RNA dependence of the bacteriophage phi 29 DNA packaging ATPase.

The activity of the DNA packaging adenosine triphosphatase (ATPase) of the Bacillus subtilis bacteriophage phi 29 is dependent upon prohead RNA. The 174 nucleotide viral-encoded RNA is positioned on the head-tail connector at the portal vertex of the phi 29 precursor shell (prohead). Here, the RNA interacts with the ATP-binding gene 16 product (gp16) to constitute the DNA-packaging ATPase and initiate DNA packaging in vitro. Both the prohead connector (gene 10 product, gp10) and gp16 may utilize an RNA recognition motif characteristic of a number of RNA-associated proteins, and the binding of gp16 by proheads shields the prohead RNA from RNase A. The ATPase activity of gp16 is stimulated fourfold by RNA and tenfold by proheads with RNA. RNA is needed continuously for the gp16/RNA ATPase activity and is essential for the gp16/prohead ATPase activity. The prohead, with its connector, RNA and associated gp16 in an assembly-regulated configuration, hydrolyzes ATP and drives phi 29 DNA translocation.     

Isolation of a series of single missense mutants of a dna gene of phage psi29, gene 1, utilizing their inhibitory effect on E. coli growth

Gene 1 product (gp1) of Bacillus subtilis phage psi29 is known to promote DNA replication of the phage. Although its role in the DNA replication is not clear, gp1 is reported to exhibit multiple characteristics, including RNA binding, cell membrane localization, and self-association. To investigate these characteristics, we undertook the isolation of a series of missense mutants of gene 1 bearing substitutions at various regions. During cloning of gene 1, we found that its expression severely inhibited the growth of its host Escherichia coli cells. In this study, we utilized this growth-inhibition phenomenon to screen a random library muta-genized by error-prone PCR, expecting that mutants which could not inhibit cell growth would be affected in the authentic functions of gp1. Using this approach, we obtained 31 different mutants bearing single amino acid substitutions at 26 positions along the entire length of gp1. As a preliminary analysis of these mutants, we compared the deduced amino acid sequences of gp1s from psi29 and its related phages PZA, B103 and M2. Alignment of these sequences revealed three conserved regions, i.e. a hydrophobic region near the carboxyl terminus (assumed to be involved in the membrane localization and self-association of gp1), coiled-coil motif (essential for self-association), and a region of unknown function near the amino terminus. Interestingly, many of the substitutions in the isolated mutants occurred at strongly conserved residues in these regions and affected characteristic features of the regions (e.g. hydrophobicity of the hydrophobic region). These substitutions are expected to affect authentic functions of gp1, and the mutants will be useful for studies of the structure and functions of gp1.     

Mechanism of stimulation of DNA replication by bacteriophage phi 29 single-stranded DNA-binding protein p5

Protein p5 is a Bacillus subtilis phage phi 29-encoded protein required for phi 29 DNA replication in vivo. Protein p5 has single-stranded DNA binding (SSB) capacity and stimulates in vitro DNA replication severalfold when phi 29 DNA polymerase is used to replicate either the natural phi 29 DNA template or primed M13 single-stranded DNA (ssDNA). Furthermore, other SSB proteins, including Escherichia coli SSB, T4 gp32, adenovirus DNA-binding protein, and human replication factor A, can functionally substitute for protein p5. The stimulatory effect of phi 29 protein p5 is not due to an increase of the DNA replication rate. When both phi 29 DNA template and M13 competitor ssDNA are added simultaneously to the replication reaction, phi 29 DNA replication is strongly inhibited. This inhibition is fully overcome by adding protein p5, suggesting that protein p5-coated M13 ssDNA is no longer able to compete for replication factors, probably phi 29 DNA polymerase, which has a strong affinity for ssDNA. Electron microscopy demonstrates that protein p5 binds to M13 ssDNA forming saturated complexes with a smoothly contoured appearance and producing a 2-fold reduction of the DNA length. Protein p5 also binds to ssDNA in the phi 29 replicative intermediates produced in vitro, which are similar in structure to those observed in vivo. Our results strongly suggest that phi 29 protein p5 is the phi 29 SSB protein active during phi 29 DNA replication.     

Prohead and DNA-gp3-dependent ATPase activity of the DNA packaging protein gp16 of bacteriophage phi 29

The ATPase activity of the DNA packaging protein gp16 (gene product 16) of bacteriophage phi 29 was studied in the completely defined in-vitro assembly system. ATP was hydrolyzed to ADP and Pi in the packaging reaction that included purified proheads, DNA-gp3 and gp16. Approximately one molecule of ATP was used in the packaging of 2 base-pairs of phi 29 DNA, or 9 X 10(3) ATP molecules per virion. The hydrolysis of ATP by gp16 was both prohead and DNA-gp3 dependent. gp16 contained both the "A-type" and the "B-type" ATP-binding consensus sequences (Walker et al., 1982) and the predicted secondary structure for ATP binding. The A-type sequence of gp16 was "basic-hydrophobic region-G-X2-G-X-G-K-S-X7-hydrophobic", and similar sequences were found in the phage DNA packaging proteins gpA of lambda, gp19 of T7 and gp17 of T4. Having both the ATP-binding and potential magnesium-binding domains, all of these proteins probably function as ATPases and may have common prohead-binding capabilities. The phi 29 protein gp3, covalently bound to the DNA, may be analogous in function to proteins gpNul of lambda and gpl of phi 21 that bind the DNA.     

An efficient phage plaque screen for the random mutational analysis of the interaction of HIV-1 gp120 with human CD4.

A lambda phage expression methodology was adapted to dissect protein/ligand interactions efficiently through the creation and rapid screening of large numbers of mutants. Here we describe the method and its specific application to the interaction between the external envelope glycoprotein of the human immunodeficiency virus (HIV-1), gp120, and the human cell surface protein CD4. Random substitutions were introduced throughout the gp120 binding region (amino acids 38-62) in the amino-terminal domain of CD4 by oligonucleotide mutagenesis. These mutations were expressed within phage plaques and directly screened for their effect on binding of gp120 using a modified phage plaque lift procedure. Plaques showing increased, decreased, and no effect on binding were identified and mutations were verified by sequence analysis. In this manner, 25 unique mutations were identified that altered CD4 binding to gp120. A new site was identified at which mutations reduced binding to gp120 and several novel amino acid substitutions were defined at sites previously implicated in binding. Of particular interest, this in vitro genetic approach identified a mutation which significantly increased binding to gp120. The phenotypes of several of these mutants were further characterized by quantitative measurement of their binding affinity. The results confirmed the accuracy of the phenotypic selection and demonstrated that the sensitivity of the system allowed detection of a 3-4-fold increase or decrease in affinity. In the context of the recently determined atomic structure of CD4, these results further implicate residues in the CDR2-like region and in an adjacent loop in recognition of gp120. This methodology should be generally applicable to other high affinity protein/ligand interactions that are compatible with expression in Escherichia coli.     

Interaction of gp16 with pRNA and DNA for genome packaging by the motor of bacterial virus phi29

One striking feature in the assembly of linear double-stranded (ds) DNA viruses is that their genome is translocated into a preformed protein coat via a motor involving two non-structural components with certain characteristics of ATPase. In bacterial virus phi29, these two components include the protein gp16 and a packaging RNA (pRNA). The structure and function of other phi29 motor components have been well elucidated; however, studies on the role of gp16 have been seriously hampered by its hydrophobicity and self-aggregation. Such problems caused by insolubility also occur in the study of other viral DNA-packaging motors. Contradictory data have been published regarding the role and stoichiometry of gp16, which has been reported to bind every motor component, including pRNA, DNA, gp3, DNA-gp3, connector, pRNA-free procapsid, and procapsid/pRNA complex. Such conflicting data from a binding assay could be due to the self-aggregation of gp16. Our recent advance to produce soluble and highly active gp16 has enabled further studies on gp16. It was demonstrated in this report that gp16 bound to DNA non-specifically. gp16 bound to the pRNA-containing procapsid much more strongly than to the pRNA-free procapsid. The domain of pRNA for gp16 interaction was the 5'/3' paired helical region. The C18C19A20 bulge that is essential for DNA packaging was found to be dispensable for gp16 binding. This result confirms the published model that pRNA binds to the procapsid with its central domain and extends its 5'/3' DNA-packaging domain for gp16 binding. It suggests that gp16 serves as a linkage between pRNA and DNA, and as an essential DNA-contacting component during DNA translocation. The data also imply that, with the exception of the C18C19A20 bulge, the main role of the 5'/3' helical double-stranded region of pRNA is not for procapsid binding but for binding to gp16.     

Binding of pRNA to the N-terminal 14 amino acids of connector protein of bacteriophage phi29

During assembly, bacterial virus phi29 utilizes a motor to insert genomic DNA into a preformed protein shell called the procapsid. The motor contains one twelve-subunit connector with a 3.6 nm central channel for DNA transportation, six viral-encoded RNA (packaging RNA or pRNA) and a protein, gp16, with unknown stoichiometry. Recent DNA-packaging models proposed that the 5-fold procapsid vertexes and 12-fold connector (or the hexameric pRNA ring) represented a symmetry mismatch enabling production of a force to drive a rotation motor to translocate and compress DNA. There was a discrepancy regarding the location of the foothold for the pRNA. One model [C. Chen and P. Guo (1997) J. Virol., 71, 3864–3871] suggested that the foothold for pRNA was the connector and that the pRNA–connector complex was part of the rotor. However, one other model suggested that the foothold for pRNA was the 5-fold vertex of the capsid protein and that pRNA was the stator. To elucidate the mechanism of phi29 DNA packaging, it is critical to confirm whether pRNA binds to the 5-fold vertex of the capsid protein or to the 12-fold symmetrical connector. Here, we used both purified connector and purified procapsid for binding studies with in vitro transcribed pRNA. Specific binding of pRNA to the connector in the procapsid was found by photoaffinity crosslinking. Removal of the N-terminal 14 amino acids of the gp10 protein by proteolytic cleavage resulted in undetectable binding of pRNA to either the connector or the procapsid, as investigated by agarose gel electrophoresis, SDS–PAGE, sucrose gradient sedimentation and N-terminal peptide sequencing. It is therefore concluded that pRNA bound to the 12-fold symmetrical connector to form a pRNA–connector complex and that the foothold for pRNA is the connector but not the capsid protein.     

Multifunctional roles of a bacteriophage phi 29 morphogenetic factor in assembly and infection

Low copy number proteins within macromolecular complexes, such as viruses, can be critical to biological function while comprising a minimal mass fraction of the complex. The Bacillus subtilis double-stranded DNA bacteriophage phi 29 gene 13 product (gp13), previously undetected in the virion, was identified and localized to the distal tip of the tail knob. Western blots and immuno-electron microscopy detected a few copies of gp13 in phi 29, DNA-free particles, purified tails, and defective particles produced in suppressor-sensitive (sus) mutant sus13(330) infections. Particles assembled in the absence of intact gp13 (sus13(342) and sus13(330)) had the gross morphology of phi 29 but were not infectious. gp13 has predicted structural homology and sequence similarity to the M23 metalloprotease LytM. Poised at the tip of the phi 29 tail knob, gp13 may serve as a plug to help restrain the highly pressurized packaged genome. Also, in this position, gp13 may be the first virion protein to contact the cell wall in infection, acting as a pilot protein to depolymerize the cell wall. gp13 may facilitate juxtaposition of the tail knob onto the cytoplasmic membrane and the triggering of genome injection.     

Single amino acid substitutions at conserved residues of human interferon-alpha can effect antiviral specific activity

Site-specific in vitro mutagenesis was used to direct various amino acid substitutions at conserved positions within the sequence of human interferon-alpha 1 (IFN-alpha 1). The antiviral specific activity of IFN-alpha 1, expressed in M13 as a fusion protein [IFN-alpha 1 (phi WT)], could be altered by single amino acid substitutions. The substitution of glycine for tyrosine at position 123 results in a loss of more than 99% of the antiviral specific activity on human cells, but causes no significant change in the antiviral specific activity on primary bovine cells. The tyrosine at position 123 is thus implicated in determining human cell specificity. Based on analysis of IFN-alpha 2, IFN-alpha 1 contains two dulsulphide bridges between cysteine residues 29 and 139 and cysteine residues 1 and 99. IFN-alpha 1 also contains a fifth cysteine residue at position 86. IFN-alpha 1 (phi WT) carrying three serine for cysteine substitutions at positions 1, 86 and 99 retains 23% of the antiviral specific activity of IFN-alpha 1 (phi WT) on human cells. However, the antiviral activity on bovine cells is not significantly affected by this modification. The presence of the disulphide bridge between residues 1 and 99 thus appears to be required for full antiviral activity on human but not bovine cells. A single serine for cysteine substitution at position 29 reduces the antiviral specific activity on both human and bovine cells by some 95%. This data shows that the disulphide bridge between residues 29 and 139 is critical for the antiviral activity of IFN-alpha's.     

Contribution of the C-terminal tail of U1A RBD1 to RNA recognition and protein stability.

The N-terminal RNA binding domain (RBD1) of the human U1A protein interacts specifically with a short RNA hairpin containing the U1 snRNA stem/loop II sequence. Previous RNA binding studies have suggested that the C-terminal tail of RBD1 contributes to RNA recognition in addition to interactions on the beta-sheet surface of the protein. To evaluate the contributions of these C-terminal residues in RBD1 to RNA binding affinity and specificity, as well as to study the thermodynamic stability of RBDs, a number of RBD1 mutants with truncated tails, with single amino acid substitutions, and with both a truncation and an amino acid substitution, have been constructed. The thermodynamic stabilities of these mutants have been measured and compared by GdnHCI unfolding experiments. The RNA binding affinity and specificity of these mutant proteins have been assessed by measuring the binding of each protein to the wild-type RNA hairpin and to selected RNA mutants with nucleotide substitutions in the RNA loop. The results demonstrate first that, although the C-terminal tail of RBD1 makes significant contributions to RNA binding affinity, it is not required for RNA binding, and second, its contributions to binding specificity are mediated only through selected nucleotides in the RNA loop, for in the absence of the tail, the protein continues to use other nucleotides to discriminate among RNAs. In these truncated proteins, the secondary structure intrinsic to the C-terminal tail is absent, yet their affinity and discrimination for RNAs are not lost. Thus, a structured tail is not required for RNA recognition.     

Regions at the carboxyl end of bacteriophage phi 29 protein p6 required for DNA binding and activity in phi 29 DNA replication.

Series of deletions corresponding to the carboxyl end of the phage phi 29 protein p6 have been constructed and their activity in the initiation of phi 29 DNA replication and their capacity to interact with the phi 29 DNA ends have been studied. Determination of the activity of the deletion mutants in phi 29 DNA replication indicated the dispensability of the 14 carboxy-terminal amino acids of the protein. The activity of protein p6 decreased with deletions from 23 to 39 amino acids and was undetectable when 44 amino acids were removed. A similar behaviour was obtained when the interaction of the mutant proteins with the phi 29 DNA ends was analyzed. These results indicate that the stimulation of phi 29 DNA replication by protein p6 requires a specific binding to the phi 29 DNA ends.     

Analysis of the RNA recognition motifs of human neuronal ELAV-like proteins in binding to a cytokine mRNA

Human neuronal Elav-like proteins contain three RNP-type RNA recognition motifs (RRMs). Previous reports demonstrated that a single RRM of the proteins is not sufficient to bind to the uridine-rich stretch in the 3' untranslated region of mRNAs and that the bi-RRM peptide consisting of the first two RRMs is necessary for the binding. The present study was designed to examine the potential contributions of the first two RRMs when binding to a cytokine mRNA. Deletions of the internal or terminal amino acid residues of the first RRM (RRM1) of the HuC/ple21 ELAV-like protein completely abolished RNA binding. However, removal of any region of the second RRM (RRM2) except for the eight amino acid residues, which correspond to the potent fourth beta-sheet structure of RRM2, did not affect RNA binding. Conjugation of the eight amino acid residues to RRM1 enhanced the RNA binding as well as the entire RRM2, indicating that the octapeptide of RRM2 can be compensated for by the binding function of RRM2. The present study also showed that the substitutions of glutamic acid at 42 for aspartic acid and leucine at 44 for phenylalanine in the first potent alpha-helix structure of RRM1, as were seen in another ELAV-like protein Hel-N1, markedly affected the RNA binding.     

In vitro mutagenesis identifies a region within the envelope gene of the human immunodeficiency virus that is critical for infectivity.

Site-specific mutagenesis was used to introduce amino acid substitutions at the asparagine codons of four conserved potential N-linked glycosylation sites within the gp120 envelope protein of human immunodeficiency virus (HIV). One of these alterations resulted in the production of noninfectious virus particles. The amino acid substitution did not interfere with the synthesis, processing, and stability of the env gene polypeptides gp120 and gp41 or the binding of gp120 to its cellular receptor, the CD4 (T4) molecule. Vaccinia virus recombinants containing wild-type or mutant HIV env genes readily induced syncytia in CD4+ HeLa cells. These results suggest that alterations involving the second conserved domain of the HIV gp120 may interfere with an essential early step in the virus replication cycle other than binding to the CD4 receptor. In long-term cocultures of a T4+ lymphocyte cell line and colon carcinoma cells producing the mutant virus, revertant infectious virions were detected. Molecular characterization of two revertant proviral clones revealed the presence of the original mutation as well as a compensatory amino acid change in another region of HIV gp120.     

Reevaluation of amino acid variability of the human immunodeficiency virus type 1 gp120 envelope glycoprotein and prediction of new discontinuous epitopes

To elucidate the evolutionary mechanisms of the human immunodeficiency virus type 1 gp120 envelope glycoprotein at the single-site level, the degree of amino acid variation and the numbers of synonymous and nonsynonymous substitutions were examined in 186 nucleotide sequences for gp120 (subtype B). Analyses of amino acid variabilities showed that the level of variability was very different from site to site in both conserved (C1 to C5) and variable (V1 to V5) regions previously assigned. To examine the relative importance of positive and negative selection for each amino acid position, the numbers of synonymous and nonsynonymous substitutions that occurred at each codon position were estimated by taking phylogenetic relationships into account. Among the 414 codon positions examined, we identified 33 positions where nonsynonymous substitutions were significantly predominant. These positions where positive selection may be operating, which we call putative positive selection (PS) sites, were found not only in the variable loops but also in the conserved regions (C1 to C4). In particular, we found seven PS sites at the surface positions of the alpha-helix (positions 335 to 347 in the C3 region) in the opposite face for CD4 binding. Furthermore, two PS sites in the C2 region and four PS sites in the C4 region were detected in the same face of the protein. The PS sites found in the C2, C3, and C4 regions were separated in the amino acid sequence but close together in the three-dimensional structure. This observation suggests the existence of discontinuous epitopes in the protein's surface including this alpha-helix, although the antigenicity of this area has not been reported yet.     

Assembly of phage phi 29 genome with viral protein p6 into a compact complex.

The formation of a multimeric nucleoprotein complex by the phage phi 29 dsDNA binding protein p6 at the phi 29 DNA replication origins, leads to activation of viral DNA replication. In the present study, we have analysed protein p6-DNA complexes formed in vitro along the 19.3 kb phi 29 genome by electron microscopy and micrococcal nuclease digestion, and estimated binding parameters. Under conditions that greatly favour protein-DNA interaction, the saturated phi 29 DNA-protein p6 complex appears as a rigid, rod-like, homogeneous structure. Complex formation was analysed also by a psoralen crosslinking procedure that did not disrupt complexes. The whole phi 29 genome appears, under saturating conditions, as an irregularly spaced array of complexes approximately 200-300 bp long; however, the size of these complexes varies from approximately 2 kb to 130 bp. The minimal size of the complexes, confirmed by micrococcal nuclease digestion, probably reflects a structural requirement for stability. The values obtained for the affinity constant (K(eff) approximately 10(5) M-1) and the cooperativity parameter (omega approximately 100) indicate that the complex is highly dynamic. These results, together with the high abundance of protein p6 in infected cells, lead us to propose that protein p6-DNA complexes could have, at least at some stages, during infection, a structural role in the organization of the phi 29 genome into a nucleoid-type, compact nucleoprotein complex.     

Nucleic acid binding and unfolding properties of ribosomal protein S1 and the derivatives S1-F1 and m1-S1.

The nucleic acid binding and unwinding properties of wild-type Escherichia coli ribosomal protein S1 have been compared to those of a mutant form and a large trypsin-resistant fragment, both reported recently [J. Mol. Biol. 127, 41-45 (1979) and J. Biol. Chem. 254, 4309-4312 (1979). The mutant (m1-S1) contains 77% and the fragment (S1-F1) 66% of the polypeptide chain length (approximately 600 amino acid residues) of protein S1. The mutant is active in protein synthesis in vitro; the fragment, although retaining one or more of the functional domains of S1, is inactive in protein synthesis. We find that m1-S1 is is almost as effective as S1 in binding to poly(rU), phage MS2 RNA and simian virus 40 (SV40) DNA, and in unfolding poly(rU) and the helical structures present in MS2 RNA and phi X174 viral DNA. S1-F1, however, binds to poly(rU) and denatured SV40 DNA, but not to MS2 RNA. It unfolds neither poly(rU), nor the residual secondary structure of MS2 RNA or phi X174 viral DNA. Thus, there appears to be a correlation between the loss in ability of S1 to unwind RNA and the loss in its ability to function in protein synthesis.     

Comparison of the A-T rich regions and the Bacillus subtilis RNA polymerase binding sites in phage phi 29 DNA.

By using a modification of the BAC spreading method for mounting the DNA for electron microscopy, partial denaturation maps of protein-free phi 29 DNA and of phi 29 DNA containing protein p3 were obtained. In phi 29 p3-DNA1 the protein does not seem to influence the melting of the ends of the molecules. The comparison of the partial denaturation map and the B. subtilis RNA polymerase binding sites indicates that five of the seven early promoters (A1, A2, A3, B2 and C2) are located in A-T rich DNA regions whereas the other two early promoters (B1 and C1) are located in less A-T rich sites.     

Mapping the anatomy of the immunodominant domain of the human immunodeficiency virus gp41 transmembrane protein: peptide conformation analysis using monoclonal antibodies and proton nuclear magnetic resonance spectroscopy.

Thirty-six monoclonal antibodies from mice and three from rats were raised against a peptide corresponding to the immunodominant domain of the transmembrane gp41 protein of human immunodeficiency virus (HIV) type 1 (LGLWGCSGKLIC; amino acid residues 598 to 609). Of these, three monoclonal antibodies from the mice and one from a rat also reacted with the corresponding peptide derived from the HIV type 2 transmembrane gp41 protein (amino acid residues 593 to 603; NSWGCAFRQVC). Immunochemical studies using a variety of synthetic peptides indicated that the cross-reactivity was due to antibody binding to CSGKLIC of HIV type 1 or CAFRQVC of HIV type 2. Single amino acid substitutions for a cysteine at either the amino or the carboxy end of the peptide interrupted antibody binding, indicating that the site recognized was the Cys-XXXXX-Cys loop. Similar results were obtained when the 11-mer HIV type 2 gp41 peptide (amino acids 593 to 603) was inoculated into mice to raise monoclonal antibodies. In this instance, of 30 monoclonal antibodies developed, 4 reacted with both HIV type 1 and HIV type 2 peptides. The conformation of a seven-residue peptide, CSGKLIC, corresponding to residues 603 to 609 of the gp41 immunodominant epitope of HIV-1 was investigated by proton nuclear magnetic resonance spectroscopy. The immunologically active form of CSGKLIC contains an intramolecular disulfide bond and maintains a preference for a folded conformation, apparently including a type I reverse turn about the residues SGKL. No such preference is observed for the reduced form of the peptide, which contains two thiol groups. The presence of the disulfide bond is thus integral to the formation of the structure of the loop in solution. In agreement with this finding, elimination of the possibility of loop formation by substitution of S for C at the amino or carboxy termini of the 7-mer is accompanied by the failure of antibody binding to this peptide.     

Modular assembly of chimeric phi29 packaging RNAs that support DNA packaging

The bacteriophage phi29 DNA packaging motor is a protein/RNA complex that can produce strong force to condense the linear-double-stranded DNA genome into a pre-formed protein capsid. The RNA component, called the packaging RNA (pRNA), utilizes magnesium-dependent inter-molecular base-pairing interactions to form ring-shaped complexes. The pRNA is a class of non-coding RNA, interacting with phi29 motor proteins to enable DNA packaging. Here, we report a two-piece chimeric pRNA construct that is fully competent in interacting with partner pRNA to form ring-shaped complexes, in packaging DNA via the motor, and in assembling infectious phi29 virions in vitro. This is the first example of a fully functional pRNA assembled using two non-covalently interacting fragments. The results support the notion of modular pRNA architecture in the phi29 packaging motor.