Double-Stranded RNA-Specific Adenosine Deaminase: Nucleic Acid Binding Properties

METHODS: A Companion to Methods in Enzymology 15, 199 –205 (1998) Article No. ME980624

Double-Stranded RNA-Specific Adenosine Deaminase: Nucleic Acid Binding Properties Yong Liu,* Alan Herbert,† Alexander Rich,† and Charles E. Samuel*,‡,1 *Department of Molecular, Cellular and Developmental Biology, University of California, Santa Barbara, Santa Barbara, California 93106; ‡Interdepartmental Graduate Program of Biochemistry and Molecular Biology, University of California, Santa Barbara, Santa Barbara, California 93106; and †Department of Biology, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139

The RNA-specific adenosine deaminase (ADAR1, herein referred to as ADAR) is an interferon-inducible RNA-editing enzyme. ADAR catalyzes the C-6 deamination of adenosine in double-stranded (ds) structures present in viral RNAs and cellular pre-mRNAs as well as synthetic dsRNA substrates. ADAR possesses three functionally distinct copies of the highly conserved double-stranded RNA binding R motif (RI, RII, RIII) implicated in the recognition of dsRNA structures within the substrate RNAs. ADAR is also a Z-DNA-binding protein. Two Z-DNA binding motifs (Za and Zb) present in ADAR correspond to repeated regions homologous to the N-terminal region of the vaccinia virus E3L protein. Here we describe assay methods for measurement of ADAR enzymatic activity, dsRNA binding activity, and Z-DNA binding activity. © 1998 Academic Press

Double-stranded RNA-specific adenosine deaminase (ADAR) (1) is an interferon-inducible enzyme that possesses both double-stranded (ds)RNA binding and Z-DNA binding activities (2–9). ADAR catalyzes the covalent modification of dsRNA substrates by hydrolytic C-6 deamination of adenosine to yield inosine (10, 11). ADAR is the candidate enzyme for two types of RNA-editing processes. First, A-to-I modifications are found at multiple sites in viral RNAs, as exemplified by the biased hypermutations observed in measles virus and other negative-stranded RNA virus genomes during lytic and persistent infections (12–14). Second, highly site-specific C-6 adenosine deaminations are ob1

To whom correspondence should be addressed. Fax: (805) 893– 4724. 1046-2023/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.

served at one or a few sites in certain viral and cellular mRNAs as exemplified by hepatitis delta virus (HDV) RNA (14 –16) and the pre-mRNAs encoding GluR receptor channel and serotonin receptor (14, 18 –21), respectively. Editing of these RNAs results in mRNAs that encode protein products that display altered functional activities (14, 20, 21). Moreover, the A-to-I deaminations observed in both viral genomes and cellular pre-mRNA transcripts are dependent on double-stranded regions within the substrate RNA (21–23). As summarized by the schematic diagram shown in Fig. 1A, characterization of ADAR cDNA clones (7, 26, 27) reveals in the central region of the deduced protein three copies of the dsRNA binding R motif, a subdomain now found in several known double-stranded RNA-binding proteins (28). The R motif, also known as dsRBM, was first described in the interferon-inducible RNA-dependent protein kinase PKR (29). The core amino acid residues of the R-motif subdomain of the PKR kinase, established by mutagenesis as crucial in the case of the dsRNA binding activity of PKR (30 –33), are fully conserved in each of the three R repeats found in ADAR (7, 24). Dissociation constants of PKR R-motif subdomains for synthetic and natural RNAs are in the nanomolar to micromolar range, dependent on the specific RNA examined (34, 35). ADAR is also a DNAbinding protein (8). ADAR specifically binds to the left-handed Z-DNA conformation (8, 9). A repeated domain present in the N-terminal region of ADAR that is homologous to the N-terminal region of the vaccinia virus E3L protein (7) corresponds to two Z-DNA binding domains, Za and Zb (9). Za pos199

200

LIU ET AL.

sesses a higher affinity (Kd 5 4 nM) than Zb for Z-DNA (9). The presence of two immunologically related forms of the human ADAR deaminase was demonstrated in a variety of human cell lines: an interferon-inducible ;150-kDa protein present in both the cytoplasm and nucleus; and a constitutively expressed ;110-kDa protein present predominantly if not exclusively in the nucleus (7). The gene encoding these proteins, now referred to as ADAR1 (1), maps to human chromosome 1q21.1–21.2 (36). A second ADAR gene, referred to as ADAR2 (1), which encodes an ;80-kDa protein, maps to human chromosome 21q22.3 (37). The ADAR1 and ADAR2 dsRNA adenosine deaminases display differential editing activities for different RNA substrates (14, 19, 21, 38). Furthermore, the existence of three naturally occurring alternative splice-site variant forms of ADAR1, herein referred to as ADAR, has been demonstrated (25).

Two ADAR splice-site variants are differentially expressed in different tissues (25). In comparison to the full-length 1226-amino-acid ADAR protein (7, 26, 27), designated ADAR-a, the variant designated ADAR-b is a 59-splice-site variant that contains a deletion of 26 amino acids between the RIII motif subdomain and the catalytic C domain. The ADAR-c variant is a 39-splice-site variant that has an additional deletion of 19 amino acids between the RII and RIII motif subdomains. Thus, a family of dsRNA adenosine deaminase enzymes with selective substrate specificities likely exist that bind and edit different viral and cellular RNA targets. We describe in this article two methods that we have used for the expression of recombinant ADAR that permit measurement of the enzyme activity with minimal endogenous background activity, as well as methods for measurement of the dsRNA binding and Z-DNA binding activities of the recombinant ADAR proteins.

FIG. 1. Schematic structure of ADAR cDNA. (A) Schematic representation of the 6613 nucleotide dsRNA-specific adenosine deaminase cDNA and deduced 1226-amino-acid protein. The N-terminal portion of ADAR encodes two Z-DNA binding domains (Za and Zb), and the central region encodes three copies of dsRNA binding motif (RI, RII, and RIII), also referred to as dsRBM motifs. The C-terminal region constitutes the catalytic (C) domain. The two solid lines indicate the full-length (M1) and N-truncated (M296) versions of ADAR protein that have been expressed and analyzed (7, 24, 25). Numbers shown below the schematic diagram are nucleotide (nt) numbers corresponding to the UTR regions and the ORF. (B) Schematic structure of the three splice variants of ADAR, a, b, and c, that have been found differentially expressed in different tissues (25). [Adapted with permission, from (25).]

ADENOSINE DEAMINASE BINDING PROPERTIES

DESCRIPTION OF METHODS Construction of ADAR Expression Vectors Construction of ADAR expression vectors using the pcDNA I/Neo vector in which ADAR translation initiates at Met-1 to give the full-length form (FL) or at Met-296 to give the N-terminally truncated form (M296) of the ADAR protein (Fig. 1A) has been previously described (24, 25). Wild-type M296 constructs include the three identified naturally occurring splice variants M296-a, -b, and -c, which use alternative exons 6 and 7 (Fig. 1B). The engineered M296-d variant, which has not yet been found in nature, contains solely the 57-nucleotide (nt) deletion between the RII and RIII motifs associated with exon 6b (25). Mutant M296 constructs in which the RI, RII, and RIII subdomains possess an amino acid substitution at a highly conserved and critical lysine residue within each of the R cores (24) are also available for each of the three splicing variants of ADAR, i.e., ADAR-a, -b, and -c (25). These mutants include seven possible constructs possessing mutations in the three R motifs, i.e., the three singlesubstitution mutants with each of the three R motifs mutated at RI(K554E), RII(K665E), or RIII (K776E); the three double-substitution mutants with two of the R motifs altered; and the triple mutant in which all three of the R motifs are altered. The doublesubstitution mutant of ADAR deficient in catalytic activity (designated ‘‘C’’ mutant) possesses the H910Q, E912A substitutions within the highly conserved CHAE sequence of the postulated C-terminal deaminase catalytic domain (24).

201

tion of about 20 mg/ml and [35S]methionine (Amersham, 1000 Ci/mmol) at 15 mCi/ml. Incubation is for 60 min at 30°C. For analysis of the ADAR proteins synthesized in vitro, 35S-labeled protein products are fractionated by 10% SDS–PAGE and visualized by autoradiography (see Fig. 4A). Quantitation is performed using either a laser densitometer (LKB ultroscan XL) or a molecular imager (BioRAD GS525) system. ADAR proteins synthesized in vitro possess dsRNAspecific adenosine deaminase activity, as illustrated by the representative deaminase assay results obtained for wild-type and mutant ADAR proteins shown in Fig. 2. In this assay for deaminase activity, inosine formation from adenosine in a 32P-labeled dsRNA substrate is measured by thin-layer chromatography (TLC) to separate IMP and AMP following hydrolysis of the edited RNA product. The synthetic dsRNA substrate is prepared by in vitro transcription using the pBS vector containing an A 1 T-rich 512-nt HindIII fragment from a PKR genomic subclone (generously provided by Dr. K. Kuhen, University of California, Santa Barbara). The plasmid is linearized with either

Expression in Vitro and Functional Analysis of ADAR Proteins In vitro transcription and translation are carried out (24) to produce 35S-labeled ADAR proteins, as well as to confirm the protein coding capacity of each mutant construction. Briefly, XhoI-linearized plasmid DNA (5 mg) of the wild-type or mutant versions of ADAR-a, -b, or -c, either M1 or M296, is transcribed in vitro using phage T7 RNA polymerase (New England Biolabs) according to the manufacturer’s instructions. Subsequent translation in vitro of the ADAR mRNA transcript is carried out using the nuclease-treated rabbit reticulocyte lysate (RRL) system (Promega) according to the manufacturer’s recommendations. The reaction mixtures (typically 50 –100 ml) contain mRNA at a concentra-

FIG. 2. Analysis of ADAR proteins synthesized in vitro for enzyme activity. Autoradiogram showing dsRNA-specific ADAR deaminase activity, measured with proteins synthesized in vitro using mRNA transcribed in vitro from wild-type and mutant ADAR cDNA clones. 32P-Labeled dsRNA substrate is incubated with equivalent amounts of ADAR WT or mutant proteins under the standard assay conditions as described. Following subsequent P1 nuclease digestion, the labeled nucleotides are analyzed by thin-layer chromatography (TLC). The positions to which the adenosine (AMP) and inosine (IMP) 59-nucleoside monophosphates migrate, as well as the origin, are indicated. Unlabeled IMP used as chromatography standard is visualized with 254-nm UV. [Adapted with permission from (24).]

202

LIU ET AL.

EcoRI or XhoI and then transcribed with either T7 or T3 RNA polymerase (Promega), respectively, resulting in the complementary transcripts. The RNA transcripts generated with T7 polymerase are 32P-labeled with [a-32P]ATP (3000 Ci/mmol, Amersham). [32P]dsRNA substrate is formed by annealing complementary single-stranded transcripts in 20 mM Tris–HCl, pH 7.9, 0.15 M NaCl by heating at 75°C for 5 min and then slowly cooling to room temperature. The standard reaction mixture (40 ml) for measurement of ADAR enzymatic activity includes 10 fmol of 32P-labeled dsRNA, varying amounts of ADAR proteins as indicated, and 50 mM Tris–HCl buffer, pH 7.9, 100 mM KCl, 5 mM EDTA, 10% (v/v) glycerol, 1 mM dithiothreitol (DTT), 0.5 mM phenylmethylsulfonyl fluoride (PMSF). After incubation at 30°C for 2 h, the dsRNA substrate is extracted with phenol:chloroform and chloroform, followed by ethanol precipitation with 0.5 mg poly(rI) (Sigma) added as carrier RNA. The precipitated RNA is washed with 70% ethanol, dried, and suspended in 10 ml of nuclease P1 buffer (30 mM potassium acetate, pH 5.3, 10 mM ZnSO4) before digestion with 1.5 mg of nuclease P1 (Pharmacia) for 1 h at 50°C. IMP and AMP are resolved from each other by TLC on cellulose NM 300 glass plates (Macherey and Nagel) in a solvent consisting of saturated (NH4)2SO4, 100 mM sodium acetate (pH 6.0), and 2-propanol [79:19:2]. Autoradiography is usually performed for 16 h at 280°C with a screen. Quantitation is carried out by excising the TLC spots and measuring radioactivity with a liquid scintillation system (Beckman LS1801) or with a molecular imager system (Bio-Rad GS525). The results shown in Fig. 2 illustrate that the three copies of the RNA binding R subdomain are functionally distinct from each other and also from the catalytic domain of dsRNA-specific adenosine deaminase. Based on the enzyme activities observed with a synthetic [32P]dsRNA substrate, the RIII copy is the most important of the three R motifs and the RII copy is the least important for ADAR deaminase activity. Site-directed mutations within the CHAE sequence of the postulated C-terminal deaminase catalytic domain destroy the deaminase activity of ADAR as illustrated by the lack of inosine formation with the ‘‘C’’ mutant.

Expression in Vivo and Functional Analysis of ADAR Proteins Monkey kidney COS-1 cells, maintained in monolayer culture in Dulbecco’s modified Eagle’s medium

(DMEM) supplemented with 10% fetal bovine serum (Hyclone), are used for the expression in vivo of ADAR proteins. Transfection with wild-type or mutant cDNA expression vectors is carried out by the DEAE-dextran– chloroquine phosphate method (39), using 5 mg of DNA per milliliter, when the cell cultures in 60-mm dishes are about 70 – 80% confluent. Transfected cells are harvested about 60 h after transfection. Cells are washed twice with phosphate-buffered saline (PBS) and cell-free extracts are prepared by Nonidet P-40 lysis with 0.4 ml of buffer A [20 mM Hepes, pH 7.5, 1 mM magnesium chloride, 10 mM potassium chloride, 0.5 mM DTT, 10% (v/v) glycerol, 0.25 mM PMSF, and 1% (v/v) protease inhibitor cocktail for mammalian cells (Sigma)] containing 0.1% Nonidet P-40. After agitation for 10 min on ice and centrifugation at 3000 rpm for 5 min, the supernatant solution is adjusted to a final concentration of 0.1 M KCl using a 1 M stock solution of KCl. After centrifugation at 14,000 rpm for 15 min, the resultant supernatant solution is used as a cytoplasmic extract. The nuclear pellet is washed twice with buffer A, suspended in 0.2 ml of buffer A containing 0.1 M KCl and disrupted by sonication, and then the crude nuclear extract is recovered by centrifugation at 14,000 rpm for 15 min. The expression level of ADAR proteins in transfected cells can be monitored by Western immunoblot analysis (7), as illustrated by the results shown in Fig. 3A. Typically 20 ml of the cytoplasmic or nuclear extract from COS-1 cells is fractionated by 7.5% SDS– PAGE, transferred to nitrocellulose filter membranes, and probed with antiserum (1:500 dilution) generated against recombinant ADAR protein expressed in Escherichia coli (7). Antibody–antigen complexes are subsequently detected with 125Ilabeled protein A (0.05 mCi/ml, ICN) and autoradiography; autoradiograms are quantified using a molecular imager system. As shown by the Western analysis in Fig. 3A, the three ADAR splice-site variant proteins are efficiently expressed in transfected COS-1 cells. The recombinant ADAR proteins are distributed in comparable amounts between the cytoplasmic (Fig. 3A, lanes b– e) and nuclear (not shown) fractions (7). However, an endogenous p110 protein (Fig.3A, lane f) that is of comparable size to M296 ADAR-a and is immunologically indistinguishable from M296 ADAR (7, 25) is present in nuclear extracts prepared from most mammalian cells, including COS cells, but not in the cytoplasm (Fig. 3A, compare lanes a and f). Therefore, the cytoplasmic fractions are rou-

ADENOSINE DEAMINASE BINDING PROPERTIES

203

tinely used for functional analysis of recombinant ADAR proteins expressed in transfected cells. The relative dsRNA-specific adenosine deaminase activity of the recombinant ADAR isoforms present in the cytoplasmic fraction prepared from transfected COS cells is shown in Fig. 3B, for both wild-type and R-mutant proteins. The three splice-site ADAR variants, ADAR-a, -b, and -c, encode active enzymes that possess comparable deaminase activity with a dsRNA substrate. Site-directed mutagenesis reveals that the functional importance of individual R motifs depends on the splice variant. For example, mutation of RII causes an enhancement of activity in ADAR-b and ADAR-c and a loss of activity in ADAR-a. This result suggests that each splice variant may target different RNA substrates in vivo.

Examination of Nucleic Acid Binding Properties of ADAR 35

FIG. 3. Functional analysis of ADAR proteins expressed in transfected COS cells. (A) Western immunoblot analysis showing expression of the N-terminally truncated M296 form of ADAR, including splice-site variant versions ADAR-a, -b, -c, and -d, as detected in the cytoplasmic fraction (lanes a– e). The ADAR-a, -b, and -c splice variants are found naturally, whereas ADAR-d is an engineered variant that has not yet been found in nature. Expression levels of the endogenous p110 protein detected in the cytoplasmic and nuclear fractions are prepared from COS-1 cells transfected with vector alone (lanes a and f). (B) Functionally distinct R motifs associated with the alternative splice-site variants. Relative enzyme activities are measured using a 32P-labeled dsRNA substrate and compared for the expressed R-motif mutants, including the three single R mutants and the triple R mutant, in the N-terminally truncated M296 isoform of the ADAR-a, -b, and -c splice variants of exons 6 and 7 expressed in transfected COS cells. The activities of R mutants for each of the three ADAR isoforms are respectively normalized to that of the corresponding wild-type isoform, since the wild-type proteins including the engineered ADAR-d (M296) variant display comparable specific enzyme activities. [Adapted with permission from (25).]

S-labeled ADAR proteins synthesized in vitro can be used to study dsRNA binding activity as described previously using an RNA–Sepharose affinity resin (7, 24). The steps for analysis of dsRNAbinding proteins, including ADAR (7, 24), by this technique, which involves dsRNA coupled to a bead resin, are described in detail by Jacobs et al. (40). Here we focus on a straightforward method to examine the Z-DNA binding activity of ADAR proteins using the same protein expression system. The Z-DNA binding activity of ADAR may be measured by an affinity binding assay conceptually similar to that used to measure dsRNA binding activity. The Z-DNA binding assay employs biotinylated poly(dC– dG) that is stabilized in the Z-DNA conformation by chemical bromination (41) and streptavidin Dynabeads M-280 (Dynal). For each Z-DNA binding reaction, 20 ml of Dynabeads prewashed twice with 0.5 M NaCl is mixed with 0.5 mg of biotinylated Z-DNA and then held at room temperature for 30 min. After washing extensively with Z buffer [10 mM Tris–Cl, pH 7.9, 10 mM MgCl2, 50 mM NaCl, 1 mM DTT, 10% (v/v) glycerol, 0.25 mM PMSF, 0.5% Nonidet P-40], 35S-labeled ADAR protein synthesized in vitro (5 ml for the M296 form or 10 ml for the FL form) is added to the binding mixture in a volume of 30 ml and incubated at room temperature for 30 min. For competition analysis, binding is carried out in the presence of varying amounts of GST-Za, a 78-amino-acid peptide possessing Z-DNA binding domain (9). The Dynabeads collected using a magnetic bar are washed exten-

204

LIU ET AL.

sively with ice-cold Z buffer, and the bound proteins are denatured by boiling for 5 min in SDS loading buffer and then analyzed by SDS–PAGE. Quantitation of binding is carried out by laser densitometry of the autoradiograms or by molecular imaging with a phosphoimager. Both the FL 1226-amino-acid ADAR-a protein and the truncated M296 form of ADAR-a possess comparable Z-DNA binding activity (Fig. 4). The Z-DNA binding activity of ADAR is gradually diminished in the presence of increasing amounts of competing GST-Za peptide (Fig. 4). The FL form of ADAR has higher binding affinity for Z-DNA than the truncated M296 form (Fig. 4B). The FL form contains two Z-DNA binding domains in the N-terminal por-

tion, with Za known to have a much higher Z-DNA binding activity than Zb (9). The M296 truncation contains only Zb (Fig. 1A). Mutation of all three R motifs, as in the triple-R mutant, does not significantly affect the binding activity of FL-ADAR. This further demonstrates that the Z-DNA binding domains are functionally separate from the dsRNA binding domains. However, mutation of the three R motifs lowers the Z-DNA binding affinity of the M296 isoform that possesses only the second Z-DNA binding domain (Zb), as compared with the M296 wild-type protein (Fig. 4B). This result suggests that mutations in the dsRNA binding motifs may cause conformational changes that impair the Z-DNA binding ability of the proximal, but not the distal, Z-DNA binding domain.

CONCLUDING REMARKS The ADAR deaminase has been shown by Northern blot analysis and by direct enzyme assay to be present ubiquitously in primary tissues and mammalian cell lines (7, 26, 27, 42), and to exist in two forms, one of which is interferon inducible (6, 7). The full-length 1226-amino-acid form of ADAR possesses multiple biochemical activities: ADAR is an RNA deaminase; ADAR is a dsRNA-binding protein; and ADAR is a Z-DNA binding protein. Herein we described two expression systems for the functional analysis of recombinant ADAR. The in vitro expression system provides a straightforward approach for examination of the nucleic acid binding properties of ADAR by affinity resin approaches. The in vivo expression system generates active ADAR proteins which can be used for a variety of functional analyses based on the measurement of deaminase activity of the recombinant ADAR protein.

ACKNOWLEDGMENT FIG. 4. Analysis of Z-DNA binding activity of recombinant ADAR proteins. In vitro-synthesized wild-type (WT) and triple-R mutant (RIRIIRIII) proteins labelled with [35S]methionine are analyzed in either the (A) full-length (FL) or the (B) N-terminally truncated (M296) form. The autoradiograms show 35S-labeled ADAR proteins bound to biotinylated bromo poly(dC-dG). The relative strength of Z-DNA binding was measured by titration with increasing amounts of Za peptide. Bound complexes were recovered before SDS-PAGE using streptavidin Dynabeads. (C) Quantitation of Z-DNA binding activity of WT and mutant ADAR proteins. The error bars represent two independent assays.

This work was supported in part by Research Grant AI-12520 from the National Institute of Allergy and Infectious Diseases, National Institutes of Health.

REFERENCES 1. Bass, B. L., Nishikura, K., Keller, W., Seeburg, P. H., Emeson, R. B., O’Connell, M. A., Samuel, C. E., and Herbert, A. (1997) RNA 3, 947–949.

ADENOSINE DEAMINASE BINDING PROPERTIES 2. Hough, R. F., and Bass. B. L. (1994) J. Biol. Chem. 269, 9933–9939. 3. Kim, U., Garner, T. L., Sanford, T., Speicher, D., Murray, J. M., and Nishikura, K. (1994) J. Biol. Chem. 269, 13480 –13489. 4. O’Connell, M. A., and Keller, W. (1994) Proc. Natl. Acad. Sci. USA 91, 10596 –10600. 5. Nishikura, K., Yoo, C., Kim, U., Murray, J. M., Estes, P. A., Cash, F. E., and Liebhaber, S. A. (1991) EMBO J. 10, 3523– 3532. 6. Patterson, J. B., Thomis, D. C., Hans, S. L., and Samuel, C. E. (1995) Virology 210, 508 –511. 7. Patterson, J. B., and Samuel, C. E. (1995) Mol. Cell. Biol. 15, 5376 –5388. 8. Herbert, A., Lowenhaupt, K., Spitzner, J., and Rich, A. (1995) Proc. Natl. Acad. Sci. USA 92, 7550 –7554. 9. Herbert, A., Alfken, J., Kim, Y.-G., Mian, I. S., Nishikura, K., and Rich, A. (1997) Proc. Natl. Acad. Sci. USA 94, 8421– 8426. 10. Bass, B. L., and Weintraub, H. (1988) Cell 55, 1089 –1098. 11. Wagner, R. W., Smith, J. E., Cooperman, B. S., and Nishikura, K. (1989) Proc. Natl. Acad. Sci. USA 86, 2647–2651. 12. Cattaneo, R., and Billeter, M. A. (1992) Curr. Top. Microbiol. Immunol. 176, 63–74. 13. Cattaneo, R. (1994) Curr. Opin. Genet. Dev. 4, 895–900. 14. Rueter, S. M., and Emeson, R. B. (1998) in Modifications and Editing of RNA, pp. 343–361, ASM Press. 15. Lai, M. M. C. (1995) Annu. Rev. Biochem. 64, 259 –286. 16. Casey, J. L., and Gerin, J. L. (1995) J. Virol. 69, 7593–7600. 17. Polson, A. G., Ley, H. L., III, Bass, B. L., and Casey, J. L. (1998) Mol. Cell. Biol. 18, 1919 –1926. 18. Hurst, S. R., Hough, R. F., Aruscavage, P. J., and Bass, B. (1995) RNA 1, 1051–1060. 19. Maas, S., Melcher, T., Herb, A., Seeburg, P. H., Keller, W., Krause, S., Higuchi, M., and O’Connell, M. A. (1996) J. Biol. Chem. 271, 12221–12226. 20. Lomeli, H., Mosbacher, J., Melcher, T., Hoger, T., Geiger, J. R. P., Kuner, T., Monyer, H., Higuchi, M., Bach, A., and Seeburg, P. H. (1994) Science 266, 1709 –1712. 21. Burns, C. M., Chu, H., Rueter, S. M., Hutchinson, L. K., Canton, H., Sanders-Bush, E., and Emeson, R. B. (1997) Nature 387, 303–308. 22. Higuchi, M., Single, F. N., Kohler, M., Sommer, B., Sprengel, R., and Seeburg, P. H. (1993) Cell 75, 1361–1370.

205

23. Yang, J.-H., Sklar, P., Axel, R., and Maniatis, T. (1995) Nature 374, 77– 81. 24. Liu, Y., and Samuel, C. E. (1996) J. Virol. 70, 1961–1968. 25. Liu, Y., George, C. X., Patterson, J. B., and Samuel, C. E. (1997) J. Biol. Chem. 272, 4419 – 4428. 26. Kim, U., Wang, Y., Sanford, T., Zeng, Y., and Nishikura, K. (1994) Proc. Natl. Acad. Sci. USA 91, 11457–11461. 27. O’Connell, M. A., Krause, S., Higuchi, M., Hsuan, J. J., Totty, N. F., Jenny, A., and Keller, W. (1995) Mol. Cell. Biol. 15, 1389 –1397. 28. Kharrat, A., Macias, M. J., Gibson, T. J., Nilges, M., and Pastore, A. (1995) EMBO J. 14, 3572–3584. 29. Samuel, C. E. (1993) J. Biol. Chem. 268, 7603–7606. 30. McCormack, S. J., Thomis, D. C., and Samuel, C. E. (1992) Virology 188, 47–56. 31. Green, S. R., Manche, L., and Mathews, M. B. (1995) Mol. Cell. Biol. 15, 358 –364. 32. McCormack, S. J., Ortega, L. G., Doohan, J. P., and Samuel, C. E. (1994) Virology 198, 92–99. 33. McMillan, N. A. J., Carpick, B. W., Hollis, B., Toone, W. M., Zamanian-Daryoush, M., and Williams, B. R. G. (1995) J. Biol. Chem. 270, 2601–2606. 34. McCormack, S. J., and Samuel, C. E. (1995) Virology 206, 511–519. 35. Bevilacqua, P. C., George, C. X., Samuel, C. E., and Cech, T. R. (1998) Biochemistry 37, 6303– 6316. 36. Weier, H.-U. G., George, C. X., Greulich, K. M., and Samuel, C. E. (1995) Genomics 30, 372–375. 37. Villard, L., Tassone, F., Haymowicz, M., Welborn, R., and Gardiner, K. (1997) Somat. Cell Mol. Genet. 23, 135–145. 38. Lai, F., Chen, C. X., Carter, K. C., and Nishikura, K. (1997) Mol. Cell. Biol. 17, 2413–2424. 39. Luthman, H., and Magnusson, G. (1983) Nucleic Acids Res. 11, 1295–1308. 40. Jacobs, B. L., Langland, J. O., and Brandt, T. (1998) Methods Companion Methods Enzymol. 15, 225–232. 41. Moller, A., Nordheim, A., Kozolowski, S. A., Patel, D. J., and Rich, A. (1984) Biochemistry 23, 54 – 62. 42. Wagner, R. W., Yoo, C., Wrabetz, L, Kamholz, J, Buchhalter, J., Hassan, N. F., Khalili, K., Kim, S. U., Perussia, B., McMorris, F. A., and Nishikura, K. (1990) Mol. Cell. Biol. 10, 5586 –5590.