The Bacteriophage l DNA packaging enzyme:
Identification of four structural domains of the gpNu1
subunit using limited proteolysis

PAMELA ARAYA,¹ MARIO ROSEMBLATT,^1,2 PABLO VALENZUELA,^1,2 and HELIOS MURIALDO. ¹

¹Fundación Ciencia para la Vida and ²Millennium Institute for Fundamental and Applied Biology. Avenida Marathon 1943, Santiago, Chile.

Author to whom correspondence should be sent: Helios Murialdo. Fundación Ciencia para la Vida, Av. Marathon 1943, Santiago, Chile. Phone : (56-2) 239-8969. Fax : (56-2) 237-2259. e-mail: hml@bionova.cl

Received: May 10, 2001. Accepted: July 31, 2001

ABSTRACT

Lambda DNA terminase, the enzyme that cleaves virion-length chromosomes from multigenomic concatemers and packages them into the bacteriophage head, is composed of two subunits, gpNu1 and gpA. Direct determination of the structure of gpNu1, the smaller subunit, has not been possible because of its insolubility in aqueous solutions. Therefore, to identify smaller and potentially water-soluble domains of gpNu1, we analyzed the nature of the products obtained by limited digestion of the protein with several proteases.
The gpNu1 subunit was obtained from E.coli cells transfected with the plasmid pH6-Nu1 that overproduces the protein. Incubation of gpNu1 solubized in 2.5 M guanidinium chloride with chymotrypsin resulted in the formation of at least eight discrete protein bands, while treatment with endoproteinase glu-C and bromelain yielded three and one major bands, respectively. The peptides generated by digestion with the various proteases were separated by two-dimensional gel electrophoresis and transferred to Immobilon membranes. Amino acid sequencing of the peptides allowed for the precise assignment of their N-terminal amino acid, while their estimated molecular weights permitted the identification of their C-terminal ends.
The results reveal that in the presence of 2.5 M guanidinium chloride, gpNu1 is partially folded in at least four distinct structural domains that correspond to functional domains as determined by previously reported genetic experiments. This information is key to design new plasmids to overproduce these domains for further structural analysis. (Biol Res 2001; 34 3-4: 207-216)

Key Terms: DNA packaging, terminase, gpNu1 structural domains, limited proteolysis.

INTRODUCTION

The bacteriophage l capsid contains a double stranded DNA molecule of 48,500 bp with complementary single-stranded ends of 12 nucleotides each. This DNA molecule, the l genome, is derived by endonuclease digestion of linear concatemers, which contain many copies of the genome in a tandem configuration. These concatemers are produced by the rolling-circle mechanism of DNA replication. The cleavage of concatemers to form virion-size DNA is called DNA maturation and is performed by the l DNA terminase. Maturation involves the recognition by this enzyme of a specific sequence in the DNA (called cos) and the introduction of two nicks, twelve nucleotides apart, each one on opposite strands. The enzyme then proceeds to bind the empty capsid (called procapsid) and to introduce the DNA in a process called packaging or encapsidation. A round of DNA packaging finishes when the enzyme encounters a second cos site along the DNA. The enzyme then cleaves the DNA and, still attached to the unpacked end, becomes ready to bind to another procapsid for a second round of packaging (for a review see Murialdo, 1991).

The l DNA terminase is composed of two subunits coded for by the viral genes Nu1 and A. The product of Nu1, gpNu1, is a protein of 181 amino acids and that of A, gpA, is made of 641 amino acids (Sanger et al., 1983). The holoenzyme purifies as a trimeric complex having the composition gpA₁-gpNu1₂ (Gold and Becker, 1983), but the composition of the packaging complex is not yet clear.

It is known that the gpNu1 subunit binds cooperatively to three repeated DNA elements called R1, R2 and R3, found in a region of cos called cosB. The assembly of gpNu1 and cosB is stimulated by a host factor called IHF (Integration Host Factor). This factor, which binds to a site called I1, also located within cosB introduces a sharp bend in the DNA (Shinder and Gold, 1988; 1989; Kosturko et al., 1989; Xin and Feiss 1988, 1993). The formation of this nucleoprotein complex increases the specificity of the enzyme and strongly enhances the binding of a dimer of gpA to cosN, a region adjacent to cosB (Cue and Feiss, 1992; 1993; Miwa and Matsubara, 1982; 1983). The larger subunit, gpA, has the nucleolytic activity that introduces the two nicks in the DNA (Davidson et al., 1991; Hwang and Feiss, 1996; Rubinchick et al., 1994a). Also, there is evidence that this subunit has the translocase activity supposed to be responsible for the introduction of the DNA into the procapsid (Rubinchik et al., 1994b).

Genetics experiments have allowed the localization of some of the activities of the terminase subunits to functional domains along the polypeptide chain. In particular, the domain that determines the DNA-binding specificity of gpNu1 has been located within the amino-terminal residues, whereas the domain responsible for the specificity of the subunits assembly with gpA is located in its carboxy-terminus (Frackman et al., 1985). Within the DNA-binding specificity domain, from amino acid residues 5 to 24, there is a helix-turn-helix motif (Becker, cited in Feiss, 1986; Kypr and Mrázek, 1986). As predicted earlier by Becker and Murialdo (1990) it has been recently demonstrated that this domain is responsible for the specificity of gpNu1 binding to cosB (de Beer et al. personal communication). GpNu1 also has a low-affinity ATPase stimulated by non-specific DNA (Babbar and Gold, 1998; Hwang et al., 1996; Tomka and Catalano, 1993; Rubinchik et al., 1994a), and its sequence, from residues 29 to 48, matches the canonical ATPase phosphate-binding loop common to many ATPases and GTPases (Becker and Gold, 1988). Mutations in Lys35 in this putative loop alters the strength and specificity of DNA binding by the holoenzyme and cause a severe post-cleavage defect in DNA packaging (Hwang and Feiss, 1997; 1999).

Central to an understanding of the mechanism of the multiple terminase activities is the knowledge of its domain structure. Numerous attempts by several laboratories to crystallize the holoenzyme have been unsuccessful. Thus, it has become apparent that the study of individual subunits should be attempted first, in particular that of the smaller one. However, it has been shown that gpNu1 is insoluble at high concentrations (Murialdo et al., 1987; Parris et al., 1988; Meyer et al., 1998) due to the presence of a highly hydrophobic region spanning residues 100 to 141 (Yang et al., 1999a; 1999b). This property has precluded the direct determination of its structure by crystallography or NMR. An alternative procedure, would be the determination of the tertiary structure of smaller individual domains. This approach requires the isolation of the structural domains of the subunit. Limited digestion of proteins with proteolytic enzymes has proven to be a successful procedure to identify and to eventually isolate structural domains (see for instance Bochkarev et al., 1997; Gomes and Wold, 1995; Gomes et al., 1996; Pfeutzner et al., 1997). In this paper we have used this procedure to show that in the presence of 2.5 M guanidinium chloride gpNu1 is partially folded in at least four distinct structural domains that correspond to functional domains previously defined by genetic experiments (Frackman et al. 1985).

MATERIAL AND METHODS

Growth media and buffers

Luria broth (LB) contained 10 g of Bacto Tryptone (Difco), 10 g of NaCl, and 5 g of yeast extract per liter of distilled water.

The pH was adjusted to 7.5 with 10 M NaOH. LB-Amp consisted of LB containing 100 µg/ml of ampicillin (Sigma), sterilized by microfiltration and added just prior to inoculation. Solid media for bacterial colony formation was LB containing 2% agar. Sonication buffer (SB) contained 20 mM Tris-HC1 pH 8.0, 100 mM NaC1 and 3 mM MgC1₂. SB-2.5G was SB containing 2.5 M guanidinium chloride and SB-6G was SB containing 6.0 M guanidinium chloride. IPTG for bacterial gene induction was obtained from Sigma.

Gel electrophoresis in one and two dimensions

To separate medium size peptides, electrophoresis was performed through polyacrylamide gels in the presence of sodium dodecyl sulphate (SDS-PAGE) using glycine running buffer as described by Laemmli (1971). Acrylamide concentration in the stacking and separating gels were 4% and 17%, respectively. To separate small size peptides, SDS PAGE using tricine as running buffer was performed according to Shägger and von Jagon (1987). Following these authors nomenclature the stacking gel was 4% T, 3% C, while the separating gel composition was 16,5% T, 3% C. Molecular weight standards were from Life Technologies or a mixture of carbonic anhidrase (29.0 Kda), cytochrome C (12.4 Kda) and apronitin (6.5 Kda) as indicated.

Two-dimensional gel electrophoresis was performed according to O'Farrell (1975) and the ampholites used (Biolabs) ranged from pH 3.0 to 10.0.

Production and partial purification of gpNu1

gpNu1 was isolated from the E. coli strain BL21(DE3)[pH6-Nu1]. The plasmid harbored by this strain contains the Nu1 gene, tagged with a methionine and six histidine codons at the 5' end, under the control of the ß-galactosidase repressor (Hang et al., 1999). The cells were grown at 37º in LB-Amp medium to an OD₆₅₀ of 0.6. Expression was induced by the addition of IPTG to a concentration of 1.2 mM. After two hours of incubation at 37º the cells were pelleted by centrifugation at 4,000g for 10 min at 4ºC. All subsequent steps were carried out at 4ºC. The pellet was resuspended in SB at 1/125 of the culture volume and lysed by several 10-sec pulses of ultrasound until the suspension became translucent. The sonicated suspension was centrifuged at 6,000g for 10 minutes. The gpNu1 protein was found in the pellet of the crude extract as inclusion bodies (Murialdo et al., 1987; Parris et al., 1988; Meyer et al., 1998).

The inclusion bodies were dissolved by incubation in 6 M guanidinium chloride adjusted to pH 8.0 with NaOH on ice for one hour. The soluble material was separated from the particulate material by centrifugation at 10,000g for 10 min. The supernatant, about 4 ml, was loaded into a 5 ml Talon Co⁺⁺ affinity column. The column was first washed with SB-6G adjusted to pH 5.9 followed by a second wash with SB-6G pH 5.5 and finally with SB-6G adjusted to pH 5.0. gpNu1 was then eluted with SB-6G adjusted to pH 4.5. This eluate was dialyzed against SB-2.5G (pH 8.0). This procedure produced gpNu1 more than 90% pure, as determined by by SDS-PAGE analysis and staining with Coomassie brilliant blue.

Limited protease digestion

The proteases used were chymotrypsin, endopeptidase glu-C and bromelain. The digestion reactions were all carried out in SB-2.5G at 37º. The proportion of protease to gpNu1 was 1:250 (w/w) for chymotrypsin and endopeptidase glu-C, and of 1:100 for bromelain. To study the kinetics of proteolysis, aliquots of the reaction mixtures were withdrawn at various times of incubation and the extent of the digestion was analyzed by SDS-PAGE. The apparent molecular weights of the most abundant peptides resulting from the protease digestions were estimated from their electrophoretic mobility. The best appropriate time for preparative digestions was chosen upon inspection of the results of the kinetics experiments.

Amino acid sequencing of the N-terminal ends of the peptides

For determination of the amino-terminal sequence of selected peptides resulting from the proteolytic cleavage of gpNu1, the protease reaction mixtures were subjected to two-dimensional gel electrophoresis and the spots were electro-transferred to Immobilon P membranes (PVDF). After staining, the regions of the membrane containing the spots of interest were cut off and the immobilized peptides were directly subjected to amino acid sequencing in a Applied Biosystems Sequencer.

RESULTS

Kinetics of gpNu1 digestion by the proteases

Several proteases were tested for their ability to digest gpNu1 with the generation of discrete bands in SDS-PAGE. Three that gave reproducible results were selected for further studies. The results of a series of time-course protease digestion of gpNu1 are presented in Figure 1. Analysis of the digestion with chymotrypsin by glycine-SDS-PAGE showed the generation of six distinct peptides labeled Q1 to Q6 (Figure 1a). Tricine-SDS-PAGE of a chymotrypsin digest showed the presence of two additional low-molecular weight peptides labeled Q7 and Q8 in Figure 1b. In this experiment, the concentration of chymotrypsin was increased to a ratio of 1:50 (protease to gpNu1). This allowed for the accumulation of peptides Q6 and Q7 but resulted in faint Q2 and Q3 bands, probably as a result of their being further hydrolyzed to smaller peptides (see below). Incubation of gpNu1 with the endopeptidase glu-C generated two faint and three major discrete bands. The latter ones are labeled G1 to G3 in Figure 1c. The same bands were observed in tricine-SDS-PAGE.

Hydrolysis of gpNu1 catalyzed by bromelain resulted in one major band, labeled B1 and a series of faint discrete fragments (Figure 1d). The same bands were observed in tricine-SDS-PAGE.

Table 1 summarizes the results of the digestion of gpNu1 with the three proteases and the molecular weights of the main peptide products, estimated from their mobilities in SDS-PAGE.

Figure 1. SDS-polyacrylamide gel electrophoresis analyses of the time-course digestion of gpNu1 by several proteases. a) Digestion with chymotrypsin. 53 µg of gpNu1 and 0.2 µg of the protease in SB-2.5G, were incubated at 37º and aliquots of the reaction were withdrawn at the indicated times and fractionated in glycine SDS-PAGE. Lanes 1 to 8 contain samples from 0, 2, 5, 10, 20, 35, 60 and 90 min of incubation respectively. Band identification is indicated at the right. b) Digestion of gpNu1 with chymotrypsin for 10 min. The reaction conditions were the same as before except that the proportion of gpNu1 to protease was 50: 1 and the peptides were fractionated in tricine SDS-PAGE. Band identification is indicated at right. c) Digestion of gpNu1 with endoproteinase glu-C. Digestion conditions were as in a). Lanes 1 to 6 contain samples withdrawn at 0, 2, 5, 10, 20, and 30 min of incubation, respectively, and resolved in glycine SDS-PAGE. The identity of the bands is indicated at left. d) Digestion of gpNu1 with bromelain. Digestion conditions were as in a) except that the ratio of protease to gpNu1 was 1:100. Lanes 1 to 5 contain samples withdrawn after 0, 30, 60, 90 and 120 min of incubation and fractionated in glycine SDS-PAGE. The identity of the bands is indicated at right. Lanes labeled Std show the electrophoretic mobility of the standards with their size indicated in kDa.

Amino acid sequencing of the fragments

An incubation time of 10 min for chymotrypsin and endoproteinase glu-C and of 60 min for bromelain was chosen to perform scaled up reactions to generate enough material for separation of the fragments by 2-D gel electrophoresis and to electrotransfer the spots to Immobilon membranes. Figure 2 shows the spot patterns in the Immobilon membranes and identifies the spots that were isolated for amino acid sequencing of the N-terminal end of the peptides. The results of the sequencing are shown in Figure 3.

TABLE I
Molecular weight and probable composition of the fragments of gpNu1 obtained by proteolysis

DISCUSSION

Analysis of the structural domains of proteins by limited proteolysis gives useful information when the structural domains are such that only a few of its peptidic bonds are spatially available and therefore susceptible to hydrolysis. In such circumstances, a limited number of peptides appear as discrete bands in electrophoresis. We have used this approach to study the domain composition of gpNu1, the smaller subunit of the bacteriophage l DNA packaging enzyme and have identified several discrete domains within this protein. The results obtained by digestion with chymotrypsin were the most informative since the composition of all of the proteolytic products detected by gel electrophoresis could be deduced. This was possible due to the precise assignment of the N-terminal amino acids of some of the peptides by sequencing and by the determination of their molecular weights inferred from their electrophoretic mobility (Table 1). The C-terminal ends of the different peptides were deduced by selecting, from among the theoretical protease-susceptible sites, the one that best fitted the estimated molecular weights. Of the eight chymotrypsin-produced peptides that were observed in SDS-PAGE, the N-terminus of four, Q2, Q3, Q5 and Q6 was directly determined by sequencing (Figure 3A). Based on the estimated molecular weights and the enzyme substrate specificity, Q2 and Q3 should extend from their determined N-termini, Tyr50 and Glu85, respectively, to the C-terminus of gpNu1 (Table 1, Figure 3A). Also, by the same reasoning, peptide Q6 corresponds to a molecule extending from its determined N-terminus, Met-7, to Trp49 (Table 1, Figure 3a). For the selection of position 150 as the C-terminus of Q5 it was assumed that cleavage of the same bond generated Q8. Given this assumption, cleavage of Q3 at the carboxy side of Phe150 would generate peptides of molecular weights closest to those estimated for Q5 and Q8. Any other alternative, based on the specificity of chymotrypsin, would improve the fitting of the electrophoretic mobility of one of the bands but it would worsen that of others. However, it is possible that Q5 would extend from Glu85 to Met159 (calculated MW = 8.6 Kda) and that Q8 would be generated by different cleavages. However, this scenario would generate a number of bands in SDS-PAGE that exceeds the eight ones that were clearly observed.

Peptides Q1, Q4, Q7 and Q8 were not sequenced, however their composition was deduced as follows. Because of its molecular weight, Q1 could not be generated by cleavages at positions 49-50 or 84-85. Thus, it is likely that this peptide extends between residues -7 to 150 or from residues -7 to 159. The composition of Q4 was inferred from its estimated molecular weight and the fact that the observed cutting at position 84-85 (fragments Q3 and Q5) was highly efficient (Figure 1a). The composition of Q7 was chosen based on the best molecular weight fit among the fragments that should be produced by the efficient cleavages observed at positions 49-50 and 84-85. The likely origin of Q8 has been discussed above.

Figure 2. Bidimensional separation of protease-generated fragments of gpNu1. Cleavage of gpNu1 with the proteases was performed as indicated in Figure 1 and the products subjected to 2D-gel electrophoresis as described in Materials and Methods. The molecular weight standards are positioned at one edge of the figure. The spots that were used for N-terminal amino acid sequencing are indicated by an arrow and labeled accordingly. a) digestion with chymotrypsin. A spot corresponding to peptide Q4, which was not sequenced, is also indicated, b) digestion with endopeptidase glu-C and c) digestion with bromelain. The times of digestion were 10 min for a) and b) and of 60 min for c).

The fragments G1 and B1 were obtained by digestion with endoproteinase glu-C and bromelain, respectively. The N-termini of these fragments were obtained by direct sequencing while their C-termini were inferred as described above for fragments Q3 and Q2, respectively. The other peptides produced by endoproteinase glu-C, G2 and G3 (Fig. 1c) did not resolve into distinct spots (Fig. 2b) and were not sequenced (Table I).

It is worth mentioning that, of all the potential sites that could be cleaved by the proteases used here, only a small number of these sites were cleaved. As expected, all of the peptides obtained were in agreement with the specificity of the respective enzyme. For instance, according to the known specificity of chymotrypsin, MetHis₆-tagged gpNu1 has 39 susceptible sites, but only 3 sites were cleaved efficiently enough to generate discrete bands in SDS-PAGE. These sites involved the bonds corresponding to amino acid residues 49-50, 84-85 and, presumably 150-151 (Figure 3 and Table 1).

The kinetics of chymotrypsin digestion showed that the most preferred cleavage occurred at position 49-50 resulting in the production of Q2 and Q6. Cleavages at position 150-151, generating fragments Q1 and Q8 and at position 84-85, which generates Q3 and Q4, were slightly less efficient than that of at position 49-50. It is also apparent that production of the peptide formed by two cuts is delayed with respect to that of those generated by a single cut (Figure 1a).

Figure 3. N-terminal amino acid sequence of gpNu1 fragments and structural domains map. A) The N-terminal sequences of the fragments obtained with the different proteases are shown aligned with the sequence of gpNu1. B) Schematic diagram of the putative structural domains of gpNu1 as determined in the present study. The four ellipses (D1 to D4) represent structural Domains 1 to 4. The protease-susceptible peptide bonds which appear as part of the connections between domains are indicated. The sequenced peptides, in the form of a line, are shown at the bottom in relation to the structural domain map. Indicated in the lines representing the peptides are their molecular weights and the positions, within gpNu1, of their N-and C-terminal amino acid residues.

Digestion of gpNu1 with endoproteinase glu-C produced numerous spots in 2-D gels that did not resolve clear from each other (Figure 2b). The analysis with this enzyme was not as extensive as that of chymotrypsin but, again, of 32 potential cleavage sites, only very few appear to be exposed to enzyme attack. Of the three fragments produced only G1 was sequenced. The analyzed fragment, G1 resulted from cleavage at site 85-86, a region also exposed to chymotrypsin attack, corroborating the solvent-exposed nature of the protein around this site.

Bromelain has very low if any specificity. This can be appreciated in Figure 1d which shows extensive protein degradation with the formation of a few, rather faint and transient bands. This allowed the sequencing of only the major band, B1. The site of the cleavage that produced B1 was 49-50, a site also susceptible to chymotrypsin attack.

It is thus clear that, under the reaction conditions utilized, there were only three narrow, well defined protein regions, among many potential ones in the amino acid sequence of gpNu1, that were exposed and, therefore, preferentially hydrolyzed by the various proteolytic enzymes.

Although the presence of 2.5 M guanidinium chloride could introduce some artifacts, such as anomalous folding, our results suggest that in the presence of this denaturant, gpNu1 is not totally unfolded, retaining instead four structural domains which correlate to some of the functional domains previously determined by genetic methods (Frackman et al. 1985). Our data indicate that these domains have the following arrangements: domain 1, spanning the protein region from residues 1 to 49; domain 2, covering residues 50 to 84-85; domain 3, a region from residues 85-86 to 150 (or 85-86 to 159) and; domain 4, which covers residues 151 (or 160) to 181 (Figure 3). A schematic diagram, summarizing these results is shown in Figure 3B.

Previous genetic experiments have localized the DNA-binding specificity of gpNu1 within its first 91 amino-terminal residues (Frankman et al., 1985). A helix-turn-helix motif is localized from amino residues 5 to 24 (Becker, cited in Feiss, 1986; Kypr and Mrázek, 1986) which, as predicted by Becker and Murialdo (1990), has been recently shown to be responsible for gpNu1 binding specificity to the cosB region of the DNA (de Beer et al., personal communication). In addition, biochemical experiments have shown that gpNu1DK, a truncated derivative of gpNu1 spanning residues 1 to 100, binds cos DNA (Yang et al., 1999b). Since our data suggests the presence of a distinct domain extending between Met1 and Trp49 it may be inferred that the DNA-binding activity of gpNu1 is contained within domain 1.

It has been found that while gpNu1DK is soluble, another derivative of gpNu1, gpNu1DP141 spanning residues 1 to 141 has a strong tendency to aggregate (Yang et al., 1999a,b). gpNu1DP141 contains, in addition to the DNA-binding domain, the highly hydrophobic region of gpNu1 which is thought to mediate self-assembly. Moreover, thermal denaturation studies done by CD spectroscopy suggest the presence of two independently folded domains in this truncated protein (Yang et al., 1999a). Our results suggest, in turn, that the two independently folded domains detected by CD spectroscopy may correspond to the proteolysis-resistant domains 1 and 2 (or 1, 2 and part of 3) identified in the present study.

Another gpNu1 activity, the specific binding of gpNu1 to gpA has been localized to a domain within the carboxy-terminus of gpNu1 (Frackman et al., (1985). Again, since we have identified a discrete domain extending from residues 150 (or 159) to the C-terminus of the protein (residue 181), it is tempting to assume that this activity would be located within our domain 4.

In summary, it can tentatively be asserted that the structural domains, as determined with the present methodology, would correspond to the functional domains as follows: Domain 1, DNA-binding domain containing the helix-turn-helix motif; Domains 2 and 3, one or both of them would be involved in gpNu1-gpNu1 interaction during formation of the packaging complex; Domain 4, gpA binding. Our results offer a rational approach for the design and construction of plasmids containing truncated genes for the overproduction and purification of large amounts of gpNu1 domains, the structure of which, could be determined by crystallography or NMR.

ACKNOWLEDGEMENTS

We thank Scott Chamberlain for protein sequencing. This research was possible due to the financial support of FONDECYT grant 1980209 to HM, and the Pablo Valenzuela and Bernardita Mendez Foundation.

REFERENCES

BABBAR BK GOLD M (1998) ATP-reactive sites in bateriophage l packaging protein terminase lie in the N-termini of its subunits, gpA and gpNu1. Virology 246: 251-264

BECKER A, GOLD, M (1988) Prediction of an ATP reactive center in the small subunit gpNu1 of the phage lambda terminase enzyme. J Mol Biol 199:219-222

BECKER A, MURIALDO H (1990) Bacteriophage l DNA: The beginning of the end. J. Bacteriol 172 : 2819-2824

BOCHKAREV A, PFUETZNER RA, EDWARDS AM, FRAPPIER L (1997) Structure of the single-stranded-DNA-binding domain of the replication protein A bound to DNA. Nature 385: 176-181

CUE D, FEISS M (1992) Genetic analysis of mutations affecting terminase, the bacteriophage l DNA packaging enzyme, that suppress mutations in cosB, the terminase binding site. J Mol Biol 228: 72-87

CUE D, FEISS M (1993) The role of cosB, the binding site for terminase, the DNA packaging enzyme of bacteriophage l, in the nicking reaction. J Mol Biol 234: 594-609

DAVIDSON A, YAU P, MURIALDO H, GOLD M (1991) Isolation and characterization of mutations in the bacteriophage l terminase genes. J Bacteriol 173: 5086-5096.

FEISS M (1986) Terminase and the recognition, cutting and packaging of l chromosomes. Trends Genet. 2: 100-104.

FRACKMAN S, SIEGELE DA, FEISS M (1985) The terminase of bacteriophage l. Functional domains for cos binding and multimer assembly. J Mol Biol 183: 225-238.

GOLD M, BECKER A (1983) The bacteriophage l terminase. J Biol Chem 258: 14619-14625

GOMES XV, WOLD MS (1995) Structural analysis of human replication protein A. Mapping functional domains of the 70-Kda subunit. J Biol Chem. 270: 4534-4543

GOMES XV, HENRICKSON LA, WOLD MS (1996) Proteolytic mapping of human replication protein A; evidence for multiple structural domains and a conformational change upon interaction with single-stranded DNA. Biochemistry 35: 5586-5595

HANG Q, WOODS L, FEISS M, CATALANO CE (1999) Cloning, expression, and biochemical characterization of hexahistidine-tagged terminase proteins. J Biol Chem 274: 15305-15314

HWANG Y, FEISS M (1996) Mutations affecting the high affinity ATPase center of gpA, the large subunit of bacteriophage l terminase, inactivate the endonuclease activity of terminase. J Mol Biol 261: 524-535

HWANG Y, FEISS M (1997) Mutations affecting Lysine-35 of gpNu1, the small subunit of bacteriophage l terminase, alter the strength and specificity of holoterminase interactions with DNA. Virology 231: 218-230

HWANG Y, FEISS M (1999) A mutation correcting the DNA interaction defects of a mutant phage l terminase, gpNu1 K35A terminase. Virology 265: 196-205

HWANG Y, CATALANO CE, FEISS M (1996) Kinetic and mutational dissection of the two ATPase activities of terminase, the DNA packaging enzyme of bacteriophage l.Biochemistry 35: 2796-2803

KOSTURKO LD, DAUB E, MURIALDO H. (1989) The interaction of the E. coli integration host factor and lambda cos DNA: Multiple complex formation and protein-induced bending. Nucl Acids Res 17: 317-334

KYPR J, MRÁZEK J (1986) Lambda phage protein Nu1 contains the conserved DNA binding fold of repressors. J Mol Biol 191: 139-140

MEYER JD, HANAGAN A, MANNING MC, CATALANO CE. (1998) The phage lambda terminase enzyme: 1. Reconstitution of the holoenzyme from the individual subunits enhances the thermal stability of the small subunit. Int J Biol Macromol 23: 27-36.

MIWA T, MATSUBARA K (1982) Identification of sequences necessary for packaging DNA into lambda phage heads. Gene 20: 267-279

MIWA T, MATSUBARA K (1983) Lambda phage DNA sequences affecting the packaging process. Gene 24: 199-206

MURIALDO H, DAVIDSON A, CHOW S, GOLD M (1987) The control of l DNA terminase synthesis. Nucl Acids Res 15 : 119-140

MURIALDO H (1991) Bacteriophage lambda DNA maturation and packaging. Ann Rev Biochem 60: 125-153

O'FARRELL P H (1975) High resolution two-dimensional electrophoresis of proteins. J Biol Chem 250: 4007-4021

PARRIS W, DAVIDSON A, KEELER JR CL, GOLD M (1988) The Nu1 subunit of bacteriophage l terminase. J Biol Chem 263: 8413-8419.

PFEUTZNER RA, BOCHKAREV A, FRAPPIER L, EDWARDS AM (1997) Replication Protein A. Characterization and crystallization of the DNA binding domain. J Biol Chem 272 430-434.

RUBINCHIK S, PARRIS W, GOLD M (1994a) The in vitro endonuclease activity of gene product A, the large subunit of the bacteriophage l terminase, and its relationship to the endonuclease activity of the holoenzyme. J Biol Chem 269: 13575-13585

RUBINCHICK S, PARRIS W, GOLD M (1994b) The in vitro translocase activity of terminase and its subunits. Kinetic and biochemical analysis. J Biol Chem 270: 20059-20066

SANGER F, COULSON AR, HONG GF, HILL DF, PETERSEN GB (1982) Nucleotide sequence of bacteriophage l DNA. J Mol Biol 162: 729-773

SCHÄGGER H, von JAGOW G (1987) Tricine-sodium dodecyl sulfate-polyacrylamide electrophoresis for the separation of proteins in the range of 1 to 100 kDa. Anal Biochem 166: 368-379.

SHINDER G, GOLD M (1988) The Nu1 subunit of bacteriophage lambda terminase binds to specific sites in cos DNA. J Virology 62: 387-392

SHINDER G, GOLD M (1989) Integration host factor (IHF) stimulates binding of the gpNu1 subunit of l terminase to cos DNA. Nucl Acids Res 17: 2005-2022

TOMKA MA, CATALANO CE (1993) Kinetic characterization of the ATPase activity of the DNA packaging enzyme from bacteriophage l. Biochemistry 32: 11992-11997

XIN W, FEISS M (1988) The interaction of Escherichia coli integration host factor with the cohesive end sites of phages l and 21. Nuc Acids Res 16: 2015-2030

XIN W, FEISS M (1993) Function of IHF in l DNA packaging I. Identification of the strong binding site for integration host factor and the locus for intrinsic bending in cosB. J Mol Biol 230: 492 _ 504

YANG Q, BRETON N, MANNING M C, CATALANO C E (1999a) Domain structure of gpNu1, a phage l DNA packaging protein. Biochem 38; 14238-14247

YANG Q, DE BEER, T, WOODS L,MEYER J D, MANNING M C, OVERDUIN M, CATALANO C E (1999b) Cloning, expression and characterization of a DNA binding domain of gpNu1, a phage l DNA packaging protein. Biochemistry 38: 465-477.