Previous Article | Next Article ![]()
Journal of Virology, May 2005, p. 6505-6510, Vol. 79, No. 10
0022-538X/05/$08.00+0 doi:10.1128/JVI.79.10.6505-6510.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Department of Molecular Microbiology and Immunology, University of MissouriColumbia, Columbia, Missouri 65212,1 Department of Virology, Haartman Institute and Helsinki University Central Hospital, POB 21, FIN-00014 University of Helsinki, Helsinki, Finland2
Received 13 August 2004/ Accepted 4 January 2005
|
|
|---|
|
|
|---|
Three species of mRNAs (2.8, 1.2, and 1.0 kb) (Fig. 1A) have been detected in infected bone marrow cells (14) and are generated following transfection of a full-length TTV clone into monkey COS cells (7), which support the expression but not the replication of TTV. Antibodies in TTV-infected individuals that react with two proteins encoded by the 2.8-kb mRNA have been detected (2, 17). The first is the large protein encoded by TTV ORF1 (the ORF1 protein [Fig. 1C]), which is predicted to be 736 amino acids in length and initiate from a methionine at nucleotide (nt) 581 (O1AUG). (All nucleotide numbers refer to GenBank accession number AY666122.) The second protein (the ORF2 protein [Fig. 1C]) is encoded from the 2.8-kb mRNA in ORF2, initiating at a methionine at nt 354 (O2AUG) and extending for 117 amino acids.
![]() View larger version (36K): [in a new window] |
FIG. 1. Determination of the genetic map of TTV genotype 6 by analysis of TTV transcripts in 293 cells. (A) Northern blot analysis of TTV RNA. mRNA purified from 20 µg of total RNA, isolated 48 h after transfection of the TTV6 full-length clone pTV6 (diagramed on top) into 293 cells (lane 1), following transfection of the parvovirus B19 full-length plasmid in COS cells to serve as a size marker (lane 2), or mock transfected (lane 3), was run on a 1% formaldehyde-agarose gel, transferred to nitrocellulose, and probed with either the full-length TTV genome (lanes 1 and 3) or the full-length B19 clone. Three major species of TTV mRNAs of approximately 2.8, 1.2, and 1.0 kb are shown. (B) Determination of the TTV genetic map by RNase protection assay. The TTV6 genome and the probes used in this study are schematically diagramed. Positions of the predicated promoter, the intron donors (1D and 2D) and acceptors (1A, 2A1, and 2A2), and the polyadenylation site [p(A)] are shown. The locations of the P1D (nt's 44 to 247), P1A (nt's 207 to 465), P2D (nt's 546 to 785), P2A (nt's 2290 to 2648), and P(A)n (nt's 2866 to 3105) probes described in this study are shown. The expected bands for each probe are depicted below each probe. Ten micrograms of total RNA isolated 48 h following transfection of pTV6 into 293 cells was subjected to RNase protections using the different probes [P1D, P1A, P2D, P2A, and P(A)n] across the promoter, intron, and polyadenylation regions, as indicated. A 32P-labeled RNA ladder (21) with the respective sizes indicated to the left is presented in lane 1. The origins of the protected bands in lanes 2 to 6 are indicated and discussed further in the text. Spl denotes spliced species, and Unspl denotes unspliced species. An arrow denotes undigested probe. The genetic map of TTV6 was deduced from the size of protected bands and confirmed by reverse transcription-PCR and sequencing results (as discussed in the text). (C) Summary of the TTV6 genetic map and expression strategy. The genetic map of TTV6 is shown on the top. The initiation site, splice junctions, and polyadenylation cleavage sites are shown with nucleotide numbers as indicated. Six potential ORF expression strategies are diagramed in the map below. The ORF1 and ORF2 proteins are encoded from the singly spliced 2.8-kb mRNA by alternative translation, using O1AUG (nt 581) and O2AUG (nt 354), respectively. The ORF2/2 and ORF2/3 proteins, which are initiated at the O2AUG, are translated from the doubly spliced 1.2- and 1.0-kb mRNAs, respectively. ORF1/1 and ORF1/2 proteins are predicted to be encoded by the 1.2- and 1.0-kb mRNAs, respectively, using the O1AUG at nt 581. The different reading frames are indicated by different fill patterns.
|
The complete sequence of TTV genotype 6 (6) (including duplication of the region between nt 3205 and 3380 [Fig. 1A]) was originally cloned into pSTBlueAccepTor (Novagen) (L. Kakkola, J. Tommiska, L. Boele, S. Miettinen, T. Blom, T. Kekarainen, R. Hoeben, K. Hedman, and M. Söderlund-Venermo, unpublished data). For ease of manipulation, pTV6 was created by inserting the complete cloned TTV6 genome into the EcoRI site of pBluescipt SK+ (Stratagene). All subsequent TTV6 plasmids used in this study were derivatives of pTV6.
Determination of the RNA profile of TTV following transfection of 293 cells. Northern blot analysis, using the full-length TTV6 genome as probe in procedures previously described (20), of TTV RNA generated following transfection of pTV6 into 293 cells using the Lipofectamine and Plus reagent (22) confirmed the presence of three mRNAs that were 2.8, 1.2, and 1.0 kb in size (Fig. 1A), as expected, and similar to the profiles previously reported for TTV genotype 1 in COS cells (7) and in TTV-infected human bone marrow cells (14). The 2.8-, 1.2-, and 1.0-kb RNAs comprised approximately 60, 5, and 35% of the total TTV RNA, respectively.
RNase protection assays, performed essentially as previously described (23), were used to more specifically map and to more quantitatively analyze the abundance of the individual TTV RNAs generated in 293 cells following transfection of pTV6. The probes used for these analyses were generated in vitro as previously described (22, 23) and are shown in Fig. 1B.
Probe P1D, which spans the putative promoter and the putative donor site (1D) of the small first intron, protected bands of approximately 135 and approximately 70 nt's in length. These bands represented RNAs which were unspliced and spliced, respectively, across the first intron donor region (Fig. 1B, lane 2). They mapped the initiation site of TTV genotype 6 RNA to nt 112 and the donor site to nt 182. Approximately 95% of TTV mRNAs detected in this assay were spliced at the small intron donor.
Probe P1A, which spans the putative acceptor of the first intron, protected bands of approximately 258 and 182 nt's (Fig. 1B, lane 3), which represented unspliced and spliced forms, respectively, of RNAs spanning the first intron acceptor region. Consistent with the results using probe P1D, greater than 95% of TTV RNAs were spliced at the small intron acceptor at nt 284, confirming that the great majority of TTV RNA is spliced in this region.
Probe P2D, which spans the putative donor site of the large second intron, protected bands of approximately 239 and 157 nt's (Fig. 1B, lane 4), which represented RNAs that were unspliced and spliced, respectively, through this region. These bands mapped the second donor site at nt 703. The relative ratio of unspliced RNAs to RNAs spliced at the first donor site was approximately 1:1.
Probe P2A, which spans the putative acceptors of the large second intron, protected bands of approximately 358, 334, and 144 nt's (Fig. 1B, lane 5), which represented unspliced RNA, RNAs spliced at acceptor 2A1, and RNAs spliced at acceptor 2A2, respectively. The 2A1 acceptor site mapped to nt 2315, and the 2A2 site mapped to nt 2505. The relative ratio of unspliced RNAs to total spliced RNAs through this region was approximately 1:1, which is consistent with the ratio obtained using probe P2D, which detected usage of the large intron donor (Fig. 1B, lane 4). The ratio of RNAs spliced at 2A1 relative to RNAs spliced at 2A2 is approximately 1:10, which confirmed that the levels of RNA spliced at 2A1 is low and is consistent with the relative levels of the 1.2- and 1.0-kb RNA detected by Northern blot analysis (Fig. 1A, lane 1).
Probe P(A)n, which spans the putative polyadenylation site, protected a band of approximately 134 nt's, which mapped the RNA cleavage site to approximately nt 3000 (Fig. 1B, lane 6). Although the polyadenylation signal (AUUAAA) at nt 2978 for TTV6 is nonconsensus, similar to the internal polyadenylation site (pA)p of human parvovirus B19 (18), it was apparently used very efficiently.
Sequencing of the products of 5' rapid amplification of cDNA ends and 3' rapid amplification of cDNA ends experiments confirmed that the TTV genotype 6 RNA initiation site is at nt 112 and the 3' cleavage site is at nt 3000 (data not shown). All of the splice junctions (1D, 1A, 2D, 2A1, and 2A2) also were confirmed by reverse transcription-PCR and subsequent sequencing (data not shown). The genetic organization of TTV genotype 6 mRNA, which has the potential capability to encode at least six proteins, is summarized in Fig. 1C.
Protein expression strategy of TTV mRNAs. Because a convenient tissue culture system to study TTV replication has not yet been developed, no clear consensus on the number, abundance, and subcellular localization of proteins that TTV produces during infection has yet emerged.
Detection of HA-tagged TTV proteins following transfection of a full-length clone of TTV. To characterize the expression of TTV proteins for which there are no available antibodies, we first inserted the 9-amino-acid hemagglutinin (HA) tag YPYDVPDYA (5'TAC CCA TAC GAT GTT CCA GAT TAC GCT3') in frame into pTV6 in the carboxyl terminus of either ORF1 (at nt 2788), ORF2 (at nt 2807), or ORF3 (at nt 2979) to generate TVf1HA, TVf2HA, and TVf3HA, respectively (Fig. 2A). TV1AUGHA was made by insertion of the HA sequence in ORF2 immediately after the O2AUG at nt 357 (Fig. 2B). The RNA transcription profile following transfection of these TTV plasmids was indistinguishable from the parent pTV6 (data not shown). Expression of these constructs following transfection into 293 cells was assayed by Western blotting using anti-HA antibody, as previously described (9).
![]() View larger version (26K): [in a new window] |
FIG. 2. Western blot analysis of HA-tagged TTV proteins. (A) Western blots of protein samples isolated 48 h following transfection of 293 cells with constructs TVf1HA, TVf2HA, and TVf3HA, in which an HA tag was inserted into parent pTV6 into the carboxyl terminus of ORF1, ORF2, and ORF3. Samples were run on 10% polyacrylamide gels, and Western blots were performed using a monoclonal antibody against the HA tag (HA-7; Sigma). Lane 1, TVf1HA generated HA-tagged ORF1/1 at a size of approximately 22 kDa (however, HA-tagged ORF1 was not detected [see text]); lane 2, TVf2HA generated both HA-tagged ORF2/2 and HA-tagged ORF1/2 at the expected sizes of approximately 31 and approximately 16 kDa, respectively; lane 3, TVf3HA generated HA-tagged ORF2/3 at the expected size of approximately 31 kDa (the amount of protein sample loaded in lane 3 was 10% of that loaded in lanes 1 and 2); and lane 4, untagged parental pTV6 generated no protein detected with anti-HA antibody. (B) Western blots of protein samples isolated 48 h following transfection of 293 cells with construct TV2AUGHA, in which an HA tag was inserted into parent pTV6 into ORF2 immediately downstream of the O2AUG at nt 354. Samples were run on 12.5% polyacrylamide gels, and Western blotting was performed using a monoclonal antibody against the HA tag (HA.11; Covance, Berkeley, CA). Lane 1, untagged parental pTV6 generated no protein detectable with the anti-HA antibody; lane 2, TV2AUGHA generated HA-tagged ORF2, running at an apparent size of approximately 11 kDa, which is slightly faster than predicted, and an apparent doublet running at the expected sizes for the HA-tagged ORF2/2 and ORF2/3 proteins, which are also detected in panel A.
|
An HA tag was also placed at the N terminus of ORF2, immediately after the O2AUG initiation codon at nt 354 (TV2AUGHA). This construct expressed a protein with a size consistent with the small ORF2 protein (predicted to be 117 amino acids and initiated at the O2AUG) and exhibited a band that likely comprised both the ORF2/2 and ORF2/3 proteins (Fig. 2B, lane 2), which were individually detected in the analysis shown in Fig. 2A, as discussed above.
Thus, all predicted TTV proteins, except the large ORF1 protein, were detected by Western blotting following transfection of the full-length clone of TTV. The ORF1 protein, however, was detected by immunofluorescence as described below. Our analysis demonstrated for the first time the presence of the TTV ORF1/1 and ORF1/2 proteins, initiated from the O1AUG from the spliced 1.2- and 1.0-kb mRNAs, respectively.
The 1.0-kb spliced mRNAs, which use the 2A2 splice acceptor and encode the ORF2/3 and ORF1/2 proteins, are approximately 10 times more abundant than the 1.2-kb spliced mRNAs that use the 2A1 acceptor and encode the ORF2/2 and ORF1/1 proteins (Fig. 1A). Thus, the ORF2/3 protein would be expected to be more abundant than the ORF2/2 protein, and the ORF1/2 protein would be expected to be more abundant than the ORF1/1 protein. Because the transcription maps of those different TTV isolates so far examined are similar (7, 14), it is likely that this holds true for the other genotypes of TTV as well.
In the absence of additional regulatory signals, 5'-proximal initiating AUGs are often translated at greater efficiency than those AUGs downstream, and this may contribute to the low levels of expression of TTV proteins initiated at the O1AUG (which is the fourth AUG in this mRNA) compared to those initiated at the O2AUG (which is the first AUG in this mRNA). However, the sequence surrounding the O1AUG site in TTV genotype 6 also deviates considerably from previously defined consensus sequences (6, 8), whereas the analogous region of the prototype TTV genotype 1 (TA278) conforms to the typical consensus signal (15). Thus, the O1AUG-initiated proteins (ORF1, ORF1/1, and ORF1/2) may be expressed at an even lower level from TTV genotype 6 than from other TTV genotypes.
Subcellular localization of TTV proteins. The subcellular localization of HA-tagged TTV proteins in fixed cells and green fluorescent protein (GFP)-tagged TTV proteins in living cells expressed following transfection was determined. The two detection strategies yielded similar results. For each type of tagging, constructs that were capable of expression of only one of the six TTV proteins were generated (as described in the supplemental material). Constructs were verified by both RNase protection assays where applicable (data not shown) and Western blot analyses (except for the ORF1 protein) to ensure that they encoded only the predicted TTV protein (see the supplemental material).
Initially, the localizations of the individual HA-tagged proteins were examined by immunofluorescence of fixed samples using a fluorescein isothiocyanate (FITC)-conjugated monoclonal anti-HA antibody (Sigma, St. Louis, MO) and counterstained with propidium iodide (P-3566; Molecular Probes, Inc., Eugene, OR), as described by the manufacturer, following transfection of individual constructs in 293 cells (Fig. 3A). Similar results were seen with GFP-tagged proteins analyzed directly by fluorescence microscopy in living cells following addition of SYTO-59 dye (Molecular Probes, Inc.) (9) (Fig. 3B). Three patterns of localization were observed. As can be seen in Fig. 3, the ORF1 and ORF2 proteins were predominately localized in the cytoplasm, perhaps consistent with a role as a structural component. Although the ORF1 protein was not detectable in Western assays, it was detectable (although at levels lower than the other TTV proteins) by both immunofluorescence in fixed cells as well as GFP detection in live cells. While the cytoplasmic localization of the ORF1 protein would be consistent with a role as a structural protein, it would be surprising for a structural protein to be present at levels below detection by Western blotting (Fig. 2A). The ORF2/2 and ORF2/3 proteins were exclusively localized to discrete foci in the nucleus, perhaps consistent with a role in genome replication and expression. The ORF1/1 and ORF1/2 proteins were distributed evenly between the cytoplasm and nuclei. Surprisingly, the three proteins which were initiated at the O1AUG (the ORF1, ORF1/1, and ORF1/2 proteins) all showed some degree of cytoplasmic localization, although they share a 17-residue arginine-rich motif in their NH2-terminal region that might be predicted to act as a nuclear localization signal (3). Although the ORF2, ORF2/2, and ORF2/3 proteins all utilize the O2AUG at nt 354, and therefore share 116 amino acids at their NH2 terminus, their localization was not uniform. While the ORF2/2 and ORF2/3 proteins were predominately localized in the nucleus, the ORF2 protein was predominately found in the cytoplasm. This suggested that if the ORF2/2 and ORF2/3 proteins contain nuclear localization signals, these likely exist in their unique regions.
![]() View larger version (72K): [in a new window] |
FIG. 3. Subcellular localization of six TTV proteins. HA-tagged expression constructs expressing each of the TTV proteins (ORF1HA, ORF2HA, ORF2/2HA, ORF2/3HA, ORF1/1HA, and ORF1/2HA) or GFP-tagged expression constructs expressing each of the TTV proteins (ORF1GFP, ORF2GFP, ORF2/2GFP, ORF2/3GFP, ORF1/1GFP, and ORF1/2GFP) (described in the supplemental material) were transfected into 293 cells. Forty-eight hours later, cells expressing HA-tagged proteins as indicated to the left were fixed and stained with FITC-conjugated anti-HA monoclonal antibody (HA-7; Sigma), and propidium iodide (PI) was used to stain nuclei (A). Live cells expressing GFP-tagged proteins as indicated at the left were stained with SYTO-59 (Molecular Probes, Inc.) at final concentration of 5 µM (B). Confocal microscopy with a x60 objective lens was used, and images were taken at a zoom of x4 as previously described (9). The columns labeled FITC anti-HA or GFP show green TTV protein expression, the column labeled PI shows nuclei counterstained red in fixed samples, and the column labeled SYTO-59 shows whole cells counterstained with the SYTO-59 nucleic acid binding dye, which gives strong red staining in the nucleus and weak staining in the cytoplasm. Overlays show merged images of the individual TTV protein expression in green with either PI or SYTO-59 counterstaining in red. The row marked control shows fluorescence in 293 cells transfected with untagged pTV6.
|
This work was supported by PHS grants RO1 AI 46458 and AI 56310 and the Program for the Prevention of Animal Infectious Disease, U.S. Department of Agriculture, Agricultural Research Service, project 58-1940-0-008, to D.J.P.
Supplemental material for this article may be found at http://jvi.asm.org/. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»