Patent Application
20050019808
The picornavirus family, which includes encephalomyocarditis viruses (EMCV) have a genome which is composed of a single-stranded, positive sense RNA of about 7,500 to 8,300 nucleotides (Belsham et al., 1996). Picornavirus RNA is functionally monocistronic and, upon infection, is translated into a single polyprotein that is processed to yield structural and nonstructural virus proteins (Rueckert, R. R., 1996).
Picornaviral proteins and their precursors take their names (L, P1, P2, P3) from their sequential locations within the polyprotein. The leader or “L” proteins are present only in cardio- and aphthoviruses. The EMCV and Mengovirus leaders are about 7 kD in molecular weight. The four P1 peptides are the capsid proteins, 1A, 1B, 1C and 1D (EMCV: 8, 28, 25, and 30 kD). Those fragments derived from a common precursor stay together as a protomer unit throughout particle morphogenesis (Arnold et al., 1987; Grigera et al., 1985; Palmenberg, 1982). The middle portion of the polyprotein contains peptides 2A, 2B and 2C (EMCV: 16, 17, and 36 kD). Protein 2C is an ATPase (Pfister et al., 2000). In the entero- and rhinoviruses, 2C is also the genetic locus of the guanidine resistance marker, a compound that affects the initiation of RNA synthesis (Anderson-Sillman et al., 1984; Pincus et al., 1986).
However, the 2C protein is not a polymerase, and its contribution to the replication cycle remains unclear. The P3 peptides, 3A, 3BVPg, 3Cpro, and 3Dpol (EMCV: 10, 2, 22, and 51 kD) are more closely associated with genome replication. Preparations of 3Dpol can catalyze the elongation of nascent RNA chains in primer-dependent reactions, an activity that identifies this enzyme as a central element of viral polymerase complexes (Flanegan and Baltimore, 1977). Protein 3B is VPg, the peptide covalently linked to the 5′ end of the genome (Pallansch et al., 1980). VPg sequences are rich in basic, hydrophilic amino acids and have only one tyrosine residue (the attachment site) at position 3 from the amino end of the peptide. Initiation of positive- and negative-strand RNA synthesis requires VPg, perhaps as free protein or as part of a larger donor peptide (Morrow et al., 1984). Protease 3Cpro is the central enzyme in the viral cleavage cascade. After a co-translational primary break, catalyzed by a peptide cassette near the COOH-end of 2A (Hahn and Palmenberg, 2001), nearly all subsequent, or secondary cleavages within cardiovirus polyproteins, are affected by 3Cpro (Palmenberg, 1989).
Furthermore, infection with most picornaviruses is characterized by a strong inhibition of host cell protein synthesis at a time when virus-species proteins are efficiently produced (Ehrenfeld, E., 1996). Enteroviruses and rhinoviruses inhibit host translation, at least partially, by inactivation of eukaryotic translation initiation factor 4F (eIF4F), which binds to the cap structure of cellular mRNAs. eIF4F is composed of three polypeptides: eIF4E, eIF4A, and eIF4G. eIF4E is the cap-binding subunit (Sonenberg, N., 1996). Picornavirus RNAs are naturally uncapped and translate by a cap- and eIF4E-independent mechanism, by which the ribosomes bind to an (IRES internal ribosome entry site) (Agol, V. I., 1991).
Enteroviruses and rhinoviruses disrupt eIF4F through cleavage of the eIF4G subunit by 2Apro. This cleavage has been reported to be direct (Haghighat, et al., 1996) or indirect (Wyckoff et al., 1992). eIF4G cleavage does not preclude but, rather, stimulates cap-independent initiation of viral protein synthesis, since the cap-binding subunit, eIF4E, remains associated with the N-terminal cleavage product (Borman et al., 1997). The C-terminal cleavage fragment of eIF4G interacts with eIF4A and eIF3 to support IRES-dependent, but not cap-dependent, translation initiation (Borman et al., 1997).
In strong contrast to enteroviruses and rhinoviruses, it is widely known that no cleavage of eIF4G occurs following infection of cells with cardioviruses. Encephalomyocarditis virus, “EMCV” (NBCI Accession No. M81861), Mengovirus (NBCI Accession No. L22089), or Theilovirus (NBCI Accession Nos. M16020, M20562, and M20301) are examples of cardioviruses. It has been learned that these viruses have relatively efficient translation of their coding RNA in cell free systems. It has also been learned that there are various techniques for culturing and producing the natural cardioviruses.
Also, it is widely known that the 2A protein of EMCV is not similar to the enterovirus and rhinovirus 2Apro and does not possess protease consensus motifs or detectable proteolytic activity (Lloyd et al., 1988). Therefore, it is believed that cardiovirus infection does not induce eIF4G cleavage, and the encoded L and 2A proteins are not proteases. Furthermore, the genome sequences from EMCV, Mengo and the Theiler's viruses, which are known to encode 2A proteins (about 150 amino acids) do not share recognizable sequence similarity with any other viral or host proteins, including the 2As from other picornavirus species.
It has long been assumed that the shutoff of host cell protein synthesis after EMCV infection results from the ability of viral RNA to efficiently compete with capped cellular mRNAs for some limiting component of the translational machinery (Lawrence et al., 1974). Recently, it was suggested that EMCV causes the shutoff of host translation by dephosphorylation and activation of a suppressor of cap-dependent translation, 4E-BP1 (eIF4E-binding protein 1) (Gingras et al., 1996). However, to date no definitive mechanism has been established with respect to how EMCV causes the shutoff of host translation. Accordingly, it would be desirable to identify viral proteins involved in mRNA transcription and translation so as to enable the design of nucleic acid constructs which can be used to inhibit virally infected cells.
The invention describes the construction and the use of nucleic acid constructs assembled from genomes of viruses in the genus Cardiovirus. The compositions and methods of the invention are largely based upon the discovery that cardiovirus protein 2A, when expressed in mammalian cells, by any of a variety of genetically engineered methods, and in the absence of any other viral proteins or genes, will localize to the nucleoli of cells, bind to ribosomes, and consequently bring about a reduction in cap-dependent translation, but not IRES-dependent translation in these cells. Therefore, in accordance with this novel finding, two important observations were made: 1) the ability of protein 2A to inhibit cap-dependent mRNA translation; and 2) the ability of protein 2A in combination with 3BCD to inhibit cellular mRNA transcription, but not rRNA transcription.
Accordingly, in a broad sense the present invention provides novel compositions in the form of nucleic acid constructs, wherein the constructs have either a polynucleotide sequence encoding a 2A cardiovirus polypeptide operably linked to at least one DNA-dependent RNA polymerase promoter or to at least one IRES sequence. These constructs can be used to inhibit cellular mRNA transcription and cap-dependent translation.
More specifically, in one aspect, the present invention provides for nucleic acid constructs having a 2A sequence in the presence of at least one IRES-driven gene, wherein the gene could be any polynucleotide encoding a protein. The construct(s) may be introduced into the eukaryotic cells as a single tricistronic cDNA; as two or more mono-cistronic cDNAs or as 2A-fusion genes. The construct(s) may also be used in combination, rather than as single cDNAs; or as RNA; and as mixed combinations of DNA and RNA.
In another aspect, the present invention provides for nucleic acid constructs having a 2A sequence introduced as a toxic translational or transcriptional element in cells or cell-free systems. The construct(s) may be introduced into the eukaryotic cells as DNA, RNA or as a polypeptide.
In another aspect, the present invention provides for nucleic acid constructs having a cardiovirus 2A sequence introduced into the eukaryotic cells as a mutant 2A sequence resulting in altered cellular regulatory activities. The construct(s) may be introduced into the eukaryotic cells as having mutations in the 2A nuclear localization signal (NLS) sequence or at the carboxyl terminal 2A primary cleavage sequence.
In another aspect, the present invention provides for nucleic acid constructs having a cardiovirus 2A sequence is introduced into the eukaryotic cells in combination of other cardiovirus genes (e.g., L, 1A, 2C, 3A, 3B, 3C, and 3D) such that the 2A sequence is expressed from non-replicating cDNA or RNA; or the 2A sequence is expressed from self-replicating cDNA or RNA.
In another aspect, the invention provides methods for down-regulating or inhibiting mRNA transcription and/or cap-dependent translation by inhibiting pol-II activity through introducing suitable 2A construct(s) described herein into cells or cell-free systems.
In another aspect, the invention provides methods for inhibiting mRNA transcription in a virally-infected cell by introducing into a cell at least one nucleic acid construct having a combination of a cardiovirus 3BCD and a 2A sequence.
In yet another aspect, the invention provides kits comprising at least one nucleic acid construct comprising a cardiovirus polynucleotide sequence encoding a 2A polypeptide.
Also, a reporter gene may be positioned downstream of the 2A sequence, the IRES sequence, or both in any of the nucleic acid constructs described herein.
Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although suitable methods and materials for the practice or testing of the present invention are described below, other methods and materials similar or equivalent to those described herein, which are well known in the art, can also be used.
Other objects, advantages and features of the present invention will become apparent from the following specification taken in conjunction with the accompanying drawings.
FIGS. 4A-G represent various monocistronic, bicistronic, and tricistronic 2A sequence cDNA constructions (i.e., 2A fusion genes). “Gene1” and “Gene2” denote any reporter gene or other desired gene of interest to be expressed. “2A” denotes the cardiovirus sequence encoding protein 2A, along with its COOH terminal autocatalytic primary cleavage motif. “Promoter” designates any DNA-dependent RNA polymerase promoter that will produce mRNA transcripts from the cDNA. This includes pol-I, pol-II, or pol-III variants which may express in a constitutive (e.g., CMV1E) or inducible manner (e.g., CMVmin).
FIGS. 5A-D profile the various recombinant viral proteins. Panel A shows results from an antibody activity experiment. A map of the EMCV genome shows the protein order within polyprotein reading frame. HeLa cells infected with vEC9 (m.o.i. of 10) were harvested (4 hr PI), and analyzed for viral protein content in Western assays with mAbs to 2A (5A12), 2B (3G3), 2C (9H11), 3A (3D2), 3BVPg (2E4), 3Cpro (1D12), and 3Dpol (H. Duque). Panel B shows results from cell staining experiments. The same mAb panel (and mAb-1CD from V. Frolov), was used to stain infected cells for visualization by confocal laser microscopy using an anti-mouse, FITC-conjugated secondary antibody. Panel C shows results from protein 2A accumulation experiments. Infected HeLa cells were harvested at the indicated times, fractionated by SDS-PAGE, then probed by Western analyses using mAb-2A (5A12) or Ab-2A (murine polyclonal serum). Recombinant 2A (r2A) is included as a marker. Panel D shows L-292 cells were infected with vMwt (m.o.i. of 10), and harvested at the indicated times for confocal microscope visualization, using the above mAb-2A.
FIGS. 6A-E show the localization of protein 2A to the nucleoli. Panel A provides results from RNA synthesis experiments. HeLa cells were infected with vEC9 or vEC9-2AΔ58. After 2, 4 or 6 hrs, the cells were pulse-labeled with FITC-conjugated UTP, and 30 min later, they were fixed and stained with mAb-2A. The secondary anti-mouse antibody was conjugated to Texas Red. Panel B shows that BHK-21 cells were infected with vMwt as in Panel A, then co-stained with Ab-B23 (goat) plus mAb-2A then visualized with appropriate secondary antibodies. Panel C provides results from cell transfections with a 2A transcription vector. Transcript RNAs from pCITE-2A were transfected into HeLa cells as described in Methods. At the indicated times, cells were fixed and stained with mAb-2A and Ab-B23, as indicated. Panel D provides results from cell transfections with a 2AB transcription vector. Transcript RNAs from pCITE-2AB were treated as in Panel C. The cells were fixed and stained with mAb-2A, mAb-2B, Ab-B23, Ab-ribophorin II, or WGA, as indicated. Panel E shows results from HeLa cells transformed with pIRES-2A cDNA, induced by tetracycline, harvested at the indicated times, and stained similar to Panel C, except that SYTOX was used to visualize DNA.
FIGS. 8A-C show the interaction between protein 2A and B23. Panel A provides results from a B23 shift. Electromobility shift assays were as described in the Methods below. The samples were treated with RNaseA (50 μg/ml) as indicated, then fractionated by 4% native PAGE. After blotting onto membranes, the bands were visualized in Western assays using Ab-B23 (lanes 1-6) or mAb-2A (5A12, lanes 7-8). Panel B shows nuclear localization of Protein B23. HeLa cells infected with vEC9 (m.o.i. of 10) were harvested at the indicated times PI, fractionated by 4% native PAGE, and the B23 and 2A reactive bands were visualized in Western assays using Ab-B23 or mAb-2A. Panel C provides results from a series of immunoprecipitation experiments. Infected (lanes 4, 5, 6, 7) or mock-infected (lanes 1, 2, 3) HeLa cell extracts were harvested at 4 hrs PI. Reactive proteins were immunoprecipitated with mAb-2A1 (5A12, lanes 2 and 4), mAb-2A2 (5F5, lanes 3 and 5), or Ab-B23 (lane 6), then fractionated by SDS-PAGE. Lane 7 contains a sample of the infected cell lysate. The bands were stained using Ab-B23 or mAb-2A (5A12) as indicated. Lane 1 contains mock-infected HeLa cell lysate (B23 marker) that was not immunoprecipitated. The mouse, anti-goat secondary antibody used for ECL (lanes 1-5), also reacted with IgG chains from the murine mAb-2A. The rabbit, anti-mouse secondary antibody (lanes 6 and 7), did not have a similar reactivity with the residual (goat polyclonal) Ab-B23 chains.
FIGS. 9A-C show that protein 2A has the ability to shut off cap-dependent translation. Panel A: shows nucleic acid construct maps. EMCV-based replicon pE-luc, and transformation plasmids pIRES-luc, pIRES-luc-Δ3Dpol, pIRES-luc-2A and pIRES-luc-2A□58 are described herein the methods. Panel B provides results from replicon assay experiments. Luciferase activity in RNA transfected HeLa cells was determined as in Methods. All samples (20 μl of lysate) were prepared and assayed in parallel. Panel C provides results transformation assays. HeLa cells were transformed (or mock transformed) with plasmid cDNAs containing constitutively active CMV promoter sequences, as described in Methods. At the indicated times post-transformation, cell extracts were harvested and tested for luciferase activity. All samples (20 μl of lysate) were prepared and assayed in parallel. Error bars indicate the variation in duplicate samples.
FIGS. 10A-C show localization of P3-region precursor(s) (i.e., 3BCD) in nuclei of infected cells. Panel A shows confocal microscopy images of vEC9-infected HeLa cells (m.o.i. of 10) harvested at the indicated times (2, 3, 4, 6 h) PI, then stained with mAb-3B, mAb-3C, or mAb-3D, and then with FITC-conjugated, anti-mouse antibodies. All images used identical filter settings. Panel B shows HeLa cells infected with vMwt (m.o.i. of 10) were harvested at the indicated times PI, fractionated by SDS-PAGE, then probed in Western analyses with mAb-3D. Panel C shows a densitometry plot. Gel lanes from the Western blot (B) were scanned by densitometry, and the relative band intensity due to each protein was normalized for each lane and plotted.
FIGS. 11A-F show 3Dpol contains a nuclear localization signal and that the P-3 region is involved in nuclear transport. Panel A shows results from HeLa cells that were infected with vEC9. After 3, 4 or 6 hrs, the cells were pulse-labeled with FITC-conjugated UTP, and 30 min later they were fixed and stained with mAb-3B, mAb-3C or mAb-3D. The secondary anti-mouse antibody was conjugated to TRITC. Panel B provides results from chromatin visualization experiments. Samples were similar to A (mAb-3D) except that SYTOX stain was used to visualize chromatin. The bottom triptych is at higher magnification. Panels C-F provides results from transfection experiments. Genome-length RNA transcripts from pEC9 (C), pEC9-Δ3D (D), pEC9-3D-K17E (E) or pEC9-3D-R16D (F) were transfected into HeLa cells as described in the Methods below. At 6 hrs post transfection (PT), the cells were fixed and stained with WGA (Golgi marker) or Ab-B23 (goat), and with mAb-3B, mAb-3C, or mAb-3D, as indicated. The secondary anti-mouse antibody was conjugated to FITC. The secondary anti-goat antibody was conjugated to TRITC.
FIGS. 12A-B show the distribution of viral precursor proteins in infected cells. Panel A shows HeLa cells infected with vEC9 were harvested at 2, 4 or 6 hr PI, then fractionated into membrane, cytoplasm or nuclear fractions as described in the Methods below. Samples were fractionated by SDS-PAGE, then probed in Western assays with mAb-3D. Subsequently, the membranes were stripped and reprobed (inserts) using antibodies against actin (cytoplasmic marker) and C23 (nuclear marker). Panel B shows that infected HeLa cells were harvested at the indicated times PI, and nuclear extracts (as in A) were subject to immunoprecipitation with Ab-B23. The reactive material was fractionated by SDS-PAGE, then probed by Western analyses using mAb-3D.
FIGS. 16A-B show a comparative microarray analyses plotting the number of transcripts observed with high-end (decreased) or low end (increased) change p-values, when compared to the control (mock) array. Equivalent (10 μg) samples of total cellular RNA were isolated from mock-infected, or vEC9, vE-2A□58 or HRV-16 infected HeLa cells at 2 or 5 hr PI, processed and hybridized to microarrays (Affymetrix HU-95A) as described in Methods. GeneChip software (Affymetrix) was used to compared the data from each infected sample, to the mock-infected control, and assign a “p-change value” for each transcript (12,626 values, or “genes” per array). The comparative data were queried (Excel) for the number of transcripts meeting or exceeding specific p-change values between 0.000 and 0.001 (“increased”) or between 0.990 and 1.000 (“decreased”), and the numbers were plotted for each sample. The software default cutoffs of 0.003 and 0.997 (vertical lines) indicate very highly significant changes in relative transcript concentration.
We have discovered that cardiovirus protein 2A, when expressed in mammalian cells, through any number of genetically engineered methods, and in the absence of any other viral proteins or genes, will localize to the nucleoli of cells, bind to ribosomes, and consequently bring about a reduction in cap-dependent translation, but not IRES-dependent translation in these cells. Furthermore, we have discovered that the presence of 2A in these cells, in the absence of other viral sequences, also results in a general down-regulation of polymerase-II-dependent mRNA transcription of cellular genes in such cells, but 2A does not inhibit pol-I (rRNA) or pol-III (tRNA and 5S RNA) activity. Because cap-independent IRES-driven mRNA translation is not affected (negatively) by 2A, and continues in these cells, there is a net reduction of cap-dependent protein expression relative to (increased) IRES-dependent protein expression, when compared to cells or protein expression systems that lack 2A.
Thus, in broad terms, Applicants have designed a variety of nucleic acid constructs relating to methods of regulating CAP-dependent or CAP-independent, internal ribosomal entry site (IRES) driven mRNA transcription and translation through introduction of these novel constructs in cells or cell-free systems. These novel nucleic acid constructs of the present invention encode a 2A polypeptide operably linked to and promotable by suitable promoters described in detail below.
As used herein the term “nucleic acid construct(s)” is understood to mean artificial nucleic acid structures (i.e., non-replicating cDNA or RNA; or replicons, self-replicating cDNA or RNA) which can be transcribed and/or translated in the target cells or cell-free systems described herein. Such a construct may be inserted into a vector. Nonviral vectors such as plasmid expression vectors or viral vectors may be used. The kind of vectors and the technique of insertion of the nucleic acid construct according to this invention is known to the artisan. A nucleic acid construct according to the invention does not occur in nature in the arrangement described by the present invention. In other words, the gene of the nucleic acid construct is not naturally combined with the promoter sequence(s) and the foreign polynucleotide modules.
Accordingly, as used herein, the term “2A” refers to a small protein (16 kDa) that encompasses the 133 amino acid, protein fragment (TME or Theiler's murine encephalomyomyelitis virus) or 143 amino acid length (EMCV or encephalomyocarditis virus) encoded as part of the viral polyprotein, which lies between the protein 1D (VP1) capsid coding fragment, and the protein 2B polymerase auxiliary protein encoded by a cardiovirus genome, including the EMCV and the TMEV. For EMCV-R this represents amino acids 902-1044 of the 2292 length. Applicant's note that for convenience, the base numbering systems of Meng-M (#L22089), and EMCV-R (#M81861) are used in all cloning descriptions described hereinbelow.
In a viral polyprotein context, the amino end of 2A is usually released by a viral protease 3C-catalyzed event. The carboxyl end of the 2A protein is released by a co-translational autocatalytic event, dependent upon the terminal 19 amino acids of the protein, as described in U.S. Pat. No. 5,912,167 (incorporated by reference herein in its entirety). Also, the active 2A protein can be produced in a recombinant manner by genetically engineering similar synthetic ends into cDNA, and expressing the protein by cap-dependent or IRES-dependent translation. It is also noted that the autocatalytic carboxy-terminal sequences of 2A allows cleavage of the active protein from a fusion-protein precursor, reporter gene, or desired expressed protein.
The term “TME” or “Theiler's murine encephalomyomyelitis virus” as used herein refers to any member isolate or strain of the Theilo virus species of the Cardiovirus genus of the family Picornaviridae. Examples are: TME-BeAn strain, TME-DA, TME-GDVII, and Vilyuisk virus.
Furthermore, the term “IRES” as used herein refers to an internal ribosomal entry site. In general an IRES is a feature in the secondary structure near the 5′ end of a picornaviral RNA genome that allows eukaryotic ribosomes to bind and begin translation without binding to a 5′ capped end. For purposes of this invention an IRES is any single-stranded RNA fragment that directs cis-acting cap-independent translation of a eukaryotic mRNA. A “type II IRES” refers specifically to the naturally occurring viral IRESs from isolates in the Cardiovirus, Aphthovirus, Kobuvirus and Achivirus genera, which share recognizable sequence and secondary structural similarities. Other IRESs (e.g., type I, or type III, or hepatitis C [HCV]), as found in the genomes of other viral genera or in certain cellular mRNAs, are functionally analogous, but may differ significantly from the type II elements, in sequence or structure. Thus, in accordance with the invention, the IRES region is derived from a picornavirus IRES region sequence, specifically selected from an enterovirus, rhinovirus, cardiovirus, and aphthovirus IRES sequence.
The term “EMCV” or “encephalomyocarditis virus” as used herein refers to any member isolate or strain of the encephalomyocarditis virus species of the Cardiovirus genus of the family Picornaviridae. Examples are: EMCV-R (Rueckert) strain, Mengovirus, Mus Elberfeld virus, Columbia-SK virus.
The term “CAP” or “cap” as used herein refers to a modified nucleotide, generally a 7-methyl guanosine, linked 3′ to 5′ (7meG-ppp-G), to the 5′ end of a eukaryotic mRNA, that serves as a required element in the normal translation initiation pathway during expression of protein from that mRNA.
In accordance with the invention, Applicants have described and enabled the construction and use of a variety of different nucleic acid constructs hereinbelow. The various embodiments are categorized into the following four general types of nucleic acid constructs. In this first series of constructs the 2A sequence is provided in the presence of an IRES-driven gene. The construct(s) may be introduced into the eukaryotic cells as a single tricistronic cDNA; as two or more mono-cistronic cDNAs or 2A-fusion genes, to be used in combination, rather than as single cDNAs; or as RNA; and as mixed combinations of DNA and RNA. In the second series of constructs the 2A sequence is introduced as a toxic translational or transcriptional element in cells or cell-free systems. Applicants note that those skilled in the art of cell-free protein expression are well aware of the various cell-free systems available for use in expressing the 2A constructs of the invention. Suitable cell-free systems encompassed within the scope of this invention include wheat germ, xenopus, rabbit reticulocyte, and S-30.
The construct(s) may be introduced into the eukaryotic cells as DNA, RNA or as protein. In the third series of constructs the 2A sequence is introduced into the eukaryotic cells as a mutant 2A sequence resulting in altered cellular regulatory activities. The construct(s) may be introduced into the eukaryotic cells as having mutations in the 2A nuclear localization signal (NLS) sequence or at the carboxyl terminal 2A primary cleavage sequence. In the fourth series of constructs the 2A sequence is introduced into the eukaryotic cells in combination of other cardiovirus genes, such that the 2A sequence is expressed from non-replicating cDNA or RNA; or the 2A sequence is expressed from self-replicating cDNA or RNA.
Applicants note that so far no definitive mechanism has been established for how 2A alone and in combination with 3BCD function. However, it is believed that during cardiovirus infection, the 2A protein is found to be associated with ribosomes as part of a unique host-translational shut-off mechanism that probably involves the synthesis or modification of “toxic” ribosomes. Moreover, it is believed that 2A expression within cells can bring about a 10-20 fold reduction of host mRNA transcription as well, by a mechanism that may involve inhibition of the cellular DNA Polymerase II enzyme. The DNA Polymerase II, however, does not shut off translation directed by mRNAs containing internal ribosomal entry sites (IRESs), such as that encoded by the parental virus, EMCV. Furthermore, it is believed that the 2A protein does not appear to shut off cellular transcription by DNA polymerase I or III enzymes that lead to the synthesis of new ribosomes and tRNA. Furthermore, it is believed that when the 2A protein is introduced into cells, IRES-dependent translation is significantly enhanced because it appears that the host cannot manufacture competing mRNAs, and of those host mRNAs in the cell, the ribosome preference is directed towards IRES-containing mRNAs. Accordingly, it appears that the 2A protein is capable of acting as an IRES enhancer protein that can down-regulate host cap-dependent protein translation and boost IRES-dependent translation.
The following examples are provided as further non-limiting illustrations of particular embodiments of the invention.
In the first embodiment, the 2A sequence is provided in the presence of an IRES-driven gene. In this embodiment, the construct(s) may be introduced into eukaryotic cells from a single DNA. For example,
By “reporter gene”, applicants refer to a foreign polynucleotide or cDNA that when expressed yields a polypeptide that is readily detectable, either by providing a colorimetric signal under certain environmental conditions or some other signal well known to those of ordinary skill in the art, as described below.
Furthermore for example,
This embodiment also encompasses a construct where the IRES element that is driving the expression of gene2 (or 2A-gene2), could be derived from any other cap-independent protein expression system, such as another type-II IRES, a type-I IRES, the IRES from hepatitis C, or another virus, or even a naturally occurring cellular IRES.
Another construct envisioned by this embodiment is where the transcriptional promoter is not pol-2 dependent (e.g., pol-I or pol-III-dependent), but still produces capped mRNAs that are exported to the cytoplasm, and translated there.
Also encompassed by this embodiment is an example where both gene1 and gene2, either one of which may be fused to 2A, are driven by two different, non-competitive IRES sequences. In this case, 2A will down regulate host mRNA synthesis, but the expression of both cDNA genes will proceed, because they are not affected by 2A.
It is encompassed within the scope of this embodiment that the construct(s) may be introduced into eukaryotic cells from multiple DNAs. For example, the constructions described in
In yet another example, shown in
In this embodiment, the construct(s) may also be introduced into eukaryotic cells from RNA. Nearly all constructions in 1A and 1B above could also be introduced into cells by transfection with RNA instead of transformation with DNA. The advantage of an RNA method would be the elimination of a cellular transcriptional promoter sequences from any of the constructs, and in some embodiments, a consequent lack of capping requirements on the introduced RNAs.
Also, within the scope of the invention is that cells could be transfected with a tricistronic mRNA synthesized in vitro using T7 polymerase and a cap analogue. The RNA could be of the form: cap, 2A-gene1 fusion, EMCV IRES, gene 2. Alternatively, as in
In yet another aspect of this embodiment it is encompassed that the cells could be transfected with multiple mRNAs, each encoding a different IRES-driven gene. At least one of these RNAs must encode 2A, as an IRES-driven or cap-driven gene, or as part of a self-releasing 2A fusion protein.
Also, in this embodiment, it is encompassed that the construct(s) may also be introduced into eukaryotic cells from combinations of DNA and RNA. Applicants note that all constructions described in FIGS. 4A-C above, involving multiple cDNAs or RNAs, and could be introduced in mixed combinations.
In this embodiment, it is envisioned that the cells could be stably transformed with a cDNA encoding an inducible pol-II promoter (e.g., CMV), followed by a 2A gene, or 2A gene fused to a reporter. Subsequent transfection with an mRNA encoding an IRES-driven gene2, followed by induction of gene1, would result in expression of 2A (via a pol-II produced capped mRNA), 2A-dependent down regulation of pol-Il transcription and cap-dependent translation, yet there would be continued expression of IRES-dependent gene2. The net result would be a higher expression level of the gene2 product per cell, than if this RNA were transfected without parallel 2A expression.
In the second embodiment, it is within the scope of the invention to have the 2A sequence is introduced as a toxic translational or transcriptional element in cells or cell-free systems. For example, the 2A protein by itself, and in the absence of any other viral or cellular proteins is toxic to cells because it will induce a down regulation of pol-II dependent mRNA transcription, and an inhibition of the translation of capped (cellular) mRNA. In all above described examples (section 1), the introduction of 2A was used as an aid to enhance translation expression from IRES-driven genes that were introduced in tandem (or in timely sequence) with 2A. However, the singular introduction of 2A into cells will also target them into an apoptotic or necrotic death cycle.
In this embodiment, it is envisioned that the 2A gene can be introduced as DNA. For example, the 2A gene is introduced into any cell-targeting delivery system as a freestanding gene, or as the amino-terminal segment in a fusion protein gene, to be released autocatalytically during translation. In a preferred construction, 2A expression itself could be driven by an IRES (to increase the expression level, but this is not a requirement). Induction, mRNA transcription and translation of 2A protein would lead to 2A-induced shutoff of all cellular translation and pol-II transcription, leading eventually to cell death in any cell that expressed 2A. Specific targeting (e.g.,) of tumor cells, by delivery of 2A-encoding cDNAs would result in their necrotic death.
Also, it is envisioned that the cells may be stably transformed with a plasmid cDNA encoding an IRES-2A gene sequence (or just the 2A gene), under the control of an inducible promoter sequence (e.g., CMV). Cells could be propagated or amplified in the absence of 2A induction, but after induction, all cells carrying this gene would undergo 2A-mediated apoptotic or necrotic death within (about) 24 hrs).
Another example of this embodiment is that a 2A gene may be introduced simply as RNA. Here the 2A gene transcripts (capped or driven by any IRES segment) would be introduced into cells by transfection, or by any RNA gene-vector delivery system. Expression of 2A would result in targeted cell death.
Also, encompassed within this embodiment is that 2A may be introduced to a cell as a protein: For examine from eukaryotic cells will express proteins from capped or IRES-driven mRNAs. The introduction of recombinant 2A protein into such extracts would result in the expression of proteins only from the IRES-driven mRNAs. This method could result in lower background levels of endogenous (capped) protein expression in such extracts or act as a timing switch between capped and IRES-driven protein expression.
In another example, injection or electroporation on of recombinant 2A protein into cells would be toxic for those cells and kill them.
In the third embodiment, the 2A sequence is introduced as Mutant 2A sequences having altered cellular activities. In this embodiment, the introduction (or synthesis) of 2A protein in cells has two effects: a reduction of cap-dependent translation (in the cytoplasm) and a reduction of pol-II dependent mRNA transcription within the nucleus. The inhibition of pol-II requires transport of 2A into the nucleus via a nuclear localization signal (NLS) encoded within the protein. The cytoplasmic translational inhibition activity of 2A does not require nuclear transport, or the activity of the NLS. Neither the cytoplasmic nor the nuclear activities require the endogenous, autocatalytic activities of 2A.
In this embodiment, a 2A NLS mutation is introduced into the cells, such as is shown in
Furthermore, it is encompassed within the scope of the invention that the mutation may be located in the 2A primary cleavage sequence, such as is shown in
In the fourth embodiment, it is encompassed that the 2A can be used in combination with other cardiovirus genes. One example enables the expression of 2A from non-replicating cDNA or RNA. Specifically, it is possible the 2A induced shutoff of pol-II transcription and cap-dependent translation could be further enhanced or made to occur even more rapidly if one or more additional EMCV viral proteins were added to the expression systems, as shown above.
In this embodiment, it is envisioned that the tricistronic or bicistronic cDNA or RNA constructions described above with the gene1 sequence could be replaced with an EMCV gene encoding protein 3BCD, a known form of the viral protease that is capable of targeting the nucleus. This would (by itself) bring about a faster rate of pol-II inhibition, via a mechanism that amplifies the parallel 2A-mediated inhibition of the same enzyme. The combination of these inhibitory two viral activities should hasten cellular transcriptional shutoff, and consequently, the switch from cap-dependent to IRES-dependent translational activity for the enhanced expression of any IRES-driven foreign gene (i.e. gene2).
In another example, it is encompassed that the 2A expressed from self-replicating cDNA could be introduced into cells. Specifically, picornaviral replicon sequences have been described, in which the viral capsid sequences of a viral cDNA are replaced with a reporter gene, or gene encoding a desired expression protein. In particular, replicons which use two 2A-encoded primary cleavage sequences to release the protein product of the added gene have been described (Binder, J. J., Hoffman, M. A., and Palmenberg, A. C. (2003) Virology, 312(2):481-94.
Thus, as used here the term “replicon(s)” includes, but is not limited to, self-replicating recombinant positive strand RNA molecules. The term “positive strand” as used herein includes, but is not limited to an RNA strand that directly encodes a protein. Replicons can be constructed by deleting all or part of capsid coding sequences and retaining all coding and non-coding sequences necessary for replication. Retention of genomic replication sequences allows the expression of viral and/or heterologous gene products in appropriate cells. For example, the CRE, found in the Mengo virus VP2 gene, is essential for replication. Also, the term “expression” or any variation thereof as used herein includes, but is not limited to, giving rise to or encoding the production of a protein or part of a protein.
Furthermore, in accordance with the invention, replicons can be prepared by several approaches as described in detail below. For example, one approach, the appropriate DNA sequences are transcribed in vitro using a DNA-dependant RNA polymerase, such as bacteriophage T7, T3, or SP6 polymerase. In another approach, replicons can be produced by transfecting animal cells with a plasmid containing appropriate DNA sequences and then isolating replicon RNA from the transfected cells.
Applicants note that as used herein, the term “transfection” includes, but is not limited to, the introduction of DNA or RNA into a cell by means of electroporation, DEAE-Dextran treatment, calcium phosphate precipitation, liposomes (e.g., lipofectin), protein packaging (e.g., in pseudo-viral particles), protamine condensation, or any other means of introducing DNA or RNA into a cell.
Also, in this embodiment, it is envisioned that an EMCV or Mengo replicon cDNA sequence linked to an upstream pol-II promoter, such as for example, a constitutive or inducible CMV promoter sequence, and also and encoding an exogenous reporter gene or expression gene of interest (e.g., human interferon), would be transfected into cells. The constitutive (or induced) transcription of this sequence would produce an mRNA, exported to the cytoplasm where it would undergo translation, producing the 2A protein, and all the other viral proteins from the P2-P3 region that were necessary and sufficient for transcript amplification in an RNA-RNA replication cycle. The endogenous, heterologous gene would thus be amplified, and its expression, as driven by the viral IRES inherent in the replicon sequence, would not be shut off by subsequent 2A activities. The combination of a DNA-based gene-introduction system, transcript amplification by RNA replication, and 2A-induced cap-dependent mRNA shutoff, should result in huge levels of targeted gene expression, relative to any other known method (see
Furthermore, the present invention provides for kits, wherein the kits are composed of at least one of the novel nucleic acid constructs described hereinabove. More preferably, it is envisioned that a kit of the invention will include at least one nucleic acid construct having either a polynucleotide sequence encoding a 2A polypeptide operably linked to a DNA-dependent RNA polymerase promoter or to an IRES sequence.
Applicability of the Invention
Applicants envision that the nucleic acid constructs described by the present invention may be suitable for use in a wide variety of potential applications. For example, the constructs can be used to express heterologous proteins in eukaryotic or animal cells or an animal host by inserting sequences coding for heterologous polypeptides into the constructs and introducing the constructs into the animal cells or the animal host. Host cells used for this application are preferably human but can be from dog, cat, pig, cow, chicken, mouse, or horse.
It is envisioned that the replicon constructs can be introduced into the host by several means, including intramuscular injection, gold particle-coated gene gun delivery, protein-packaged injection (e.g., packaged in pseudo-viral particles), protamine-condensed injection, or liposome-encapsulated injection. For example, a Mengo virus-derived construct allows the transient expression of a toxic protein or a proapoptotic protein in a solid tumor by direct injection, thus providing a form of anti-tumor gene therapy. Thus, the novel constructs can be used to down-regulate mRNA cellular transcription and/or cap-dependent translation in eukaryotic cells.
In addition, recombinant replicon constructs of the invention can be used in vitro or in vivo in order to express conveniently detected reporter protein, as described in detail below. These constructs can be used to monitor RNA replication and RNA delivery, thereby allowing for optimization of animal cell transfection or RNA delivery into an animal host. Finally, these constructs can for example, be used to increase protein expression to conduct further studies on protein characterization, protein production, or protein localization.
It is further envisioned that the constructs of the invention could be used to induce an immune response against the encoded heterologous protein in an animal recipient. Thus, the constructs of the instant invention along with a pharmaceutically acceptable carrier can comprise a vaccine. Possible pharmaceutical carriers include, but are not limited to, sterile liquids, such as water, oils, including petroleum oil, animal oil, vegetable oil, peanut oil, soybean oil, mineral oil, sesame oil, saline solutions, aqueous dextrose, glycerol solutions, liposomes, gold particles, and protamine or any other protein or molecule able to condense the RNA. It is noted that the suitably designed constructs of the invention can also, for example, be injected in the form of naked RNA. The term “naked” as used herein includes, but is not limited to, an RNA molecule not associated with any proteins.
Materials and Methods
Viruses and Cells
Recombinant vMwt (Mengo), vMC0 (Mengo) and vEC9 (encephalomyocarditis) have been described (Hahn and Palmenberg, 1995; Martin et al., 1996) (species encephalomyocarditis virus, genus Cardiovirus, family Picornaviridae). Mutant derivatives vE-2A□58 and VE-2A□120 have deletions in the 2A coding-region relative to vEC9 that remove 58 and 120 amino acids, respectively (Svitkin et al., 1998). The 19 amino acid COOH-terminal primary cleavage cassette (PCC) is intact and functional in both of these deleted viruses (Hahn and Palmenberg, 2001). Viruses were cultured in HeLa, L-929 or BHK-21 cells, at 37° C. under 5% CO2-air, using RPMI-1640 medium supplemented with penicillin, streptomycin and 10% fetal bovine serum as described (Rueckert and Pallansch, 1981). Typically, subconfluent cell monolayers (5×106 cells per 6 cm plate) were infected with virus at a multiplicity (m.o.i.) of 10 plaque-forming units (PFU) per cell.
Recombinant Proteins
Isolation of recombinant Mengo 3Cpro and 3Dpol from E. coli has been described (Hall and Palmenberg, 1996; Duque and Palmenberg, 1996). For convenience, the base numbering systems of Meng-M (#L22089), and EMCV-R (#M81861) are used in all cloning descriptions. To isolate recombinant Mengo 2A protein, the gene (429 b) from pMC0 (b 3462-3890) was amplified by PCR. One primer contained an engineered Nde I site, an ATG codon, and 19 nucleotides derived from the 5′ end of the 2A coding region. The second primer contained nucleotides complementary to the 3′ end of 2A region, the complement of a TAG codon, and a Bam HI site. The amplicon was subcloned into pET-41b (Novagen), transformed into E. coli (strain BL21-DE3-pLysS, Novagen), and the bacteria were amplified in 2xYT broth (37° C.), supplemented with kanamycin (15 mg/ml) and chloramphenicol (30 mg/ml). After induction (5 hrs, with 1 mM isopropylthio-β-D-galactoside), the cells were pelleted and then resuspended in buffer A ({fraction (1/10)} volume, 50 mM Tris-HCl pH 8.5, 1 mM EDTA, 1 mM DTT, 10% glycerol, 1% deoxycholate), lysed by freeze-thaw and subject to centrifugation (27,000×g, 15 min). The pellet was washed twice (buffer A). The inclusion bodies were collected by centrifugation (12,000×g, 5 min), then resuspended in buffer B (0.1% Triton X100, 100 mM NaCl, 6M urea, 50 mM Tris-HCl, pH 8.0, 1 mM PMSF). Insoluble material was removed (12,000 g, 5 min), and the supernatant was dialyzed against TE buffer (10 mM Tris-HCl pH 8.0, 1 mM EDTA, 0.1% Triton X100), then concentrated (Centricon-10, Amicon).
Recombinant Mengo 2B protein (b 3891-4343) was produced in E. coli from a pET-41b plasmid (Novagen) that had been engineered to contain the pMCo 2B gene (450 b), in a manner similar to that described above, except the 2B-specific primers used Nde I and Bam HI sites. Recombinant Mengo 2C protein (b 4344-5318) was also produced from pET-41b, but as a fusion protein that linked a GST-His-Stag segment to the N-terminus of the viral sequence (2C-specific primers, had added Nco I and Bam HI sites). The 2C-containing protein was isolated from disrupted cells and purified by fractionation on a His-column (Novagen). EMCV-R proteins 3A and 3AB (b 5391-5654 and b 5391-5714 respectively) were expressed using procedures similar to those described for 2A and 2B, except the vector, pET-11a (Novagen), was used. Bacteria harboring these plasmids were grown in M9 media supplemented with glucose (0.2%) and rifampicin (1 mM, Sigma). The 3A and 3AB-containing inclusion bodies were resuspended in buffer B1 (0.5% Triton X100, 500 mM NaCl, 50-mM Tris-HCl, pH 8.0, 1 mM PMSF) before clarification and dialysis.
EMCV Replicons
Mengo replicons which express active firefly luciferase after RNA transcript transfection of HeLa cells have been described (Duque and Palmenberg, 2001). A similar EMCV-based replicon (pE-luc) was created by replacing an Spe I to Ava I fragment of pEC9 (b 2426-3509 in 1C-1D coding region) (Hahn and Palmenberg, 1995) with an in-frame firefly luciferase gene (from plasmid pGL2, Novagen), proceeded by an autocatalytic primary cleavage cassette (PCC) derived from the pMC0 2A-2B junction (84 b of 2A, 30 b of 2B). Translation of T7 RNA transcripts from this replicon produced a full-length, active luciferase enzyme with 10 additional viral amino acids at the N-terminus (from the Mengo PCC), and 29 additional viral amino acids at the C-terminus (from EMCV 1D). A derivative, pE-luc-2A□58, deleted 174 bases (b 3777-3951) encoding 58 amino acids (Beretta et al., 1996) from the middle of the EMCV 2A gene. A second derivative, pE-luc-Δ3Dpol, had a 555 base deletion (b 7177-7732) that effectively removed the active site of viral 3Dpol (Eco47 III to Mlu I fragment) Proteins directed by these transcripts were inactive for viral RNA synthesis. Luciferase assays in extracts from transfected HeLa cells were as described (Duque and Palmenberg, 2001).
Eukaryotic 2A Expression Vectors
The Mengo 2A and 2AB genes were engineered into vectors designed to evaluate protein expression in HeLa cells after RNA transcript transfection. The vectors were based on plasmid pCITE-4b (Novagen), which contains an EMCV IRES under the control of a T7 polymerase promoter. Appropriate viral amplicons encoding 2A or 2AB were derived from pMC0 using the Nde I and Bam HI-containing primers described above. Gel-purified fragments were ligated into pCITE DNA that had been digested with these enzymes. When translated in reticulocyte extracts (Shih et al., 1979), T7 transcripts from pCITE-2A and pCITE-2AB produced viral proteins of the expected size and immunogenicity, and the PCC within 2AB cleaved to near completion (results not shown).
The Mengo 2A gene was also engineered into a DNA transformation vector (Pmin CMV), where protein expression was under the control of an inducible, minimal cytomegalovirus promoter (pTRE2hyg, Clontech). PCR reactions included one primer with a Bam HI segment linked to a 5′ fragment of the EMCV IRES, and another primer complementary to the 3′ end of the 2A gene, linked to an Mlu I segment to amplify the IRES-2A sequence from within pCITE-2A. The amplicon was used to replace the Bam HI-Mlu I fragment of pTRE2hyg. When transformed into HeLa cells and induced by tetracycline (2 μg/ml, according to manufacturer's instructions), pIRES-2A produced (capped) transcripts that directed 2A translation under control of the EMCV IRES.
The EMCV 2A gene and derivative 2A□58 were engineered into bicistronic, luciferase-containing plasmids under the control of the immediate early promoter sequence of cytomegalovirus (PCMV IE). Commercial plasmid pIRES (Clontech), was used to link this constitutive CMV promoter to a firefly luciferase gene, followed by an intact EMCV IRES and a multiple cloning site (MCS). The 2A and 2A□58 segments from pEC9 and pEC9-2A□58 were amplified using flanking primers containing Xba I or Not I sequences, then ligated into the pIRES MSC, after digestion with appropriate enzymes. The pIRES-luc (control), pIRES-luc-2A and pIRES-luc-2A□58 plasmids (5 μg) were transformed into HeLa cells (5×106 per plate) with transfectin (Qiagen), then incubated (37° C., 5% CO2-air) in the presence of neomycin, following manufacturer's instructions. Luciferase assays in extracts from transformed cells were as described (Duque and Palmenberg, 2001).
Antibodies
Murine monoclonal antibodies (mAbs) raised against recombinant Mengo 3Dpol and capsid protein 1CD were generously supplied by Dr. H. Duque, Dr. V. Frolov and Dr. O. Frolova (Duque and Palmenberg, 1996). Murine monoclonal and polyclonal antibodies to Mengo 2A, 2B, 2C (GST-HIS-Stag fusion), 3A, 3AB, and 3Cpro were developed from recombinant proteins (above) according to described methods (Duque and Palmenberg, 1996; Hall and Palmenberg, 1996). To raise anti-3B antibodies, a water-solubilized synthetic peptide (GPYNETARVKPKTLQLLDIQ) SEQ. ID. NO. 1, corresponding to the complete EMCV 3BVPg protein (b 5655-5714) was conjugated with KLH (keyhole limpet hemocyanin) and used to immunize mice (40 μg, over 6 doses, during 18 weeks). Antibody reagents raised against each viral protein were tested for high-titer ELISA reactivity with the corresponding recombinant protein. The antibody subtypes (Ig) produced by each line (hybridoma or ascites) were identified with commercial test kits (Sigma). Goat polyclonal antibodies to nucleolar-specific proteins, B23 and C23, were purchased (Santa-Cruz Biotechnology, Inc.)
Immunodetection of Proteins
Western assays used cells harvested at appropriate times post infection (PI). The cells were washed with phosphate-buffered saline, lysed by freeze-thaw (3×), and the clarified supernatants were fractionated by SDS-PAGE, then blotted onto polyvinylidene fluoride membranes (Immobilon-P, Millipore) as described (Duque and Palmenberg, 2001). Bands with positive reactions against the appropriate antibodies (typically, 1:2,000 dilution of ascites), were visualized by chemiluminescence (ECL kit, Amersham Pharmacia Biotech, Inc.), after secondary reactions with appropriate anti-mouse antibodies, conjugated with horseradish peroxidase (1:2,000 dilution, Amersham Biosciences). Mobility shift assays with viral proteins by non-denaturing 4% PAGE were similar to those described (Kiessig et al., 2001; Stern and Frieden, 1993). Briefly, recombinant 2A protein (30 ng), and nuclear extracts or whole cell lysates from HeLa cells (20 μl, Promega) were co-incubated in buffer (20 mM HEPES, pH 8.0, 100 mM KCl, 1 mM EDTA, 5 mM MgCl2, 0.1% Triton X100 and 1 mM DTT) for 1 hour at 30° C. The samples were fractionated by non-denaturing PAGE (20° C., TBE buffer, 200 V), transferred to membranes, and the bands were visualized by Western assay (Ab-B23 or mAb-2A), using ECL (anti-goat or anti-mouse secondary antibodies, respectively).
Immunofluorescence Microscopy
HeLa or L-929 cell monolayers were grown on coverslips to 50-70% confluence then infected (10 PFU/cell), transfected (1.5 μg RNA per 5×106 cells) or transformed (5 μg DNA per 5×106 cells) as described above. At appropriate intervals (37° C. under 5% CO2-air), the cells were rinsed with phosphate-buffered saline, fixed with paraformaldehyde, permeabilized (0.3% Triton X-100), then incubated (1 hr, 20° C.) with the desired primary antibody (1:2000 dilution in blocking solution), as described (Amineva et al., 2003). After reaction with a corresponding secondary antibody, the slips were mounted (Vectashield mounting medium, Vectorlab) and label location was visualized by laser-scanning confocal microscopy. Image capture used MCR 1024 LaserSharp software (Bio-Rad, Inc.). For double-label experiments, primary antibodies raised in different species (e.g., mouse and rabbit) were selected, and the samples were developed with appropriate, corresponding secondary antibodies (1:100 dilution) conjugated with fluorescein isothiocyanate (FITC) or Texas Red (Santa Cruz Biotechnology, Inc.). RNA labels incorporated fluorescein-12-UTP (Sigma) into permeabilized cells before the paraformaldehyde fixation step, as described previously (Aminev et al., 2003; Amineva et al., 2003). A fluorescent-tagged wheat germ agglutinin (WGA, Molecular Probes, Inc.) was used to highlight and identify Golgi and nuclear membrane locations. SYTOX stain (Molecular Probe Inc.) was used to localize dsDNA within cells. Rabbit polyclonal serum against ribophorin II.
Results and Discussion
Recombinant Viral Proteins
In accordance with the present invention applicants have described the cloning and isolation of recombinant Mengo 3Cpro and 3Dpol, and the characterization of murine monoclonal antibodies raised against these proteins (Duque and Palmenberg, 1996; Hall and Palmenberg, 1996). Using similar procedures, the Mengo genes for 2A, 2B, 2C, 3A and 3AB were cloned into bacterial expression vectors. The 2C gene was linked to a GST-His-Stag fragment in a construction which facilitated subsequent protein solubility and isolation, but the other proteins were synthesized as unmodified, non-fusion sequences. Most of the expressed material was produced in inclusion bodies which were readily solubilized by detergent and dialysis. A typical yield was about 100 mg of protein per liter of culture. After verification of protein size and purity by SDS-PAGE, the preparations were inoculated into mice. To complete the panel, a synthetic peptide with the Mengo 3BVPg sequence was inoculated in parallel. Animals which seroconverted to their respective proteins (ELISA) were exsanguinated (polyclonal sera) and their spleens were used to develop hybridomas producing monoclonal antibodies and subsequent high-titer ascites (Duque and Palmenberg, 1996). Each reagent was tested for ELISA reactivity with its recombinant protein and also in Western assays with lysates from Mengo-infected HeLa cells, so the precursor recognition capacity could be characterized. As an example, five IgG-producing hybridoma lines were generated to 2A (5A12, 5F5, 4D10, 3A6, 1C7), each of which reacted strongly with recombinant protein in ELISA, immunoprecipitated its targets from infected cell lysates, and stained infected cells by confocal microscopy. But only two of these 2A mAbs (5A12 and 5F5) also gave positive signals in Western assays with infected lysates. When used at appropriate concentrations (typically, 1:2,000 dilution of ascites), only a few mAbs within the entire panel of new reagents showed any reactivity with cellular proteins. Nevertheless, care was taken in every experiment to include uninfected cell samples for the identification of spurious signals.
The Western profiles of selected mAbs are shown in
Protein 2A Localizes to Nucleoli
Having identified the cohort of viral precursors (or multimers) that gave rise to immunogenic signals, the next step was to identify within cells, their preferred location. The replication of infectious picornaviruses is a cytoplasmic process. Indeed, infectious poliovirus can be completely synthesized in cell-free extracts (Molla et al., 1991), and this virus will replicate in enucleated cells (Detjen et al., 1978). Electron microscopy studies with poliovirus (using mAbs to 2C and 3Dpol) have confirmed repeatedly that genome translation and replication take place in cytoplasmic foci where the smooth and rough endoplasmic reticulum (ER) are in near conjunction (Bienz et al., 1983; Bienz et al., 1992; Egger et al., 2000; Bienz and Egger, 1995). Our first microscopy experiments with cardiovirus mAbs were in agreement with these findings, in that signals from capsid proteins (1CD) and nonstructural proteins 2B, 2C and 3A (
As described above and shown in
Protein 2A and B23
The sequence and structural elements that allow mammalian proteins to shuttle into nuclei and nucleoli are not very well understood. No cardioviral proteins, including 2A, have canonical nuclear localization signals (NLS), that exactly match the three classical types described for eukaryotic proteins using the importin α or importin β pathways (Michael, 2000; Hicks and Raikhel, 1995). Nor do the viral sequences suggest an obvious small nuclear RNA (snRNA) binding motif that could easily explain their targeting. On the other hand, 2A does have a rather basic amino acid profile (18% H+K+R, pI of 9.23), and contains at least one segment resembling the [KR][KR]X10[3 of 5: KRHW] sequence that defines about 50% of known nuclear proteins in eukaryotic cells (Michael, 2000). Moreover, a subset of this segment (
The nucleolar signals within 2A are yet another unknown. Nucleoli, of course, are the genome sites where pol-I transcribes dense clusters of rRNA genes into a 45S pre-rRNA. These ribosome factories splice pre-rRNA into 18S, 28S and 5.8S fragments as the 40S and 60S ribosomal subunits are assembled on site. Ribosomal 5S rRNA, synthesized by pol-III, and ribosomal proteins translated in the cytoplasm, are shuttled into complex nucleolar scaffolds as the ribosomes are built (Melese and Xue, 1995). The most curious observation about nucleoli in Mengo-infected cells was that they never shut off their synthesis. Nucleolar antibodies (B23 and C23), viral mAbs (2A), and fluorescein-UTP labels showed progressively brighter foci throughout the infection and continuous rRNA synthesis (
The requirements for protein accumulation in nucleoli are still unclear, although binding to nucleolar proteins or nucleolar RNAs (rRNA) is probably a common theme (Valdez et al., 1994; Li et al., 1996). In this regard, the close association between B23 and cardiovirus 2A (or the 2A dimer?) by microscopic localization, could not be coincidental (
In parallel experiments, two different mAb-2A samples (5A12, 5F5) were tested for their ability to immunoprecipitate B23 from infected cell extracts (
Protein 2A Shuts Off Cap-Dependent Translation
In accordance with the present invention, applicants believe that viruses need such an unusual nuclear function, or nucleolar-targeting protein because RNA viruses live short, intense predominantly cytoplasmic lifecycles. Within 6-8 hrs, they infect, translate, synthesize RNA, package their progeny and exit the cell, all while in direct competition with host for metabolic resources. Preferably, they can simultaneously avoid activating the plethora of host-encoded antiviral defense systems that could trigger an abortive infection. A virus can gain an edge in this competition if it inhibits detrimental cellular responses or effectively sequesters metabolic resources. Picornaviruses do both. Observations dating back at least 40 years report a marked shutoff of host mRNA synthesis following Mengo infection of L-929 cells (Baltimore and Franklin, 1962). Careful measurement of the DNA-dependent RNA polymerase levels (pol-II) in infected cells showed these enzymes were not damaged or denatured, but nonetheless were unable to initiate new mRNA synthesis in the manner of mock-infected cells (Apriletti and Penhoet, 1978; Baltimore and Franklin, 1962). At the same time, there was a strong, parallel shift from cap-dependent (host) to IRES-dependent (viral) translation, during the EMCV infectious cycle (Svitkin et al., 1978). More recently, Nahum Sonenberg's group conducted an elegant series of experiments clearly connecting EMCV 2A to the shutoff of host protein synthesis during viral infection (Svitkin et al., 1998). Those experiments implicated the P13 kinase-FRAP signaling pathway in this process, but did not determine the precise mechanism by which cap-dependent inhibition was achieved. The 2A proteins of polio- and rhinoviruses are proteases, which, among other duties, cleave translational factor eIF-4G to inactivate the host's cap-binding complex (Krausslich et al., 1987; Haghighat et al., 1996; Liebig et al., 1993). In aphthoviruses, the leader protein carries out similar cleavages (Guame et al., 1998). By inactivating eIF-4G, these viruses effectively prevent host mRNA translation while they themselves continue to translate by virtue of their IRESs. But in EMCV-infected cells, eIF-4G is not cleaved, 2A is not a protease (nor is the leader), and the shut off of host protein synthesis, while clearly evident, is neither as rapid nor as extensive as that caused by polio (Jen et al., 1980).
To resolve these questions, we are only now in the process of probing cells for specific phenotypes associated with the cardiovirus 2A protein, in the absence of infection. The pIRES-2A cDNA described above, is certainly a useful tool, because of the rapidity and totality with which the induced 2A kills cells. On the other hand, rapid cell death can sometimes make it even harder to tease apart the lethal pathways if there is an inadvertent triggering of unwarranted apoptotic responses. Among our more subtle approaches are new EMCV cDNAs encoding RNA replicons or bicistronic reporter mRNAs. A replicon is a non-infectious viral RNA transcript, where a portion of the capsid-coding region has been replaced by a reporter gene, typically luciferase (
Materials and Methods
Cells and Viruses
Recombinant viruses vMC0 (Mengovirus), vMwt (Mengovirus), vEC9 (EMCV), vEC4 (EMCV) and wild-type HRV-16 (human rhinovirus) have been described (Hahn and Palmenberg, 1995; Martin et al., 1996; Lee et al., 1995), as has EMCV mutant strain vE-2A□58 with a deletion in the 2A coding-region relative to vEC9 that removed 58 amino acids (Svitkin et al., 1998). The 19 amino acid COOH-terminal primary cleavage cassette (PCC), was intact and functional in this deleted virus (Hahn and Palmenberg, 2001). Viruses were cultured in HeLa cells at 37° C. under 5% CO2-air, using RPMI-1640 medium supplemented with penicillin, streptomycin and 10% fetal bovine serum as described (Rueckert and Pallansch, 1981). Typically, subconfluent cell monolayers (5×106 cells per 6 cm petri dish) were infected with virus at a multiplicity of infection (m.o.i.) of 10 plaque-forming units (PFU) per cell (Rueckert et al., 1980).
Cellular Techniques
Subconfluent HeLa monolayers were infected with vEC9 (m.o.i. of 10) as above. At appropriate times post-infection (PI), the cells were collected and fractionated into nuclear, cytoplasmic and membrane components as described (Hu et al., 1998). Briefly, HeLa cells (5×105) were washed 3 times with cold PBS, scraped into a tube, and pelleted (400×g, 1 min, 4° C.). The cells were resuspended in buffer A (500 μl, 10 mM Tris-HCl, pH 7.8, 5 mM MgCl2, 10 mM KCl, 0.3 mM EGTA, 0.5 mM DTT, 0.3 M sucrose, 10 mM β-glycerol phosphate, 2 mM ZnCl2) and then incubated on ice for 15 min before NP40 detergent was added to 0.5%. The samples were vortexed, and nuclei were harvested after a centrifugation step (7200×g for 20 sec, 4° C.). The cytoplasmic (supernatant) fractions were removed, and the pellets (nuclei) were washed 3 times with buffer A, then resuspended in buffer B (125 μl, 20 mM Tris-HCl, pH 7.8, 5 mM MgCl2, 320 mM KCl, 0.2 mM EGTA, 0.5 mM DDT, 2 mM ZnCl2). After incubation on ice (15 min, 0° C.) and sonication (10 sec), nuclear extracts and membrane fractions were separated by centrifugation (13,500×g, 15 min, 4° C.). The membrane fractions (pellets) were washed 3 times with buffer B and resuspended (125 μl) in the same buffer.
EMCV Replicons
An EMCV-based replicon (pE-luc) replaced the 1CD coding region of pEC9 with an in-frame firefly luciferase gene, proceeded by an autocatalytic primary cleavage cassette (PCC) derived from the pMC0 2A-2B junction (Aminev et al., 2003). Translation of T7 RNA transcripts from this replicon produced a full-length, active luciferase enzyme with 10 additional viral amino acids at the N-terminus (from the Mengovirus PCC), and 29 additional viral amino acids at the C-terminus (from EMCV 1D). A derivative, pE-luc-Δ3Dpol, had a 555 base deletion (Eco47 III-Mlu I fragment) that effectively removed the active site of viral 3Dpol. Proteins directed by these transcripts were inactive for viral RNA synthesis. Luciferase assays in extracts from transfected HeLa cells were as described (Duque and Palmenberg, 2001).
Mutations
Engineered mutations that inactivated the 2C/3A (QG-QV), 3A/3B (QG-QV), 3B/3C (QG-RG), or the 3C/3D (QG-QC) cleavages sites within the polyprotein sequence of vEC4, have been described (Hall and Palmenberg, 1996a). The genome and polyprotein numbering system used here is according to the published sequence of EMCV-R (GenBank #X00463). Site-specific mutations near the N-terminus of the 3Dpol gene (putative NLS sequence) and at the 2B/2C cleavage site (QS-QY) were engineered with a 2-step overlapping PCR method similar to that used to create the above cleavage site mutations (Hall and Palmenberg, 1996a). For 3Dpol mutation R16D, amplicons resulting from primer pairs 573 (CGT ACC Aga taa aAC AGC ACT ACG CCC CAC, SEQ. ID. NO. 2) plus 95 (CCC AGT GGA AAC CGG G, SEQ. ID. NO. 3), and 575 (GTG CTG Ttt tat cTG GTA CGT GAA TAC GG, SEQ. ID. NO.4) plus 576 (CTA CGA CAT CTG TTC TGC GCA TTC C, SEQ. ID. NO. 5) were combined, then reamplified with primer pair 95 plus 576. For 3Dpol mutation K17E, the primer pairs were 572 (CGT ACC Acg tga aAC AGC ACT ACG CCC CAC, SEQ. ID. NO. 6) plus 95, and 574 (A GTG CTG Ttt cac gTG GTA CGT GAA TAC Gg, SEQ. ID. NO. 7) plus 576. The resultant fragments were digested with Bgl I and Xma I, then ligated into similarly digested pEC4 or pE-luc plasmids to create full-length mutated viral genomes (pEC4-R16D and pEC4-K17E), or mutated replicons (pE-luc-R16D and pE-luc-K17E). For 2B/2C cleavage site mutation (QS-QY), the primer pairs were 375 (CAA CAA CAA tat CCC TTG AAA, SEQ. ID. NO. 8) plus 377 (AAC TAC CCG TCA ATG GAC TCT, SEQ. ID. NO. 9), and 376 (TTT CAA GGG ata TTG TTG TTG, SEQ. ID. NO. 10) plus 378 (TTG ATA AAG ATT TCC CTT GCC, SEQ. ID. NO. 11), respectively. These two fragments were mixed then amplified by PCR using primers 377 and 378. The amplicon was digested with Sac I and Sac II, then ligated into similarly digested pEC4 cDNA.
Antibodies
Murine monoclonal antibodies (mAbs) raised against recombinant Mengovirus protein 3Dpol (8D10) and structural precursor 1CD were generously supplied by Drs. H. Duque, V. Frolov and O. Frolova (Hall and Palmenberg, 1996b; Lee et al., 1995). Murine monoclonals raised against Mengovirus (or EMCV) proteins 2B, 2C, 3A, 3AB, 3BVPg, and 3Cpro are described herein. Antibodies (mAb) reactive with HRV-16 protein 3Cpro have also been described herein. Rabbit polyclonal antibodies to proteins, B23 (nucleophosmin) and C23 (nucleolin), and actin were purchased (Santa-Cruz Biotechnology, Inc.). Anti-mouse, anti-goat, and anti-rabbit secondary antibodies, conjugated with fluorescein-5-isothiocyanate (FITC) or tetramethylrhodamine-5- (and 6-)isothiocyanate (TRITC) were purchased (Sigma Inc). Western assays used HeLa cell extracts harvested at appropriate times post infection (PI). The cells were washed with phosphate-buffered saline (PBS), lysed by freeze-thaw (3×), and the clarified supernatants were fractionated by SDS-PAGE, then blotted onto polyvinylidene fluoride membranes (Immobilon-P, Millipore) as described (Duque and Palmenberg, 2001). Bands with positive reactions against the appropriate antibodies (typically, 1:2,000 dilution of ascites), were visualized by chemiluminescence (ECL kit, Amersham Pharmacia Biotech, Inc.), after secondary reactions with appropriate anti-mouse, anti-rabbit, or anti-goat antibodies, conjugated with horseradish peroxidase (1:2,000 dilution, Sigma Inc.).
Labeling of Nascent RNA
In situ RNA synthesis was visualized in cells as described previously (Ko et al., 2000). Briefly, the transfected (or infected) HeLa cells were cultured on cover slips, then washed consecutively with PBS (pH 7.4, 2×), with permeabilization buffer (20 mM Tris-HCl, pH 7.4, 5 mM MgCl2, 0.5 mM EGTA, 0.5 mM PMSF), then with the same buffer containing triton X-100 (0.05%), before pulse-labeling was initiated by immersing the slips in reaction buffer (50 mM Tris-HCl, pH 7.4, 100 mM KCl, 5 mM MgCl2, 2 mM MnCl2, 0.5 mM EGTA, 25 U/ml RNasin, 0.5 mM of ATP, CTP, GTP, and 0.2 mM flourescein-12-UTP, Sigma Inc.). After incubation (30 min at 37° C.), the cells were washed (2×, PBS, 25 U/ml RNasin), then fixed (PBS, 10% formaldehyde, 25 U/ml RNasin, 20° C., 20 min). Subsequent cell permeabilization, blocking, antibody reactions, and visualization by confocal immunofluorescence microscopy were as described below.
Immunofluorescence Microscopy
HeLa monolayers were grown on cover slips to 50-70% confluence then infected (m.o.i. of 10), or transfected (1.5 μg RNA per 5×106 cells) (Duque and Palmenberg, 2001) as required. At appropriate intervals (37° C. under 5% CO2-air), the cells were rinsed with PBS, fixed with formaldehyde, permeabilized by 0.3% Triton X-100, then incubated (1 hr, 20° C.) with the desired primary antibody (1:2,000 dilution in blocking solution), as described (Amineva et al., 2003). After reaction with a corresponding secondary antibody, the slips were mounted (Vectashield mounting medium, Vectorlab), and label location was visualized by laser-scanning confocal microscopy. Image capture used MCR 1024 LaserSharp software (Bio-Rad, Inc.). For double-label experiments, primary antibodies raised in different species (e.g., mouse and rabbit) were selected, and the samples were developed with appropriate, corresponding secondary antibodies (1:100 dilution) conjugated with FITC or TRITC. A fluorescent-tagged wheat germ agglutinin (WGA, Molecular Probes, Inc.) was used to highlight and identify Golgi and nuclear membrane locations. SYTOX stain (Molecular Probe Inc.) was used to localize dsDNA within cells. Photography used an 488 nM filter for detection FITC and an 647 nM filter for detection of TRITC or SYTOX.
RNA and Microarray Analyses
Confluent HeLa cell monolayers (3×106 cells per 60 mm plate) were infected (m.o.i. of 10) with vEC9, vE-2A□58 or HRV-16, then incubated (37° C., under 5% CO2-air) for 2 or 5 hrs. The total RNA (3 plates per sample) was isolated and purified using RNasy Total RNA Isolation Kits (Qiagen Inc). Quantitative northern assays (dot blots) used DNA primers specific for 5S rRNA (M51545, GCAACCCTACAGAACCCGGTG, SEQ. ID. NO. 12), 18S rRNA (M10098, GCCGGTCCAAGAATTTCACCTCTA, SEQ. ID. NO. 13), 28S rRNA (M27830, GTCGAGGGCTGACTTTCAATA, SEQ. ID. NO. 14), or for the mRNA of glyceraldehyde-3-phosphate dehydrogenase (M33197, CCACGATACCAAAGTTGTCATGGAT, SEQ. ID. NO. 15), as described previously (Sambrook et al., 1989). The Superscript Choice System (Gibco BRL Life Technologies) was used to convert a sample of each cellular RNA fraction (10 μg) into cDNA using SSII reverse transcriptase with an oligo(dT)-T7 promoter primer. The second strand of cDNA was synthesized in reactions (self-primed) with DNA polymerase I, RNase H, and DNA ligase according to the manufacturer's directions. Biotin-labeled RNA probes were generated from these DNAs using RNA Transcript Labeling Kits (Affymetrix Inc.) with T7 RNA polymerase and biotin-NTPs. After fragmentation by boiling, the labeled-RNAs were hybridized to microarrays (Affymetrix, H-95A) that were developed, scanned and analyzed with GeneChip software (Affymetrix), after staining with R-phycoerythrin streptavidin conjugate (Molecular Probe).
Results and Discussion
Proteins 3BVPg, 3Cpro and 3Dpol Localize to Nucleoli
In accordance with the methods of the invention, applicants have cloned 7 Mengoviral and EMCV proteins into bacterial expression systems and used the material to immunize mice for the isolation of monoclonal antibodies (mAbs). Antibodies (mAbs) against EMCV 3BVPg were raised to a synthetic protein. Given the 93.6% average amino acid identity between polyproteins, it was not surprising to find most mAbs in the panel reacted equally well against EMCV (i.e. vEC9) or Mengovirus (i.e. vMwt) proteins, and indeed this was a selection criterion for any mAb used in subsequent cell labeling or Western analyses as described herein. The initial rational for development of these mAbs was to help follow polyprotein precursor fates through the labyrinth of the 3Cpro processing events in cell-free extracts. Eventually, however, it also seemed like a good idea to pursue some of the precursors temporally and physically within infected cells by immunofluorescence microscopy, using the mAbs as labels. The first images were astounding. Without question, mAbs against four mature viral proteins, 2A, 3BVPg, 3Cpro and 3Dpol, lit up nuclear as well as cytoplasmic loci during the entire time course of Mengovirus or EMCV infection (e.g.,
From the earliest times after EMCV infection of HeLa cells (or L-929 and BHK-21 cells, not shown), mAbs against 3BVPg, 3Cpro and 3Dpol formed bright, punctate spots in nuclei (
Precursor Identification
Cell labeling by mAbs could not distinguish whether the mature proteins or their precursors (e.g., 3CD, 3BCD, or 3BC) were responsible for nuclear and nucleolar targeting. Nor could they distinguish turnover between the nuclear and cytoplasmic protein pools as the infection progressed, and indeed, kinetic labeling of this sort is not trivial. Rather, as an initial approximation of the process, we instead tried to characterize the repertoire of precursors and mature proteins that could have given rise to the various signals. Infected HeLa cells were lysed, fractionated by SDS-PAGE, then probed in Western assays for reactive bands using each of the P2 or P3 protein mAbs in turn.
Western assays with mAbs against 3BVPg and 3Cpro using fresh gel blots or membranes that were stripped and reprobed, gave similar results, in that viral precursors dominated the immunogenic signals from infected cells, especially at the earliest times PI (
Applicants have found that mAbs proved quite adept at detecting and identifying transient precursors such as 2C3A, 2C3AB, 3BC, 3BCD and 3ABC, and at monitoring their conversion into mature forms, like 3AB, 3CD, 3Cpro and 3Dpol as described herein. To determine where these protein pools were distributed, infected cells were fractionated in the presence of a mild detergent (NP40) before SDS PAGE and sequential probing with the mAbs (
To test the effects of cleavage site mutations on the nuclear localization of 3BVPg, 3Cpro and 3Dpol mAb signals, point mutations were engineered into the codons for each of five processing site in the P2 and P3 regions of vEC4, using alterations known to abrogate 3Cpro activity (Parks et al., 1989). The 3A/3B, 3B/3C and 3C/3D mutations are lethal to infectivity (Hall and Palmenberg, 1996a). The 2B/2C and 2C/3A mutations allow weak viral growth, but only of revertant progeny, and all five cleavage site mutations are known to impede viral RNA synthesis (Hall and Palmenberg, 1996a; Parks et al., 1989). None, however, was able to prevent nuclear labeling by 3BVPg, 3Cpro and 3Dpol mAbs, after transfection with genome-length RNA transcripts (
3Dpol NLS Identification
As reported for the EMCV 2A nuclear signals, mAbs against 3BVPg, 3Cpro and 3Dpol always co-localized with B23 (nucleophosmin) and C23, the ribosomal chaperone proteins (Hingorani et al., 2000; Szebeni et al., 1995).
To explore whether this region could be involved in nuclear or nucleolar transport of 3BCD, point mutations converting two of the YRP motif codons (R16D and K17E) were engineered into the 3Dpol region of recombinant vEC9 (
rRNA Transcription
Rhinovirus and enterovirus infections cause rapid shutoff of pol-I (Rubinstein et al., 1992), pol-II (Yalamanchili et al., 1997a; Yalamanchili et al., 1997c) and pol-III transcription (Clark et al., 1991). Pol-II shutoff by Mengovirus infection was first documented more than 40 years ago in experiments that led to the initial characterization of this important cellular enzyme (Baltimore and Franklin, 1962), but the effects of cardioviruses on pol-I and pol-III are less well understood. Cells pulse-labeled with FITC-UTP showed a very different pattern following vEC9 infection (
Direct attempts to quantitate the continued pol-I and pol-III synthesis by these nucleoli were complicated by several factors. The first was the surprising observation that the total RNA in EMCV (or Mengovirus)-infected cells, as measured by OD260 on a per cell basis, was 25-50% lower than for mock-infected or HRV-16 infected cells (Table 1). The decrease in RNA content was evident as early as 2 hr PI (6% relative to mock), even before the onset of vRNA synthesis. Larger losses were measured by 5 hr PI (22% relative to mock). The shortfall in cellular RNA could be documented by a variety of techniques (not shown), and was not attributed to viral-induced cell lysis, which does not occur until 7-8 hrs PI. Viral mutants with 2A gene deletions (e.g., vE-2A□58), are defective in nucleolar localization and defective in host protein translational shutoff (Svitkin et al., 1998), but found to have even lower RNA contents (63.5% relative to mock infected cells). Another complication with rRNA measurements was the knowledge that the cellular mRNA pool changed dramatically over infection, as did the vRNA pool. Polyadenylated host mRNA usually represents about 1-3% of total RNA in an uninfected cell (Darnell et al., 1990), but after pol-II shutoff by infection, the turnover rate for existing mRNA was unknown. Therefore, OD alone could not be used, even as a rough estimate, of rRNA content.
Measurement of the relative cellular mRNA content was eventually resolved with microarray analyses (
aHeLa cells (15 × 106) were infected as described in Methods. At 2 or 5 hr PI, the total RNA per sample was isolated, purified then quantitated using a scanning spectrophotometer.
bThe samples in A were brought to equivalent concentrations (OD260), diluted serially, then probed in quantitative Northern assays for reactivity with rRNA-specific radiolabeled probes as described in the Methods herein. The data were captured by phosphoimaging and digitized. The intercept value for the linear regression analysis of each dilution curve is represented as percent of the mock-infected value for that rRNA.
cThe data in B were corrected for relative concentration in the original, total RNA pools.
P mRNA Transcription
New ribosome synthesis requires proteins in addition to rRNA. It was therefore of interest to learn whether the ribosomal protein genes could have been selectively spared from cardiovirus-induced shutoff of pol-II, or alternatively, whether these proteins were recycled from ribosome turnover. Accordingly, HeLa cells were infected with vEC9, HRV-16, or vE-2A□58. After 2 hrs to 5 hrs PI, the total RNA was isolated from these cells. The poly(A)-containing RNA in equivalent samples (10 μg of total RNA) was copied into cDNA with reverse transcriptase, then probed for transcript content using microarrays (Affymetrix, Hu-95A). Each array used 20 match/mismatch probe pairs to assess whether each of 10,000 human gene transcripts was detected (present) or not detected (absent), and to assign a measure of relative abundance to each transcript according to statistical thresholds (Snedecor et ai., 1980). RNA from mock-infected cells (10 μg, 5 hr PI) was included as a baseline, and when the array data were compared, this was the standard against which statistical calls of “increase”, “decrease”, or “no change”, were assigned to the infected samples (
Microarray analyses with vEC9 (a and b) and vE-2AΔ58 (c) infected HeLa cell RNA samples are described in Methods. Select examples of “increased” or activated transcripts (p-change values of <0.0025 relative to control samples) and “decreased” or inactivated transcripts (p-change values of >0.9985) are listed, along with relevant GenBank accession numbers.
Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it will be readily apparent to those of ordinary skill in the art in light of the teachings of this invention that certain changes and modifications may be made thereto without departing from the spirit or scope of the appended claims.
This application claims the priority benefit of U.S. Provisional Application Ser. No. 60/485,681, filed Jul. 9, 2003, incorporated by reference herein.
This invention was made with government support under grant AI-17331 awarded by the National Institutes of Health. The United States Government has certain rights in this invention.
| Number | Date | Country | |
|---|---|---|---|
| 60485681 | Jul 2003 | US |
