessential amino acids (0.1 mM), glutamine (2 mM), HEPES (28 mM), and bovine fetal calf serum (10%). OAdV was propagated in these cells and purified by CsCl gradient centrifugation as previously described (Boyle et al., 1994). Preparation of RNA, RT-PCR, and analysis of products CSL503 cells were infected at an m.o.i. of 1000 particles/cell (particle:infectious particle ratio, 35:1) with OAdV623, a recombinant virus that has the wild-type complement of OAdV genes (Lockett and Both, 2002). Polyadenylated RNA was prepared from infected cells at 2, 6, 12, and 24 h p.i. and at 6 and 24 h from uninfected cells using a kit (Amersham Pharmacia Biotec, Piscataway, NJ) and a protocol provided by the supplier. A portion of each sample was reverse transcribed into cDNA using oligo(dT) 15 primer and a first-strand synthe-
Bai, C., Sen, P., Hofmann, K., Ma, L., Goebl, M., Harper, J. W., and Elledge, S. J. (1996). SKP1 connects cell cycle regulators to the ubiquitin proteolysis machinery through a novel motif, the F-box. Cell 86, 263–274. Benko, M., and Harrach, B. (1998). A proposal for a new (third) genus within the family Adenoviridae. Arch. Virol. 143, 829–837. Both, G. (2001). Atadenovirus. In “The Springer Index of Viruses” (C. Tidona, and G. Darai, Eds.), pp. 11–17. Springer, Heidelberg. Boyle, D. B., Pye, A. D., Kockerhans, R., Adair, B. M., Vrati, S., and Both, G. W. (1994). Characterisation of Australian ovine adenovirus isolates. Vet. Microbiol. 41, 281–291. Cenciarelli, C., Chiaur, D. S., Guardavaccaro, D., Parks, W., Vidal, M., and Pagano, M. (1999). Identification of a family of human F-box proteins. Current Biol. 9, 1177–1179. Dan, A., Ruzsics, Z., Russell, W. C., Benko, M., and Harrach, B. (1998). Analysis of the hexon gene sequence of bovine adenovirus type 4 provides further support for a new adenovirus genus (Atadenovirus). J. Gen. Virol. 79, 1453–1460. Davison, A. J., Wright, K. M., and Harrach, B. (2000). DNA sequence of frog adenovirus. J. Gen. Virol. 81, 2431–2439. Deshaies, R. J. (1999). SCF and Cullin/Ring H2-based ubiquitin ligases. Ann. Rev. Cell Dev. Biol. 15, 435–467. Fuchs, S. Y., Chen, A., Xiong, Y., Pan, Z. Q., and Ronai, Z. (1999). HOS, a human homolog of Slimb, forms an SCF complex with Skp1 and Cullin1 and targets the phosphorylation-dependent degradation of IkappaB and beta-catenin. Oncogene 18, 2039–2046. Harrach, B., and Benko, M. (1999). Phylogenetic analysis of adenovirus sequences—Proof of the necessity of establishing a third genus in the Adenoviridae family. In “Adenovirus Methods And Protocols” (W. S. M. Wold, Ed.), Vol. 21, pp. 309–339. Humana Press Inc., Totowa, NJ. Hatakeyama, S., Kitagawa, M., Nakayama, K., Shirane, M., Matsumoto, M., Hattori, K., Higashi, H., Nakano, H., Okumura, K., Onoe, K., Good, R. A., and Nakayama, K. I. (1999). Ubiquitin-dependent degradation of IkappaBalpha is mediated by a ubiquitin ligase Skp1/Cu1 1/F-box protein FWD1. Proc. Natl. Acad. Sci. USA 96, 3859–3863. Hess, M., Blocker, H., and Brandt, P. (1997). The complete nucleotide sequence of the egg drop syndrome virus: An intermediate between mastadenoviruses and aviadenoviruses. Virology 238, 145–156. Khatri, A., and Both, G. W. (1998). Identification of transcripts and promoter regions of ovine adenovirus OAV287. Virology 245, 128–141. Koepp, D. M., Harper, J. W., and Elledge, S. J. (1999). How the cyclin became a cyclin: Regulated proteolysis in the cell cycle. Cell 97, 431–434.
ATADENOVIRUS F-BOX PROTEINS Kozak, M. (1984). Selection of initiation sites by eukaryotic ribosomes: Effect of inserting AUG triplets upstream from the codon sequence for preproinsulin. Nucleic Acids Res. 12, 3873–3893. Lockett, L., and Both, G. (2002). Complementation of a defective human adenovirus by an otherwise incompatible ovine adenovirus recombinant carrying a functional E1A gene. Virology 294, 333–341. Lorick, K. L., Jensen, J. P., Fang, S., Ong, A. M., Hatakeyama, S., and Weissman, A. M. (1999). RING fingers mediate ubiquitin-conjugating enzyme (E2)-dependent ubiquitination. Proc. Natl. Acad. Sci. USA 96, 11364–11369. Moberg, K., Bell, D., Wahrer, D., Haber, D., and Hariharan, I. (2001). Archipelago regulates cyclin E levels in Drosophila and is mutated in human cancer cell lines. Nature 413, 311–316. Nevels, M., Rubenwolf, S., Spruss, T., Wolf, H., and Dobner, T. (2000). Two distinct activities contribute to the oncogenic potential of the adenovirus type 5 E4orf6 protein. J. Virol. 74, 5168–5181. Patton, E. E., Willems, A. R., Sa, D., Kuras, L., Thomas, D., Craig, K. L., and Tyers, M. (1998). Cdc53 is a scaffold protein for multiple Cdc34/ Skp1/F-box protein complexes that regulate cell division and methionine biosynthesis in yeast. Genes Dev. 12, 692–705. Querido, E., Blanchette, P., Yan, Q., Kamura, T., Morrison, M., Boivin, D., Kaelin, W. G., Conaway, R. C., Conaway, J. W., and Branton, P. E. (2001). Degradation of p53 by adenovirus E4orf6 and E1B55K proteins occurs via a novel mechanism involving a Cullin-containing complex. Genes Dev. 15, 3104–3117. Schulman, B. A., Carrano, A. C., Jeffrey, P. D., Bowen, Z., Kinnucan, E. R.,
433
Finnin, M. S., Elledge, S. J., Harper, J. W., Pagano, M., and Pavletich, N. P. (2000). Insights into SCF ubiquitin ligases from the structure of the Skp1-Skp2 complex. Nature 408, 381–386. Skowyra, D., Craig, K. L., Tyers, M., Elledge, S. J., and Harper, J. W. (1997). F-box proteins are receptors that recruit phosphorylated substrates to the SCF ubiquitin-ligase complex. Cell 91, 209–219. Spruck, C., Strohmaier, H., Watson, M., Smith, A. P., Ryan, A., Krek, T. W., and Reed, S. I. (2001). A CDK-independent function of mammalian Cks1: Targeting of SCF(Skp2) to the CDK inhibitor p27Kip1. Mol. Cell. 7, 639–650. Strohmaier, H., Spruck, C., Kaiser, P., Won, K.-A., Sangfelt, O., and Reed, S. (2001). Human F-box protein hCdc4 targets cyclin E for proteolysis and is mutated in a breast cancer cell line. Nature 413, 316–322. Vrati, S., Brookes, D. E., Strike, P., Khatri, A., Boyle, D. B., and Both, G. W. (1996). Unique genome arrangement of an ovine adenovirus: Identification of new proteins and proteinase cleavage sites. Virology 220, 186–199. Winston, J. T., Koepp, D. M., Zhu, C., Elledge, S. J., and Harper, J. W. (1999). A family of mammalian F-box proteins. Current Biol. 9, 1180– 1182. Xu, Z. Z., Hyatt, A., Boyle, D. B., and Both, G. W. (1997). Construction of ovine adenovirus recombinants by gene insertion or deletion of related terminal region sequences. Virology 230, 62–71. Yew, P. R. (2001). Ubiquitin-mediated proteolysis of vertebrate G1- and S-phase regulators. J. Cell. Physiol. 187, 1–10.
Identification of a Unique Family of F-Box Proteins in Atadenoviruses Gerald W. Both 1 CSIRO Molecular Science, North Ryde, New South Wales, Australia 2113 Received June 3, 2002; returned to author for revision July 25, 2002; accepted August 5, 2002 Ovine adenovirus isolate 287 (OAdV-7) is the prototype of the atadenoviruses, a genus whose strategy for infection and replication is still being elucidated. A transcription unit at the right end of the genome contains four related genes (ORFs RH1, 2, 4, and 6), at least three of which are nonessential for replication in vitro. Related genes are also present in the genomes of bovine and avian atadenoviruses. To investigate how these apparently redundant genes are decoded, a more detailed transcription map of the right end of the OAdV-7 genome has been deduced. Eight transcripts that were derived from a promoter in the terminal repeat sequence were identified. Five were potentially bicistronic. The transcripts could encode all the potential proteins of the region subject to efficient reinitiation of translation. However, the most interesting and surprising finding in this work was that the related RH proteins carry an F-box motif. This was first identified in OAdV-7 RH1 and subsequently found in RH2, 4, and 6 proteins and the related reading frames from the bovine and avian atadenoviruses. Although very rare among viral proteins, several hundred cellular proteins contain F-box motifs. F-box proteins facilitate the degradation of a variety of important regulatory proteins via SCF ubiquitin ligase complexes. Thus, it appears that atadenoviruses have adopted a strategy to regulate a key cellular pathway(s) that distinguishes them from the other adenovirus genera and from most viruses in general. © 2002 Elsevier Science (USA)
INTRODUCTION
reading frames (ORFs 1, 2, 4, and 6) that were first recognized in OAdV-7 by their high level of amino acid identity (Fig. 1) (Xu et al., 1997). This relatedness is conserved across viruses for ORFs 8 and 9 from DAdV-1 (Hess et al., 1997; Khatri and Both, 1998) and for ORFs RH0, 1, 2, 4, and 6 of BAdV-4 (GenBank Accession No. NC002685) (Fig. 1). This was not previously noticed for ORF RH0 because it lacks an initiating methionine. In OAdV-7, deletion of ORFs RH2–6 is not lethal (Xu et al., 1997), suggesting that some of these ORFs are redundant, at least in vitro. In addition, the short ORF RH3 is not conserved in BAdV-4 and ORF RH5 is not present in DAdV-1. These observations raised questions as to whether each of the ORFs could actually code for a protein. In this work a more detailed transcription map of the region has been elucidated to investigate coding potential. Additional computer searches have also provided an important clue as to the role of these proteins. The indications are that atadenoviruses have a very different strategy for modulating the host cell compared with other adenoviruses, and indeed, most known viruses.
Adenoviruses have recently been classified into at least three genera based on phylogenetic analyses of the hexon and protease proteins and the distinct genome arrangements of members of the mast-, avi-, and atadenoviruses (Benko and Harrach, 1998; Dan et al., 1998; Harrach and Benko, 1999). Members of the atadenoviruses, so named because of the high A/T content of their genomes, include the prototype ovine adenovirus type 7 (OAdV-7), Egg Drop syndrome virus (DAdV-1), and bovine adenovirus type 4 (BAdV-4), all of which have been completely sequenced. Other viruses from cattle (BAdV-7 and -8), black-tailed deer, goats, and possum are also included in the genus but are incompletely characterized (Both, 2001). While the central core of the characterized genomes encodes the DNA replicating enzymes and structural protein, not all capsid proteins are conserved between the mast- and atadenoviruses (Vrati et al., 1996). Emphasizing their distinct nature, the left and right ends of the atadenovirus genomes also lack strong homologues of the mastadenovirus E1A/B and E4 regions and instead carry genes that are unique to the atadenoviruses and sometimes to the individual virus. The functions of these genes have yet to be determined. Of particular interest in this work is a family of four open
RESULTS AND DISCUSSION Transcription at the right end of the OAdV-7 genome The right end of the genome is transcribed in a right to left direction. To identify proteins that could be produced from the right end of the OAdV-7 genome, RT-PCR was used to identify transcripts and their temporal relationship. RNA harvested from infected cells at 2, 6, 12, and
1 To whom correspondence and reprint requests should be addressed at: CSIRO Molecular Science, P.O. Box 184, North Ryde, NSW 1670, Australia. Fax: 61 2 9490 5005. E-mail: [email protected].
425
0042-6822/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved.
426
GERALD W. BOTH
FIG. 1. Comparison of the right ends of the genomes for OAdV-7, BAdV-4, and DAdV-1. Related reading frames within and across virus species are indicated by similar cross-hatching. The relative location and sense of primers GB1-6 and MF6, 7 used for RT-PCR is indicated with arrowheads. E4 1, 2, and 3 refer to the ORFs for the proposed E4 region of OAdV-7.
24 h postinfection (p.i.) and from uninfected cells was reverse transcribed and cDNAs were amplified using appropriate primer pairs (Table 1). Primer pairs GB1/5 (Fig. 2, top) produced two PCR products that were amplified from infected but not uninfected cells, in agreement with an earlier study (Xu et al., 1997) and these were detectable at 6, 12, and 24 h p.i. (lanes 3, 5, and 7), but not at 2 h p.i. (lane 2). Extending the number of PCR cycles from 33 to 40 did not change this profile (data not shown). Cloning and sequencing of these products confirmed their identity and splice junctions (Xu et al., 1997)
(Table 2) and their derivation from transcripts 1 and 2 (Fig. 3). Primers GB3/5 amplified three virus-specific products that were sequenced, the largest of which (⬃1580 bp, Fig. 2, top, lane 13) was shown by sequencing to be unspliced and therefore probably derived from traces of viral DNA that contaminated the mRNA extracted at 24 h p.i. The smaller ⬃1.4-kb product was evident by 6 h p.i. (lanes 9 and 11) and was derived from transcript 3 that was contiguous with the genome, except for an intron that was spliced out between ORFs RH1 and RH2 (Fig. 3). This splicing event was also identified by
TABLE 1 Primers Used for RT-PCR and Their Location in the OAdV Genome Primer GB1 a GB2 GB3 GB4 GB5 a GB6 MF6 MF7 a
Sequence
Location
CGGGATCCATACAGGCCAAGCATCA (⫹) CGGATCCAACTTTGTGAGCAGGCC (⫹) CGGGATCCAACCGCATTCAGGCCAA (⫹) CCGGATCCAACTGATTTAACTGCCC (⫹) GCCGGATCCGGTATGATTGCTTTTCTTCC (⫺) CGGGATCCTCCCAGTGGGTGTGGC (⫺) ACT CAT CAG GAC AAC GGC ATT TAA G (⫹) GCATTTTAAGCTCTCTGCTTCAC (⫹)
RH6 3⬘ end RH5 3⬘ end RH4 3⬘ end RH2 3⬘ end RH1 5⬘ end Adjacent to ITR RH5 3⬘ end RH5 3⬘ end
Primers GB1 and GB5 are identical to PCR1 and PCR2 (Xu et al., 1997).
ATADENOVIRUS F-BOX PROTEINS
427
FIG. 2. Products derived by RT-PCR. RNA prepared from infected (I) and uninfected (U) CSL503 cells at various times postinfection were reverse transcribed and amplified by PCR with the primer pairs indicated. Species indicated with a dot were cloned and sequenced to determine transcript identity and to identify RNA splice junctions.
the products of GB4/GB5 and GB4/GB6 (data not shown). The minor ⬃1-kb species (lane 13) was not evident until 24 h p.i. and resulted from the loss of an additional intron that spliced sequences within ORF RH2 to a site at the 5⬘ end of ORF RH4 (Table 2) (Fig. 3, transcript 5). No splicing event that created an ORF RH2-specific transcript was identified. The promoter that produces right-end transcripts probably lies within the ITR of the genome (Xu et al., 1997). Primer GB6, which was located adjacent to the right ITR at nucleotides 61–46 of the genome, was used in conjunction with downstream primers GB1, 2, or 3 to determine whether additional transcripts could be amplified.
PCR products were detected at 12–24 h postinfection with each combination (Fig. 2, bottom, lanes 5, 7, 9, and 11), confirming that the promoter included the ITR sequence. Nucleotide sequencing of these products revealed a major splice junction that was located two nucleotides upstream of the methionine for ORF RH1. Splicing occurred from this point to different sites within ORFs RH4 (transcript 4), RH5 (transcripts 7 and 8), and RH6 (transcript 6) (Table 2), thus eliminating ORF RH1 coding sequences (Fig. 3). In addition, GB2/GB5 and GB2/GB6 combinations were used to look for transcripts that might encode the ORF RH5 reading frame. However, only short PCR prod-
428
GERALD W. BOTH TABLE 2 Splice Sites in Transcripts from the Right-Hand End Promoter
Transcript 1 2 3 4 5 6 7 8 a
Primer pair GB GB GB GB GB GB GB GB
1/5 1/5 3/5 3/6 3/5 1/6 1/6 2/6
First intron 5⬘ splice site a
First intron 3⬘ splice site a
Second intron 5⬘ splice site a
Second intron 3⬘ splice site a
TTTAAACG⬘GTGAGT TTTAAACG⬘GTGAGT TTTAAACG⬘GTGAGT CTTCTACG⬘GTATGA TTTAAACG⬘GTGAGT CTTCTACG⬘GTATGA CTTCTACG⬘GTATGA CTTCTACG⬘GTATGA
GTTGTTAG⬘TTTCTT TTTTACAG⬘GTCATG TTTAATAG⬘GAAGAA TTTTGTAG⬘GCGCCT TTTAATAG⬘GAAGAA TTTTACAG⬘GTCATG GTTGTTAG⬘TTTCTT GTTTTAAG⬘AGAAGA
na na na na GTGGTCTG⬘GTTGGT na AAAGTTTT⬘GTAAGT na
na na na na TTTTGTAG⬘GCGCCT na TTTTACAG⬘GTCATG na
GenBank Accession No. U40839.
ucts that were derived from transcripts 1, 7, or 8 were obtained. When ORF RH5-specific primers MF6 and MF7 from the intron region were used in conjunction with GB6 (Fig. 3), a larger product of ⬃800 bp that was visible at 6 h but not 2 h p.i. was obtained (Fig. 2, lanes 13–15). Sequencing revealed that this product was derived from transcript 4 (Fig. 3). Thus, eight transcripts from the region were identified (Fig. 3). No ORF RH3-specific transcript was identified and as this ORF is not con-
served in the closely related BAdV-4 genome (Fig. 1) it is unlikely to encode a genuine viral protein. It is likely that all transcripts terminate at the polyadenylation signal that was mapped just beyond the ORF for RH6 (Xu et al., 1997). Transcript four, at least, reads through a potential poly(A) signal that lies within ORF RH5. The failure to identify longer PCR products with primers GB1 and GB5 or 6 was probably due to the preferential amplification of short PCR products under the conditions used.
FIG. 3. Transcription map of the right end of the OAdV-7 genome. The black box to the right of each transcript indicates the promoter within the ITR. Numbers with short tick marks above transcripts refer to primers also shown in Fig. 1. Bold and broken lines indicate sequenced and unsequenced regions of transcripts 1–8, respectively. Thin lines are introns. (*) Polyadenylation site. (M) Signifies the first methionine codon within an ORF. (ˆ) Translation stop codons. M is 11 and 6 nucleotides downstream of ˆ in spliced transcripts 3 and 4, respectively. (#) Methionine codon in ORF RH4 is within the F-box motif. (m) Splicing truncates and frameshifts reading frame for ORF RH2. (Hr pi) Signifies when PCR products were first detected. Product signifies possible protein(s) encoded.
ATADENOVIRUS F-BOX PROTEINS
Proteins from the right end of the OAdV-7 genome The eight transcripts could code for proteins RH1, 2, 4, 5, and 6. Transcripts 1, 2, 3 (detectable at ⬃6 h p.i.), and 5 code for protein RH1 which begins at the first methionine in the genome at that end. RH2 and RH6 proteins could also be produced from transcripts 3 and 2, respectively (Fig. 3), by reinitiation of translation because the initiation codons for these proteins occur only 11 and 7 nucleotides downstream from the terminator for ORF RH1 (Kozak, 1984). An extended form of RH6 could also be produced from transcript 1 using an upstream initiation codon (Fig. 3) but this is not conserved in the BAdV-4 RH6 homologue (Fig. 4). At later times p.i. (⬃24 h), modified splicing produces transcripts 6 and 7 that can only code for ORF RH6. Transcript 4 could produce RH4 protein and RH5 as the second product by reinitiation of translation five nucleotides downstream of the ORF RH4 stop codon (Fig. 3). Transcript 5 could produce RH1 and an aberrant form of RH2 (due to a splice and frameshift) (Fig. 3) but is unlikely to produce RH4 because translation from ORF RH2 does not terminate before the ORF RH4 initiation codon (Kozak, 1984). Thus, the major protein product of the right end is likely to be RH1 because its ORF occupies the 5⬘ location in transcripts 1, 2, 3, and 5. Protein RH6 could be similarly produced from transcripts 6, 7, and 8 and may be more abundant later in the infection. Protein RH4 may be produced from transcript 4 at early times. However, RH2 and RH5 proteins are predicted to arise only as the second product from transcripts 3 and 4, respectively (Fig. 3), and may be less abundant. A unique family of F-box proteins Clues as to the function of this family of atadenovirus proteins were sought from BLAST-P searches of databases using the predicted amino acid sequence of OAdV-7 ORF RH1. Such searches confirmed the relatedness of these sequences as shown in Fig. 1 and uniqueness of these genes as a family. The searches also identified homology between the aminoterminal portion of RH1 protein and an F-box consensus motif (Fig. 4A). OAdV-7 RH2 was also listed as a member of the F-boxcontaining protein family smart00256 (including the pFam00646 subfamily) that was identified by the NCBI Conserved Domain search programme. Only two other proteins of viral origin, from Chlorella virus 1 and from baculovirus, were listed. Thus, the F-box motif is rare among viral proteins but is present in ⬎600 cellular proteins from a diverse range of plants (Arabidopsis), lower eukaryotes (e.g., Caenorhabditis elegans, yeast) (Bai et al., 1996; Patton et al., 1998; Skowyra et al., 1997) and mammalian species including humans (Cenciarelli et al., 1999; Fuchs et al., 1999; Hatakeyama et al., 1999; Winston et al., 1999). The cellular polypeptides also carry a wide range of other structural motifs that reflect their
429
diverse functions (Cenciarelli et al., 1999; Winston et al., 1999). Although BLAST-P searches did not detect F-box homology for other atadenovirus RH proteins, manual inspection of the open reading frame amino acid sequences revealed that similar motifs exist at the Nterminus of OAdV-7 RH4 and RH6, in BAdV-4 RH0, 1, 2, 4, and 6 and in DAdV-1 ORFs 8 and 9 (Figs. 1 and 4B). However, in several ORFs the first methionine falls within the F-box motif (DAdV-1 ORFs 8 and 9, OAdV-7 and BAdV-4 ORF RH4), or is well downstream (BAdV-4 ORFs RH1, 2, 6) or completely absent from the reading frame (BAdV-4 ORF RH0) (Fig. 4B). Consequently, these motifs were not previously recognized. Clearly, in these instances, if the motifs are functional, a splicing event must provide an upstream initiation codon. For BAdV-4, several initiation codons between the ITR and ORF RH0 could fulfill this function. As DAdV-1 has additional genes at this end of the genome (Fig. 1), the promoter for ORFs 8 and 9 could be located some distance away. F-box protein function The importance of F-box proteins is that they interact with key regulatory proteins that are involved in the control of a variety of cellular functions. F-box motifs, which consist of approximately 40–50 amino acids, may be located at the aminoterminus or internally, as in the Skp2 protein. According to structural data (Schulman et al., 2000) the F-box motif in Skp2 mediates an interaction with Skp1, a protein that forms part of the Skp-1, Cul-1, F-box (SCF) complex. Together with Rbx-1 and Cks1 (Spruck et al., 2001) these proteins interact with the E3 class of ubiquitin ligases in the cell (Bai et al., 1996). The F-box protein recruits a phosphorylated partner that becomes polyubiquitinated by the complex (Skowyra et al., 1997) and targeted for degradation via the 26S proteasome of the cell. The SCF system and the families of F-box proteins are thus utilized to modulate various cellular pathways by targeting key proteins for degradation. Examples include signaling cascades involving I kB␣ and -catenin (Fuchs et al., 1999; Hatakeyama et al., 1999) and proteins involved in cell-cycle control (Deshaies, 1999; Koepp et al., 1999; Moberg et al., 2001; Strohmaier et al., 2001; Yew, 2001). The lack of homology between the atadenovirus family of proteins and other F-box proteins (apart from the F-box itself) suggests two possibilities. First, that members of this family interact with a unique target(s) whose identity is presently unknown. If the partner were a key protein such as p53 or I kB␣, for example, the atadenoviruses could potentially modulate host cell metabolic pathways to create conditions favorable for virus replication or survival in vivo. Second, because of their small size relative to most other F-box proteins, it may be that these proteins simply bind to SCF complexes and block ubiquitination, thus affecting multiple metabolic pathways.
430
GERALD W. BOTH
FIG. 4. Identification of an F-box motif in atadenovirus proteins. (A) Homology between a consensus F-box motif and OAdV-7 RH1 (residues 1–45) as detected by the NCBI Conserved Domain search. Identical and similar residues are indicated with X and ( 䡠 ), respectively. Residues conserved in Skp2 (amino acids 109–152), an F-box protein whose structure has been determined in conjunction with Skp1 (Schulman et al., 2000), a component of the SCF ligase complex, are also shown. F-box residues in Skp2 that contact Skp1 in the structure are also indicated (E). (B) Alignment of the related reading frames in the three atadenoviruses using the programme SeqVu (Garvan Institute, Sydney, NSW). The GenBank sequence for BAdV-4 ORF RH6 required a single base insertion to restore the reading frame and this was verified as being present (B. Harrach, personal communication). The F-box motif is overlined and the first Met in each ORF is underscored.
ATADENOVIRUS F-BOX PROTEINS
The aminoterminal F-box motif appears to be present in all the atadenovirus proteins (Fig. 4B), but it is unclear whether they are all functional. Not all motifs were identified by the BLAST-P search. Assuming that an upstream methionine is provided by RNA splicing, there is generally strong conservation of key structural residues in the first part of the F-box motif (Fig. 4A) with more variation among the sequences in the second half of the motif. However, this pattern is consistent with the variation observed among F-box motifs in cellular proteins (see NCBI Conserved Domain search programme). At the C-terminal end of the RH protein family the sequences are very heterogenous (Fig. 4B). ORF RH1 is truncated, relative to the other ORFs in both the bovine and the ovine viruses. The other main feature is the presence of many charged residues, particularly lysines and arginines. As ubiquitin is covalently attached through its C-terminus to lysine residues (Yew, 2001), it is possible that these proteins themselves are targets for turnover by the SCF complex. The existence of these apparently redundant sequences in otherwise compact genomes is puzzling. The number of related ORFs varies from two (DAdV-1) to five in BAdV-4. It would be surprising if the genes were not transcribed and translated after appropriate splicing events. Indeed, the precedent at the transcriptional level is set by the data presented here for OAdV-7, although for ORF RH4 no transcript that provided an initiation codon upstream of the F-box was identified and this putative protein therefore has an incomplete motif. A similar situation could exist for ORF RH4 in BAdV-4 where the same methionine is conserved (Fig. 4B). However, there is clearly redundancy with respect to the function of these genes in vitro because ORFs 2–6 can be deleted without loss of virus viability (Xu et al., 1997). It is conceivable, therefore, that this region might only be of functional importance in vivo as is the mastadenovirus E3 region. Thus, the F-box protein(s) might modulate the immune response either directly or indirectly, as discussed above. The caveat is that we have not yet succeeded in rescuing a virus in which all the RH reading frames were deleted (G. W. Both, unpublished results), although this negative result should be interpreted with caution. An alternative explanation is that the region really is redundant and that during evolution related genes have been acquired multiple times, the one nearest the RH promoter being functional. The existence of spliced RNAs 6–8 that only code for ORF RH6, however, argues against this. Recently it was demonstrated that the degradation of p53 protein in cells infected by a human mastadenovirus occurred via cooperation between the adenovirus E4orf6 and E1B55k proteins and a cullin-containing complex of which Rbx1, elongins B and C, and E2 ubiquitinligase were also components (Querido et al., 2001). This was similar to the proposed von Hippel–Lindau tumor
431
FIG. 5. Schematic drawings of proposed ubiquitination ligase complexes. Top shows the SCF complex with which OAdV-7 F-box proteins may interact. The partner protein (?) is unidentified. The bottom shows how mastadenovirus E1B 55 kd and E4 orf6 proteins are thought to cooperate to degrade p53 (Querido et al., 2001).
suppressor complex. A cysteine-rich motif in E4-orf6 has been implicated in p53 protein binding and ubiquitination (Lorick et al., 1999; Nevels et al., 2000). This motif was first identified in OAdV-7 E4-orf2 and orf3 proteins (Vrati et al., 1996), which raises the question as to whether these OAdV-7 proteins, together with the E1B 55 kDa homologue (Vrati et al., 1996), might be involved in a complex of the type shown in Fig. 5 (bottom). Such complexes may target other regulatory proteins and would be distinct from the SCF E3 ubiquitin–ligase complexes with which atadenovirus F-box proteins may associate (Fig. 5, top). Although the F-box protein partner remains to be identified, these observations highlight the distinct biological strategies that appear to be used by atadenoviruses in their interactions with the host. Last, the role of the RH5 protein is unknown. The ORF is conserved in OAdV-7 and BAdV-4 but not in the avian virus DAdV-1 (Fig. 1). It is intriguing that a comparison of RH5 sequences of OAdV-7, BAdV-4 with the proposed E3 product of frog adenovirus-1 and turkey adenovirus-3, both members of the proposed siadenoviruses (Davison et al., 2000), revealed limited amino acid identity, especially six conserved cysteine residues (Fig. 6) and significant amino acid homology between the proteins (data not shown), suggesting that they may share a similar architecture and function. This relationship together with the nonessential nature of the right end of OAdV-7 would be consistent with the notion that the region is the “E3” region of atadenoviruses. The distant relationship between the sequences (Fig. 6) suggests that the genes
432
GERALD W. BOTH
sis kit (Roche Diagnostics, Mannheim, Germany). PCR primers were synthesized by Geneworks (Adelaide, South Australia). PCR reactions were carried out using 2⫻ RT-PCR mix (Promega, Madison, WI) together with appropriate cDNA samples and primers. PCR products were analyzed by agarose gel electrophoresis. Products of interest were cloned directly from PCR reactions or after fragment purification into pCR2.1-TOPO vector (Invitrogen, Groningen, The Netherlands). Nucleotide sequencing of selected clones was performed using standard primers by SUPAMAC (Camperdown, NSW). ACKNOWLEDGMENTS The author thanks Balazs Harrach, Veterinary and Medical Research Institute, Budapest for clarification in relation to ORF RH6 sequences, and David Boyle and Jason Smythe for comments on the manuscript.
REFERENCES
FIG. 6. Sequence comparison of E3 proteins from frog and turkey adenoviruses with RH5 proteins from BAdV-4 and OAdV-7. Sequences were aligned using the SeqVu programme (see legend to Fig. 4). Identical residues are boxed and shaded and conserved cysteines are highlighted with a dot.
may have been acquired independently from a common ancestor and mutated over time in their respective hosts. MATERIALS AND METHODS Cells and virus The fetal ovine lung fibroblast line CSL503 (CSL Ltd., Parkville, VIC, Australia) was grown in MEM with Earles salts (Gibco) supplemented with nonessential amino acids (0.1 mM), glutamine (2 mM), HEPES (28 mM), and bovine fetal calf serum (10%). OAdV was propagated in these cells and purified by CsCl gradient centrifugation as previously described (Boyle et al., 1994). Preparation of RNA, RT-PCR, and analysis of products CSL503 cells were infected at an m.o.i. of 1000 particles/cell (particle:infectious particle ratio, 35:1) with OAdV623, a recombinant virus that has the wild-type complement of OAdV genes (Lockett and Both, 2002). Polyadenylated RNA was prepared from infected cells at 2, 6, 12, and 24 h p.i. and at 6 and 24 h from uninfected cells using a kit (Amersham Pharmacia Biotec, Piscataway, NJ) and a protocol provided by the supplier. A portion of each sample was reverse transcribed into cDNA using oligo(dT) 15 primer and a first-strand synthe-
Bai, C., Sen, P., Hofmann, K., Ma, L., Goebl, M., Harper, J. W., and Elledge, S. J. (1996). SKP1 connects cell cycle regulators to the ubiquitin proteolysis machinery through a novel motif, the F-box. Cell 86, 263–274. Benko, M., and Harrach, B. (1998). A proposal for a new (third) genus within the family Adenoviridae. Arch. Virol. 143, 829–837. Both, G. (2001). Atadenovirus. In “The Springer Index of Viruses” (C. Tidona, and G. Darai, Eds.), pp. 11–17. Springer, Heidelberg. Boyle, D. B., Pye, A. D., Kockerhans, R., Adair, B. M., Vrati, S., and Both, G. W. (1994). Characterisation of Australian ovine adenovirus isolates. Vet. Microbiol. 41, 281–291. Cenciarelli, C., Chiaur, D. S., Guardavaccaro, D., Parks, W., Vidal, M., and Pagano, M. (1999). Identification of a family of human F-box proteins. Current Biol. 9, 1177–1179. Dan, A., Ruzsics, Z., Russell, W. C., Benko, M., and Harrach, B. (1998). Analysis of the hexon gene sequence of bovine adenovirus type 4 provides further support for a new adenovirus genus (Atadenovirus). J. Gen. Virol. 79, 1453–1460. Davison, A. J., Wright, K. M., and Harrach, B. (2000). DNA sequence of frog adenovirus. J. Gen. Virol. 81, 2431–2439. Deshaies, R. J. (1999). SCF and Cullin/Ring H2-based ubiquitin ligases. Ann. Rev. Cell Dev. Biol. 15, 435–467. Fuchs, S. Y., Chen, A., Xiong, Y., Pan, Z. Q., and Ronai, Z. (1999). HOS, a human homolog of Slimb, forms an SCF complex with Skp1 and Cullin1 and targets the phosphorylation-dependent degradation of IkappaB and beta-catenin. Oncogene 18, 2039–2046. Harrach, B., and Benko, M. (1999). Phylogenetic analysis of adenovirus sequences—Proof of the necessity of establishing a third genus in the Adenoviridae family. In “Adenovirus Methods And Protocols” (W. S. M. Wold, Ed.), Vol. 21, pp. 309–339. Humana Press Inc., Totowa, NJ. Hatakeyama, S., Kitagawa, M., Nakayama, K., Shirane, M., Matsumoto, M., Hattori, K., Higashi, H., Nakano, H., Okumura, K., Onoe, K., Good, R. A., and Nakayama, K. I. (1999). Ubiquitin-dependent degradation of IkappaBalpha is mediated by a ubiquitin ligase Skp1/Cu1 1/F-box protein FWD1. Proc. Natl. Acad. Sci. USA 96, 3859–3863. Hess, M., Blocker, H., and Brandt, P. (1997). The complete nucleotide sequence of the egg drop syndrome virus: An intermediate between mastadenoviruses and aviadenoviruses. Virology 238, 145–156. Khatri, A., and Both, G. W. (1998). Identification of transcripts and promoter regions of ovine adenovirus OAV287. Virology 245, 128–141. Koepp, D. M., Harper, J. W., and Elledge, S. J. (1999). How the cyclin became a cyclin: Regulated proteolysis in the cell cycle. Cell 97, 431–434.
ATADENOVIRUS F-BOX PROTEINS Kozak, M. (1984). Selection of initiation sites by eukaryotic ribosomes: Effect of inserting AUG triplets upstream from the codon sequence for preproinsulin. Nucleic Acids Res. 12, 3873–3893. Lockett, L., and Both, G. (2002). Complementation of a defective human adenovirus by an otherwise incompatible ovine adenovirus recombinant carrying a functional E1A gene. Virology 294, 333–341. Lorick, K. L., Jensen, J. P., Fang, S., Ong, A. M., Hatakeyama, S., and Weissman, A. M. (1999). RING fingers mediate ubiquitin-conjugating enzyme (E2)-dependent ubiquitination. Proc. Natl. Acad. Sci. USA 96, 11364–11369. Moberg, K., Bell, D., Wahrer, D., Haber, D., and Hariharan, I. (2001). Archipelago regulates cyclin E levels in Drosophila and is mutated in human cancer cell lines. Nature 413, 311–316. Nevels, M., Rubenwolf, S., Spruss, T., Wolf, H., and Dobner, T. (2000). Two distinct activities contribute to the oncogenic potential of the adenovirus type 5 E4orf6 protein. J. Virol. 74, 5168–5181. Patton, E. E., Willems, A. R., Sa, D., Kuras, L., Thomas, D., Craig, K. L., and Tyers, M. (1998). Cdc53 is a scaffold protein for multiple Cdc34/ Skp1/F-box protein complexes that regulate cell division and methionine biosynthesis in yeast. Genes Dev. 12, 692–705. Querido, E., Blanchette, P., Yan, Q., Kamura, T., Morrison, M., Boivin, D., Kaelin, W. G., Conaway, R. C., Conaway, J. W., and Branton, P. E. (2001). Degradation of p53 by adenovirus E4orf6 and E1B55K proteins occurs via a novel mechanism involving a Cullin-containing complex. Genes Dev. 15, 3104–3117. Schulman, B. A., Carrano, A. C., Jeffrey, P. D., Bowen, Z., Kinnucan, E. R.,
433
Finnin, M. S., Elledge, S. J., Harper, J. W., Pagano, M., and Pavletich, N. P. (2000). Insights into SCF ubiquitin ligases from the structure of the Skp1-Skp2 complex. Nature 408, 381–386. Skowyra, D., Craig, K. L., Tyers, M., Elledge, S. J., and Harper, J. W. (1997). F-box proteins are receptors that recruit phosphorylated substrates to the SCF ubiquitin-ligase complex. Cell 91, 209–219. Spruck, C., Strohmaier, H., Watson, M., Smith, A. P., Ryan, A., Krek, T. W., and Reed, S. I. (2001). A CDK-independent function of mammalian Cks1: Targeting of SCF(Skp2) to the CDK inhibitor p27Kip1. Mol. Cell. 7, 639–650. Strohmaier, H., Spruck, C., Kaiser, P., Won, K.-A., Sangfelt, O., and Reed, S. (2001). Human F-box protein hCdc4 targets cyclin E for proteolysis and is mutated in a breast cancer cell line. Nature 413, 316–322. Vrati, S., Brookes, D. E., Strike, P., Khatri, A., Boyle, D. B., and Both, G. W. (1996). Unique genome arrangement of an ovine adenovirus: Identification of new proteins and proteinase cleavage sites. Virology 220, 186–199. Winston, J. T., Koepp, D. M., Zhu, C., Elledge, S. J., and Harper, J. W. (1999). A family of mammalian F-box proteins. Current Biol. 9, 1180– 1182. Xu, Z. Z., Hyatt, A., Boyle, D. B., and Both, G. W. (1997). Construction of ovine adenovirus recombinants by gene insertion or deletion of related terminal region sequences. Virology 230, 62–71. Yew, P. R. (2001). Ubiquitin-mediated proteolysis of vertebrate G1- and S-phase regulators. J. Cell. Physiol. 187, 1–10.