DNA Packaging in Bacteriophage T4

Rao VB, Black LW.


Double-stranded (ds) DNA packaging in phage T4 and other icosahedral viruses is a fascinating biological problem. During packaging, a complex, metabolically active, concatemeric DNA is translocated into an empty prohead in an ATP-driven process and condensed as a highly ordered structure of near crystalline density.1-3 dsDNA packaging serves as an excellent model system to understand fundamental biological mechanisms such as the reversible condensation and decondensation of DNA, DNA movement along protein complexes, and transduction of ATP hydrolysis energy into mechanical motion of DNA.

The phage T4 head assembly pathway produces a complex prohead consisting of six essential proteins and at least seven nonessential proteins.4 The T4 prohead maturation protease degrades the scaffolding core into peptides and cleaves off the N-termini of the major capsid protein (gp23) and the vertex protein (gp24) to generate an empty mature prohead that is competent for DNA packaging. In parallel, the T4 DNA replication pathways generate a highly branched “endless” concatemeric DNA, which is associated with a myriad of protein complexes involved in replication, transcription, recombination and repair. A terminase complex of gp16 (18 kDa) and gp17 (70 kDa) links these two pathways by recognizing the viral concatemer, making an endonucleolytic cut, and joining it to the prohead through specific interactions with the dodecameric portal vertex constituted by gp20 (Table 1). Consequently, a DNA packaging machine is assembled, with the terminase as one of the key components (fig. 1). This machine translocates an intact, unit-length, linear dsDNA genome into the capsid to form a highly ordered condensed structure. Terminase apparently also makes the second cut terminating DNA packaging, dissociates from the packaged structure, and repeats the DNA linkage to another prohead in a processive fashion.

Table 1. Major T4 packaging and prohead proteins.

Table 1

Major T4 packaging and prohead proteins.

Figure 1. Schematic representation of the putative DNA packaging machine.

Figure 1

Schematic representation of the putative DNA packaging machine. The terminase proteins gp16 and gp17 cut the DNA concatemer and attach it to the empty prohead through interactions with the gp20 portal vertex. The packaging machine drives translocation (more...)

The above represents an overall pathway for DNA packaging in T4, which, in many respects, is a common DNA packaging pathway among a number of well-characterized dsDNA phages. Common features among the dsDNA packaging phages include the dodecameric ring structure of the portal, a terminase complex formed by multiple copies of one small and one large subunit, the enzymatic functions associated with the terminase proteins, and the headful nature of the packaging process.2 Recent evidence suggests that the ATPase motifs in terminase are particularly well conserved among the phage packaging proteins,5 indeed, phage genomics reveals that these are among the signature proteins defining a phage “quasi-species”.6 The conservation appears to extend to the putative terminase from herpes viruses, which shows a particularly close resemblance to bacteriophage T4.5,7

Thus, it is likely that the dsDNA bacteriophages and herpes viruses package DNA by a common mechanism. Phage T4, with its large genome (171 kb), complex genetic make-up and rich details of head morphogenesis, provides an excellent, though challenging, model system to elucidate the molecular details of the DNA packaging machine. In this review, we focus on the components of the T4 DNA packaging machine, in particular, the terminase proteins. At the end, we discuss the emerging details of the T4 packaging mechanism and provide evidence for a DNA translocating ATPase in phage T4.

Components of the Phage T4 DNA Packaging Machine

Gp16: The Small Terminase Protein

Gp16 is an 18 kDa protein.8,9 It is the small subunit of the terminase holoenzyme. Gp16 is dispensable for in vitro DNA packaging although packaging efficiencies are low in its absence. It is essential in vivo since 16am mutations are lethal, although microscopy reveals that, in double am mutants of gene 16, packaging begins late after infection and proceeds slowly and incompletely.10 Gp16 appears to play an important role in DNA recognition and in modulating the functions of the terminase/packaging complex. It consists of the following binding sites and associated activities.

Oligomerization Site

Gp16 forms stable oligomeric rings and double rings with a diameter of ~8 nm and a central channel of ~2 nm11 (fig. 2). Each ring apparently consists of about 8 gp16 monomers. Since expression vector synthesis of gp16 with or without a his-tag, and refolding of gp16 from 6M urea lead to the ring and double-ring structures and little, if any, monomeric gp16, these structures are strongly and preferentially formed.11,12 They are also observed to be active for in vitro DNA packaging. Sequence analysis identifies a conserved stretch of hydrophobic amino acid sequence in the center of the protein, which may be responsible for the protein-protein interactions that lead to oligomerization (fig. 3).5,11 Consistent with this hypothesis, truncation experiments show that the N-terminal 55 residues do not oligomerize whereas the COOH terminal 121 residues do oligomerize (N. Malys, unpublished). An analogous site was recently identified in the phage λ small terminase subunit gpNuI.13 One model for the ring and double ring structures consistent with a single multimerization interface is that the structures are helical lock washers and double washers, where the double washers unstack. Gp16 rings may play a role in the formation of higher order packaging initiation complexes consisting of the holo-terminase and DNA bound to the portal ring (fig. 1).

Figure 2. STEM micrographs of gp16 oligomers.

Figure 2

STEM micrographs of gp16 oligomers. A) Bright field micrograph of gp16 stained with methylamine vanadate. Single- and double-headed arrows indicate the single and double rings, width of the ring monomer ~8 nm. B) Same as (A), except at higher (more...)

Figure 3. Schematic representation of putative functional sites in gp16 and gp17.

Figure 3

Schematic representation of putative functional sites in gp16 and gp17. Top) The functional motifs in gp16 and gp17 are labeled as shown. Note that an out-of-frame 15 bp sequence codes for 5 amino acids at the amino-terminal-end of gp17 as well as for (more...)

ATP Binding Site

There are two forms of gp16 synthesized in vitro and in vivo, most likely because the out-of-frame partial overlap region between genes 16 and 17 induces translational frameshifting at the ribosome binding site for gene 17, which is located within the 3'-end of gene 16 (fig. 3). The short form gp16 of 155 residues is truncated at a C-terminal R residue, 9 amino acids from the end of the full-length 164 residue protein.11 The short form predominates in vivo. In fact, expression of the short form gp16 alone is sufficient to complement gene 16 am mutants (H. Lin, unpublished observations). Gp16 is shown to have a weak ATP binding activity with the longer form gp16 showing greater ATP binding.11 A Walker-B type ATP binding motif was predicted in the central region of gp16,14 which is analogous to the predicted ATP reactive sites within the small terminase subunits of λ and SPP1. However, neither of the potential D residues appear likely to be the metal-chelating D of a Walker-B motif. In addition, the mostly α-helical secondary structure prediction for gp16 suggests that the protein does not contain the classic nucleotide-binding fold, which consists of β-strands at its core. The gp16 protein does not display an ATPase activity,11,12 among terminase small subunits, only the λ Nu1 protein has been reported to display an ATPase activity.15 The observed weak ATP-binding activity of gp16 is likely due to an anomalous ATP-binding site whose biological function, if any, is uncertain.

DNA Binding Site

T4 gp16 oligomer displays little or no DNA binding, but when renatured from urea together with DNA, gp16 is observed to bind to dsDNA but not ssDNA.11 There appears to be preferential binding to DNA containing gene 16. Analogous to the small terminase subunits of phages λ and SPP1, a helix-turn-helix (H-T-H) DNA binding motif is predicted in the N-terminus of gp16. This motif appears to be a variant of the classical H-T-H signature structure determined for many DNA binding proteins that has been characterized structurally as a winged helix-turn-helix in the case of the λ Nu1 protein.16

Stimulation of Gp17 ATPase and Packaging Activities

Gp16 stimulates the gp17-associated ATPase and in vitro DNA packaging activities by at least 50-fold12 (also, see below). Costimulation of ATPase and DNA packaging activities suggested a linkage between these two functions. Although the mechanism is unknown, it was proposed that gp16 interaction with gp17 induces a conformational change in the large subunit, transforming a basal ATPase into a stimulated ATPase with high catalytic capacity. This may be analogous to the stimulation of GTPases by interaction with the GAPs17 and hence may represent a common molecular switch (see below). Experiments are currently underway to test this hypothesis.

Gp17: The Large Terminase Protein

Gp17 is the 70 kDa large subunit of the terminase holoenzyme.8,9 Gp16 and gp17 together form the terminase holoenzyme complex although the stoichiometry of the subunits in the holoenzyme is unknown. Unlike in the case of the small subunit gp16, gp17 alone is sufficient for packaging DNA in vitro.8,12 A number of functional motifs and associated activities have been recognized, which shed light on the roles of gp17 in the DNA packaging pathway.

ATP Binding Site I

The consensus sequence of the Walker-A nucleotide binding motif, (G/A)XXXXGK(T/S), is present in a large number of enzymes capable of nucleotide binding and/or hydrolysis. Two Walker-A motifs have been identified in gp17,9,18 the N-terminus proximal SRQLGKT161-167 (Walker-AI) and a centrally located TAAVEGKS299-306 (Walker-AII). As shown recently, Walker-AI is highly conserved among all the four T4-family (T4D, RB49, KVP40, KVP20) terminase sequences (fig. 3) as well as in numerous other phage terminases and herpes virus terminases.5,19 Extensive combinatorial mutagenesis analyses of this site revealed a striking conservation of its features. No substitutions were tolerated in the highly conserved GKT signature sequence of the T4 gp17 Walker-A, including the conservative substitutions G165A, K166R, and T167A19 (fig. 4, also, see below).

Figure 4. The critical Walker A motif in gp17.

Figure 4

The critical Walker A motif in gp17. Phenotypes of Walker-A mutants. Amino acid substitutions below the native sequence resulted in a null phenotype and the ones above resulted in a functional phenotype. Some of the null phenotypes shown were inferred (more...)

ATPase sites also contain a Walker-B motif, with the consensus sequence ZZZZD (Z represent a hydrophobic amino acid), which is generally located 50-130 residues downstream of the Walker-A1 lysine. The four hydrophobic amino acids of the Walker-B motif form a β-strand that ends with the highly conserved aspartate. This well-characterized aspartate is responsible for chelating Mg2+ of the bound Mg-ATP complex and for orienting the substrate for nucleophilic attack by an activated water molecule. Sequence analysis of the T4 large terminase shows that a potential Walker-B motif, MIYID251-255, is located 94 amino acid residues downstream of the Walker A lysine5 (fig. 3). The critical elements of this Walker-B motif not only appear to be strictly conserved in the four T4-family terminases, but also in HSV-1 UL15 and other terminases.5 Recent mutagenesis and biochemical studies of the gp17-D255 further provided evidence supporting the Walker-B assignment to this sequence (M. Mitchell and V. Rao, unpublished data). In addition, a catalytic carboxylate residue (E256) and a novel helicase motif III that is presumably involved in ATPase coupling have been identified.5 The catalytic carboxylate residue is required to activate a water molecule for an in-line attack on the λ-phosphate of ATP, whereas the motif III in helicases makes a direct contact with the nucleic acid backbone and triggers subsequent conformational changes resulting in ATP hydrolysis.5

ATP-Binding Site II

A second Walker-A P-loop has been proposed in a number of terminases.18 In T4 gp17, the second ATP-binding site is represented by the sequence TAAVEGKS299-306 with its apparent Walker-B motif close to the C-terminus GVSVAKSLYMD468-478.20 Unlike the striking conservation of Walker-AI, the sequence conservation of Walker-AII and its surrounding region is quite poor. Moreover, the gp17s from phages KVP40 and KVP20 substitute isoleucine and valine respectively for the critical lysine of the canonical Walker-A motif (K305 in T4 gp17).5 Such a substitution would severely compromise ATP binding and hydrolysis. Recent mutagenesis data further showed that nonconservative single and double substitutions are tolerated at the residues K305 and S306; in fact, screening of a library consisting of all possible double substitutions showed that >30% of the substitutions, including PG, VG, RN and SL, were tolerated, suggesting that this site is not critical for gp17 function.19

Terminase Cutting Site

One or more H-X2-H type metal-binding motifs with suggested roles in DNA-binding and endonucleolytic activity have been identified in terminases.18 The histidine-rich motif in the C-terminal half of T4 gp17, H382-X2-H385-X16-C402-X8-H411-X2-H414-X15-H430-X5-H436, appeared more complex consisting of three H-X2-H type motifs.19 Site-directed mutagenesis of His436 showed that none of the twelve substitutions tested were tolerated, which is a clear indication that this is a critical residue.21 Biochemical analyses of H436R gp17 mutant further showed that this mutant lost both the DNA packaging and terminase cutting activities. Recent combinatorial mutagenesis experiments identified certain strictly conserved aspartate residues within the “histidine-rich” motif (D401 and D409) that are critical for gp17 function and some of these mutants exhibited a loss of terminase cutting activity, but not the in vitro DNA packaging activity22 (F. Rentas and V. Rao, in preparation). We suggest that the terminase DNA cutting site consists of a cluster of histidine and aspartic acid residues that are involved in metal-coordinated acid-base catalysis of DNA cleavage, as was reported in a number of nucleases.23

Prohead Binding Site

The portal-terminase interactions appear to be mediated by the negatively charged and hydrophobic residues at the carboxy-terminus of the large terminase protein. Analogous to the phages λ and T3, the negatively charged sequence ELQDMSDDYAP578-588 at the carboxy-terminus of gp17 was considered a good candidate for the portal binding site. However, it turned out that this sequence is not critical for function since a recombinant gp17 that is truncated after K577 is fully functional for in vitro DNA packaging (V. Rao, unpublished data). Based on recent genetic data, other negatively charged clusters, LYNDEDIFDD322-331, and IDYADKDD560-567 are likely candidates for the portal binding site.5,24 A peptide from phage T4 portal protein gp20 (residues 281-308), which encompasses the packaging-defective cold sensitive (cs) mutations in g20, was shown to interact with gp17 and block DNA packaging in vitro.24 Second site mutations that suppress cs20 mutations map in g17. These substitutions, I364F (tsR1) and S583N (tsL51), map close to the proposed sites, which are also well conserved among the T4 family terminases, suggesting that additional residues in the flanking sequence also provide specificity to the terminase-portal protein interactions.24

Activities Associated with gp17

DNA Packaging

Packaging efficiencies measured at 10% wild type (>108/ml infected bacteria) can be obtained using purified terminase proteins.12,25 Gp16 enhances gp17-dependent packaging activity by about 100-fold at low concentrations of gp17.8 DNA packaging in vitro normally requires both gp16 and gp17, but gp17 alone may suffice at high concentrations.11,12 Endogenous phage T4 concatemeric DNA that accumulates in packaging defective phage infections is packaged at about 100-fold greater efficiency than externally added mature DNA.12 Phenol extracted (protein-free) mature and concatemeric DNAs are packaged with comparable efficiency and predominantly by recombination into the active concatemeric DNA as judged by formation of phage recombinants in vitro.12,25 The very high competence of the endogenous DNA suggests that packaging occurs hand-in-hand with other DNA metabolic processes such as recombination and repair. Certain structural features of the metabolically active DNA and/or the associated protein complexes may provide sites for recognition and docking of the terminase onto the DNA substrate.25 Thus concatemeric DNAs produced by genes 55 and 33 am defective infection are inefficiently packaged, apparently at least in part because of the absence of gp55 which interacts with the gp17 terminase subunit and may dock it to DNA.25 The gp55 and gp33 are the T4 late transcription activator (σ-factor) and coactivator, respectively, and link replication and late transcription through attachment to the processivity sliding clamp, gp45, loaded onto the DNA.


In concurrence with the DNA packaging results, gp17, but not gp16, exhibits an ATPase activity.12 This activity hydrolyzes the β-λ phosphoanhydride bond of ATP generating ADP and Pi. The gp17 ATPase is highly specific to ATP; dATP is also cleaved, but at a reduced efficiency. However, none of the other NTPs/dNTPs are hydrolyzed to a significant extent. The km for ATP hydrolysis is about 110 μM and the kcat is about 2 ATPs hydrolyzed/gp17 subunit/min. These data, in particular the low kcat, are consistent with the notion that the basal ATPase activity of gp17 is maintained at a low level when it is not coupled to DNA translocation.12

Gp16-Stimulated ATPase

As mentioned above, the gp17-dependent in vitro DNA packaging activity is enhanced by gp16 by about 100-fold at limiting concentrations of gp17. Gp16 does not possess an ATPase activity, but stimulates the gp17-associated ATPase by about 50-fold.12 The stimulation is specific and shows dependence on the gp16:gp17 ratio, implicating a specific gp16-gp17 interaction to form a holo-terminase complex. The gp16:gp17 ratio for maximal stimulation is estimated to be 6-8:1, although further biochemical studies are underway to determine the stiochiometry of an active holoterminase complex. Analysis of the catalytic parameters of the gp16-stimulated ATPase activity showed that the increase is due to an increase in the catalytic capacity of gp17, but not due to an increase in the affinity towards the ATP substrate;12 the km and kcat for ATP hydrolysis in the presence of gp16 are about 256 μM and 107 ATPs hydrolyzed/gp17 subunit/min, respectively (km: 110 μM and kcat: 2 for gp17 alone).


Incubation of gp17 with ATP results in phosphorylation of gp17. About 10% of the total gp17 is estimated to be in the phosphorylated form, which seemed to be trapped only under conditions of low catalytic rates for ATP hydrolysis such as the absence of gp16, and incubation at low temperatures.12 This and the fact that some of the bound Pi is released in the presence of gp16 suggests that the phosphorylated gp17 is an intermediate of ATP hydrolysis rather than gp17 having an independent protein kinase activity. It is important to determine the phosphorylation site in gp17 and further characterize its role in the catalytic process. A phosphorylated intermediate of gp17 shows mechanistic similarity to some motor proteins, such as the SR ATPase, but not others, such as myosin or F1F0ATPase, which do not display a covalently attached Pi intermediate.26

DNA Cleavage (Terminase)

In phages with specific ends (e.g., λ, T7, T3) and phages with nonsequence specific pac and headful cutting ends (e.g., P22, SPP1, P1), the nonstructural packaging proteins cleave the newly replicated viral concatemeric DNA and generate the termini of the packaged DNA molecule.2,3 The small terminase protein recognizes the viral DNA substrate and directs the large terminase protein to the cleavage site, which will then be cleaved by an endonucleolytic activity associated with the large protein. Phage T4 appears to have a similar functional division, since it packages DNA by a headful mode yielding circularly permuted and terminally redundant ends, i.e., the ends of the packaged genome are widely dispersed over the genome with an estimated 3.3 kb (~2% of the genome) repetition at the ends. As in the pac site phages, T4 packaging is terminated at a random sequence following packaging of ~102% (one headful) of the viral genome, to yield the terminal redundancy of 3.3 kb. Because phenol extracted mature T4 DNA can be efficiently ligated by T4 ligase in vitro,27 either the mature DNA is blunt ended, or the same local nucleotide sequence is cleaved by terminase in packaging, the latter possibility appearing less likely. Overall, the T4 terminase must be a nonspecific but strictly regulated endonuclease, which allows only limited headful length cuts coupled to DNA packaging. The terminal headful cutting may be regulated by portal protein interaction, as is thought to be the case for phages P22 and SPP1.

Evidence thus far suggests that the large terminase protein indeed exhibits a nonspecific endonuclease activity.28,29 Expression of gp17 in E. coli from strong promoters such as phage T7 and λ pL promoters results in extensive cleavage of both the resident plasmid DNA and the E. coli genomic DNA. That this activity is associated with the large terminase protein gp17 was rigorously characterized using a number of control recombinant constructs. The cleaved DNA exhibits a characteristic smear that extends throughout the lane upon agarose gel electrophoresis. Mapping the ends of isolated cleaved DNA showed no sequence specificity. Further characterization of this activity revealed that the T4 terminase exhibits a preference towards the cleavage of transcriptionally active DNA.28,29

Since gp17 nonspecifically cleaves the E. coli genomic DNA, significant basal level expression of gp17 in leaky expression strains such as E. coli BL21 (DE3) is lethal. This characteristic was exploited to select point mutations that render this activity defective and hence would allow E. coli BL21 (DE3) to survive and form colonies.21 Such mutants, which are no longer lethal to E. coli, lack the characteristic cleavage of DNA upon gp17 expression. This strategy, as described above, allowed mapping of a terminase cutting site in the C-terminus of gp17.21,22

Franklin et al reported an in vitro terminase activity associated with the full-length gp17, which degrades single stranded DNA.20 The gp17 preferentially binds to the single stranded DNA at the junctions of single and double stranded DNA (putative recombinational or replicative intermediates) and degrades the single stranded portion of the junction in a nonspecific manner. While these data in some ways mimic the in vivo terminase activity discussed above, a significant reservation should be noted. The gp17 purified by Franklin et al binds to ssDNA cellulose columns tightly, an activity not reported with the other gp17 preparations.8,12,30 Also, their purified gp17 was not characterized by the in vitro DNA packaging assay in order to compare it with the other independently purified preparations, which were characterized by the bench-mark in vitro DNA packaging assay. The in vitro DNA packaging assay remains the only specific biological assay available to assess the functionality and identity of the purified protein.

Gp20: The Portal Protein

The T4 portal protein gp20 (61 kDa) was first determined to be directly connected to packaging by gene 20 cs mutations that blocked packaging initiation at low temperature. The prohead defect was reversed upon temperature shift or could be suppressed by specific gene 17 terminase mutations, thereby showing intimate association between terminase and portal proteins in packaging.31,32 As discussed above, a number of clustered cs mutations in the portal gene and their terminase suppressors have been sequenced.24 Early electron microscopic structure determination showed that the T4 portal is a dodecamer.33 In fact, it is likely that its 3D structure is similar to the Φ29 and SPP1 portal proteins.34 Assembly of the T4 portal is unusual in requiring a specific assembly factor or chaperone which leads to assembly on the cytoplasmic membrane.35-37 Assembly of the T4 prohead is in turn strongly dependent upon assembly of the membrane–linked portal, since in the absence of the portal protein, tubes rather than proheads form with a delay.35 Proteolytic processing of the prohead releases the prohead from the membrane—among the prohead structural proteins only the portal protein is not processed—and presumably this maturation frees the portal for interaction with the terminase in the cytoplasm, where electron microscopy reveals packaging occurs.4 In relation to the mechanism of DNA packaging and portal packaging function, modifications of the T4 portal protein by means of gene fusions reveal that bulky portal fusion proteins extending inside (gp20-gfp) or outside (HOC-gp20) the prohead can be assembled together with truncated gp20.30 Formation of active phage heads containing these altered portals poses a problem for DNA translocation models that invoke the necessity of portal rotation.38

Major Events during Phage T4 DNA Packaging

Packaging Initiation

In a common pathway for packaging initiation in dsDNA phages, the small terminase subunit recognizes the viral genome, facilitates the assembly of a holoterminase complex on the recognition site, the large terminase subunit then cuts the DNA, and the DNA end is linked to the empty prohead via interactions with the portal protein.2 Phage T4 follows this common pathway. Like the pac site phages, which recognize a unique sequence in the phage genome and cut near it to initiate a series of processive, random sequence ended phages, phage T4 generates circularly permuted ends. The evidence suggests that the phage T4 small terminase subunit gp16 also recognizes a pac site near the 3' end of its structural gene to create packaging initiation sites.

Evidence for T4 pac site recognition includes small terminase subunit gp16 binding to dsDNA containing the gene 16 pac site, analogous to the other phage terminase small subunits. 11 A variant H-T-H DNA binding motif is proposed in the gp16 sequence.5,11 In vivo genetic evidence for gp16 binding to the pac site includes the observations that: (i) under selection for higher copy numbers of gene 17 to increase its expression, gp16 is required for amplification of a gene 16 to gene 19 region of the phage T4 chromosome bounded by the pac site in gene 16 and a homologous region in gene 19; site directed mutagenesis of a 24 base pair homology region in the gene 16 pac site can eliminate the amplification,39,40,43 (ii) recombination between genes 16 and 19 in plasmids at the preferred amplification junction likewise requires synthesis of gp16;41 western blotting confirms that gp16 synthesis occurs from these gene 16 containing plasmids (H. Lin, unpublished data); (iii) plasmids and prophages containing the gene 16 pac site show enhanced transduction by the T4 transducing phage variant T4GT7, whereas site directed mutagenesis of the pac site lowers the transduction;42 and (iv) a pac site DNA fragment of ~10 kb extending from the unique BamHI site of phage T4 to gene 16 is found in mature phage T4 DNA.42 Thus, overall, a gene 16 pac sequence promotes transduction of pac-containing DNA, is found frequently to terminate mature T4 DNA, and gp16 is required to recognize the gene 16 pac sequence to promote genetic recombination in vivo between it and a homologous sequence in gene 19 to produce terminase amplification mutants.

The above observations strongly suggest that phage T4 gp16 recognizes and binds a pac sequence near the 3' end of gene 16. A model for gp16 DNA recognition and binding is that, as synthesized, the gp16 monomer binds to a preferred pac region toward the end of gene 16. Oligomerization releases gp16 from DNA binding and allows the gp16 ring to slide along the DNA and to participate in mature DNA end formation and packaging. Formation of the double ring could lead to the observed in vivo recombination-driven amplification of gp16 binding sequences due to synapsis of DNA segments each binding a single gp16 ring.42-44 This model proposes that the gp16 synapsis-related packaging function is to regulate DNA end formation by making it dependent upon DNA synthesis and accumulation, i.e., upon concatemerization of identical sequences.44 The model is consistent with the observation that the pac fragment in mature T4 DNA is more abundant when multiple plasmid gene 16 sequences are also present in the infected host,42 and with the absence of DNA binding by the gp16 multimer.11 Overall, this model for how gp16 recognizes the viral DNA substrate for packaging conforms to functions conserved among small terminase subunit proteins, including binding to a pac or cos region located within or near the small subunit gene. However, the mode of DNA binding by gp16 and the existence of a DNA bound gp16 ring form require experimental demonstration.

The old phage T4 literature is probably inadequate to distinguish between pac site and “random headful packaging” modes, which are mechanistically similar, especially if head filling is highly processive; i.e., an infrequent cut near to a pac site followed by multiple processive packaging rounds yields headfuls of apparently “randomized” DNA-containing phages. But, in fact, in the old phage T4 literature the most complete electron microscopic survey of heteroduplex DNAs formed from mature phage T4 particle DNAs concluded that although phage T4 mature DNA end sequences are dispersed over the chromosome, they are not “randomized” but distributed in a manner suggesting one or more pac sites in the genome.45 Given a gene 16 pac site, it is not excluded that other modes of end formation and packaging are also utilized; indeed, the observation that site directed mutagenesis of the gene 16 pac sequence which eliminates gene 17 amplification is not lethal,43 suggests that other packaging modes can intervene, as does the slow accumulation of filled heads in a gene 16 defective infection.10

Prohead Expansion

In T4 and other dsDNA bacteriophages, procapsid expansion accompanying DNA packaging is an important event, resulting in an increase in the inner capsid volume by about 50% in T4. In phage T4 head assembly, expansion is a striking transformation resulting in the stabilization of capsid subunit interactions, reorganization of capsid protein epitopes between the outer and inner surfaces of the capsid shell, and exposure of binding sites for the outer capsid accessory proteins Soc and Hoc.46,47 Changes in ATP- and DNA-binding properties of the assembled capsid protein have also been reported.48,49 This remarkable transformation of the capsid structure during expansion, and the timing of this transformation which coincides with DNA packaging, have raised interesting biological questions.

What is the trigger for prohead expansion? A number of studies indicate that the expansion transformation is an inherent property of the capsid protein, and can occur both in vitro and in vivo in isolation from DNA packaging.46,50,51 It is also clear that expansion is not energetically coupled to DNA packaging since much of the DNA enters an already expanded prohead and that DNA can be packaged in vitro into an expanded prohead.52 However, in vivo, it is likely that interaction of the terminase-DNA complex with the portal vertex and possibly the entrance of some DNA into the prohead, is the natural trigger for expansion. Jardine et al have identified an unexpanded, presumably cleaved, T4 prohead containing some DNA, which apparently can be chased into an expanded prohead and eventually phage.53,54 The amount of DNA associated with expansion in T4 is not characterized. There is better evidence for the above hypothesis in phages ? and T3 than in T4. It is well documented in the λ and T3 defined DNA packaging systems that expansion occurs after packaging of about 11% and 25% of the genomes, respectively.55,56 These data suggest that the terminase-prohead interactions, packaging of a small fraction of the DNA, prohead expansion, and continued packaging into an expanded head, occur in that order. Nevertheless, in both T4 and T3/T7, efficient initiation and completion of packaging into expanded proheads can be demonstrated, showing that there is no necessary mechanistic coupling between expansion and packaging.52,56,57

What is the biological significance of prohead expansion? The overall expansion transformation appears to be an energetically favorable process although it must be triggered.51 The initial thermodynamic barrier is probably overcome by specific terminase-DNA-portal interactions. It has been demonstrated that expanded T4 proheads are stabilized relative to the procapsid precursors.4 In addition to this function, expansion could play a role in DNA packaging or ejection. According to one hypothesis, the DNA entering the prohead interacts with the inner surface of the unexpanded capsid, serving as a “nucleator” for organization of the incoming DNA.1 The following expansion would weaken these interactions, allowing organization of the packaged DNA and its delivery during the infection process.1 Alternatively, expansion changes the inner capsid surface from a core-interacting surface to a DNA-interacting surface.52 The inner surface of the capsid, which is built with a core-interacting capability, should be transformed at the time of DNA packaging since a core-interacting surface would be irrelevant, if not detrimental, for the accompanying DNA packaging event. Consistent with these hypotheses, there is evidence in phage T4 for a shift in capsid protein epitopes from outside to inside of the capsid surface during expansion.58 How this bears on the structure of the packaged DNA is unknown. Phage T4 DNA is packed to ~500 mg/ml, the same density as other dsDNA phages. The condensed T4 DNA is oriented parallel to the long axis of the prolate head. Proteins encapsidated with the DNA display mobility within the condensate, since e.g., Staphlococcal nuclease can hydrolyze the packaged DNA.59

DNA Translocation

Although a number of interesting models have been proposed, viz., portal protein as a translocating rotor, topisomerase nicking and resealing, osmotic pump, tracking along the DNA, and conformational switching, the basic mechanism of DNA translocation is still very much a mystery. Although recent cryo-EM and X-ray structural analyses of the Φ29 portal ring allowed postulation of explicit details which link portal compression and rotation to translational movement of DNA,38 there is no evidence yet for a rotating portal. It is however clear that, regardless of the DNA translocation mechanism and which component does it, it is powered by an ATPase “motor”. The defined in vitro DNA packaging systems and structural data estimate that hydrolysis of one ATP is coupled to translocation of two base pairs of dsDNA.60,61 This means that roughly 105 molecules of ATP are expended to package one molecule of T4 DNA at an estimated catalytic rate of 104 ATPs per min per packaging unit. It is important to bear in mind however that these are merely estimates based on evidence from defined in vitro DNA packaging systems, which are inherently subject to a number of uncertainties.

A key question that needs resolution to further analyze the mechanistics of DNA packaging is: which component of the packaging machine (fig.1) is the ATPase motor that powers DNA translocation? A strong case can be made for the large terminase subunit gp17. Recent studies with T4 have taken the lead to address this question rigorously using a combination of molecular genetic and biochemical approaches. A combinatorial paradigm was developed to analyze the ATPase site I (fig. 3), which, as described above, appeared to be a critical motif in preliminary studies.19 Every possible amino acid substitution was introduced at every residue of the putative Walker-A motif SRQLGKT161-167 (fig. 4). Substitutions were either not tolerated or highly restricted at any of the invariant residues of the canonical G(A)XXXXGKT(S) Walker-A P-loop. Most strikingly, no substitutions, other than a conservative T167S, were tolerated at the GKT stretch, which is known to be responsible for binding ATP substrate (fig. 4). Atomic structures of a number of ATPases show that the ε-amino group of lysine interacts with the β-, γ-, phosphates of ATP, whereas the hydroxyl group of T coordinates with the Mg of the ATP-Mg complex. In the context of these well-established roles, the lethality of “conservative” substitutions such as G165A, K166R and T167A is likely due to a loss of specific and essential interactions between the ATP-Mg substrate and the catalytic center residues rather than to a major perturbation of gp17 structure or folding.19

This conclusion was further supported by the consistent biochemical phenotype exhibited by the ATP binding site I mutants. The purified K166G gp17 showed a complete loss of DNA packaging activity and the gp16-stimulated ATPase activity. The conservative mutants G165A, K166R, and T167A, showed a loss of in vitro DNA packaging activity but not the terminase (DNA cutting) activity. Thus, the two major functions of gp17 can be separated leading to the inference that the ATPase I site is required for the DNA packaging function. Recent data further show that the mutants in the proposed downstream Walker-B (MIYID251-255) residues, which are also predicted to be part of the same ATPase center, exhibited the same phenotypic pattern (M. Mitchell and V. Rao, unpublished data).

Thus, the evidence strongly supports the hypothesis that the T4 large terminase protein gp17 provides the ATPase motor function for the DNA packaging machine. In fact, a common ATPase center with conserved functional signatures has recently been discovered in the large terminase subunit of numerous dsDNA viruses, suggesting this site as the elusive translocating ATPase.5 Recent mutational and biochemical evidence on the analogous ATPase site in the large terminase subunit gpA of phage λ also corroborates this conclusion.62 However, more direct evidence is necessary in order to assign a causal role. It can be argued that DNA translocation could be catalyzed by a second “cryptic” ATPase that is activated upon the assembly of the complete packaging machine. Although unlikely, this possibility cannot be ruled out by the available evidence. Biochemical analyses of the conditionally lethal mutants are currently underway to rigorously assess the proposed linkage.

Based on the evidence presented above, a simple model for the assembly of the DNA translocating ATPase is proposed in Figure 5. This model invokes a trigger (stimulation) of the ATPase motor when it is connected to DNA translocation.12,19 Analysis of a number of packaging systems suggests that the rate of ATP consumption during DNA packaging is substantially greater than would be predicted from the kcat of the ATPase associated with the large terminase subunit. Assuming that the large terminase subunit alone is the translocating ATPase, this suggests a coupling of ATPase catalysis to DNA packaging, resulting in an increase in the catalytic capacity of the ATPase upon the assembly of a functional packaging machine and ensuing of DNA translocation. Perhaps the best evidence for the stimulated ATPase state is obtained in the T4 terminase system. As discussed above, the gp17-associated ATPase and DNA packaging activities are stimulated by about 50-fold in the presence of the small terminase subunit gp16.12,19 Considering the evidence that gp17 is hetero-disperse in solution and gp16 forms multimeric rings (or lock washers), it appears likely that the stably assembled gp16 facilitates multimerization of gp17 to form a gp16-gp17 holo-terminase complex. These interactions followed by conformational changes in gp17 result in the stimulation of the ATPase activity. In fact, gp16 dependent gp17 ATPase can be stimulated by an antiserum prepared against denatured gp17 to an activated enzymatic form hydrolyzing ~400 ATPs/gp17/min, an activity consistent with estimates of ATP hydrolysis requirements for DNA translocation (5,000 ATP/min), if gp17 forms an active ~12mer (~4800 ATP/min). And, in fact, it appears that this antibody-activated gp17 is a high molecular weight complex of roughly this dimension that does not require continued gp16 interaction for high turnover ATPase.74 Stimulation must occur also upon docking of the terminase-DNA complex with the gp20 portal since packaging in vitro (and in vivo) can occur in the absence of gp16. The proposed stimulations may represent common regulatory switches that couple ATP hydrolysis to a biological function in many systems.12,19

Figure 5. Model for the assembly of a functional packaging machine.

Figure 5

Model for the assembly of a functional packaging machine. The small T4 terminase protein gp16 exists as eight-subunit single or sixteen subunit double ring multimers in solution (see fig. 2), whereas the large terminase protein gp17 exists largely as (more...)

Interactions of DNA Packaging with Other DNA Processes

Packaging in vivo must interact with numerous other DNA processes, suggesting that regulatory mechanisms governing these interactions are necessary. In fact, in several phages initiation of DNA packaging requires terminase interactions with other proteins, which help to control the process or facilitate it. For example, in phage λ, the host coded proteins IHF and HU apparently bend DNA to promote terminase binding.3 Phage T4 generally operates relatively independently of host components thus suggesting terminase phage protein interactions may predominate. In fact, phage T4 terminase interacts with numerous phage proteins. In addition to interacting with itself, the small terminase subunit (gp16), and the portal (gp20), the phage T4 terminase gp17 subunit is known to interact with a number of phage T4 DNA binding proteins, namely gp32 (single strand DNA binding protein), gp55 (late T4 σ factor), as well as possibly gp45 (DNA sliding clamp25). The terminase interaction with the late T4 σ-factor is intriguing in view of the participation of the T7/T3 RNA polymerase in DNA packaging and DNA repair synthesis. Since gp17 shows little or no affinity for DNA, these interactions with DNA directed proteins may serve to promote terminase loading onto DNA for packaging as well as to regulate packaging initiation. Additionally, it is probable that the DNA replication machinery should be mobilized to the site of packaging initiation in order to repair the concatemer following the double strand DNA break required for DNA packaging. It is tempting to speculate that the terminase is a criticial regulatory protein whose structure and activity depend upon these multiple protein interactions.

DNA Structural Requirements for Packaging

Probably the best-understood example of DNA structure representing a roadblock to packaging arises in the case of T4 gene 49 (endonuclease VII) mutations. T4 gene 49 mutations lead to the accumulation of partially filled heads, apparently because of the accumulation of branches (recombinational intermediates) which can be visualized outside the partially filled head.63,64 The gene 49 endonuclease is able to resolve Holliday structures in DNA as well as display other activities consistent with a role in removing such branches in the concatemeric DNA. The phenotype of temperature-sensitive gene 49 mutations therefore demonstrates that DNA damage can arrest DNA packaging, and DNA repair processes can rescue arrested intermediates. Interestingly, the gene 49 endonuclease binds to the gp20 portal protein, positioning it favorably for resolution of branched structures as these encounter the prohead.65

DNA ligase is similarly required to complete DNA filling of T4 proheads in vivo. When both T4 and host DNA ligase temperature-sensitive mutant enzymes are inactivated, following the accumulation of a large DNA concatemer pool in the presence of ligase, full heads accumulate to only about 3% of the total. Restoration of ligase activity by temperature shift allows the major product, partially filled heads, to be filled to active heads.66 These experiments demonstrate that either the T4 or the E. coli ligase can satisfy the packaging requirement. Whether the ligase lesion is a nick or more extensive discontinuity in duplex DNA is unknown, although the duplex concatemer must be interrupted to allow late transcription.66 In phage T4 a number of “early” DNA replication-recombination functions, when altered by mutation, lead to defects in packaging as revealed by the accumulation of partially filled heads as the predominant product, suggesting that defects in the concatemer block continuation of DNA translocation. 67 On the other hand, heteroduplex loops of 19 bases can be packaged into λ heads, a surprising degree of deviation from dsDNA for models that gear dsDNA translocation into the head.68 Can a single strand extension of a translocating double strand also be translocated into the prohead? Greater knowledge of the DNA structural requirements for packaging would provide necessary checks on models for the translocation mechanism.

Discontinuous Headful Packaging

An interesting twist to the classical headful packaging model has been discovered during the large DNA cloning experiments in phage T4 and phage P1 in vitro packaging.69,70 Both the cloning systems package linear headful size foreign DNA molecules cloned within a unique site of the vector followed by transduction of these molecules into E. coli. Surprisingly, a fraction of the clones showed very small, less than headful length, inserts. Subsequent experiments using only the 7-30 kb size vector DNAs as the packaging substrate revealed that these phages can indeed package and transduce DNA molecules that are much smaller than the headful length DNA. But, these particles have the same density as the wild type suggesting that multiple plasmid molecules are packaged within the same head. Alternate mechanisms such as recombination or rolling circle replication that might convert the less than headful length DNA to a headful size have been eliminated.70 These results have some interesting implications: (i) the packaging machinery does not have a “ruler” capability to measure the length of the packaging substrate before initiation of packaging and discriminate against packaging of small molecules; (ii) the partially packaged, presumably expanded, head is capable of initiating and packaging a second DNA molecule; and (iii) ds DNA packaging is a remarkably flexible process, and can occur by multiple modes; normally, the unit lengths are cut from concatemers (in vivo), occasionally, unit length linear molecules are packaged from end-to-end (both in vitro and in vivo), and rarely, multiple molecules are packaged to fill the head volume.


Phage T4 fundamentally employs the same mechanisms as the other well characterized dsDNA bacteriophages for packaging DNA. However, it also presents some unique challenges. For instance, the mechanism of DNA substrate recognition and regulation of concatemer maturation and cutting represents one such problem. This is complicated by the fact that the T4 packaging substrate is an “endless” concatemer with multiple branches that are associated with protein complexes. T4 may exploit multiple modes of substrate recognition and cutting, but a close collaboration between packaging and DNA recombination and repair processes is likely essential to repair the packaging concatemer and to resolve competing DNA packaging complexes. The fact that T4 terminase associates with components (gp32, gp55) of the late transcription and replication pathways is direct evidence for probable coupling of packaging and DNA repair processes. Understanding the dynamics of these interactions and elucidating an integrated mechanism may generate novel insights into the overall mechanisms of regulation of DNA metabolic pathways.

With its similarity to the putative herpes viral terminases and the presence of a single critical ATPase center (other terminases reportedly have two or more ATPase sites18,71,72), the proposed ATPase motor in the large terminase subunit gp17 offers an excellent model to elucidate the DNA packaging mechanism. The collection of unique mutants generated by the powerful molecular genetic and biochemical approaches, including the very rare conditionally lethal mutants, offer a unique resource to dissect the molecular details of ATP energy transduction into mechanical movement of DNA. The data may also have broad implications to the general understanding of energy and signal transduction mechanisms. The ATP (GTP) consensus motif is apparently one of the most common motifs found in genomes (up to 5-10% of all expressed proteins), and may represent one of the fundamental (and ancient) motifs in biological systems.73 Numerous ATPase (GTPase) systems and molecular motors use this motif to trap the energy-rich ATP and regulate and couple its hydrolysis to power numerous biological processes. Biochemical analyses of the mutant collection of gp17 may provide a more precise understanding of the common molecular switches involved in energy/signal regulation and coupling.


The research in VBR's laboratory is supported by National Science Foundation (MCB-0110574) and that in LWB's laboratory is supported by National Institutes of Health (AI11676).


Earnshaw WC, Casjens SR. DNA packaging by the double-stranded DNA bacteriophages. Cell. 1980;21:319–331. [PubMed: 6447542]
Black LW. DNA packaging in dsDNA bacteriophages. Annu Rev Microbiol. 1989;43:267–292. [PubMed: 2679356]
Catalano CE, Cue D, Feiss M. Virus DNA packaging: The strategy used by phage lambda. Mol Microbiol. 1995;16:1075–1086. [PubMed: 8577244]
Black LW, Showe MK, Steven AC. Morphogenesis of the T4 headIn: Karam JD, ed.Molecular biology of bacteriophage T4Washington, DC: ASM Press,1994. 218–258.
Mitchell M, Matsuzaki S, Imai S. et al. Sequence analysis of bacteriophage T4 DNA-packaging/ terminase genes 16 and 17 reveals a common ATPase center in the large subunit of viral terminases. Nucleic Acids Res. 2002;30:4009–4021. [PMC free article: PMC137109] [PubMed: 12235385]
Smith MCM, Burns RN, Wilson SE. et al. The complete genome sequence of the streptomyces temperate phage fC31: Evolutionary relationships to other viruses. Nucleic Acids Res. 1999;27(No 10):2145–2155. [PMC free article: PMC148434] [PubMed: 10219087]
Davison AJ. Channel catfish virus: A new type of herpesvirus. Virology. 1992;186:9–14. [PubMed: 1727613]
Rao VB, Black LW. Cloning, overexpression and purification of the terminase proteins gp16 and gp17 of bacteriophage T4: Construction of a defined in vitro packaging system using purified terminase proteins. J Mol Biol. 1988;200:475–488. [PubMed: 3294420]
Powell D, Franklin J, Arisaka F. et al. Bacteriophage T4 DNA packaging genes 16 and 17. Nucl Acids Res. 1990;18:4005. [PMC free article: PMC331121] [PubMed: 2374730]
Wunderli H, van den Brock J, Kellenberger E. Studies related to the head-maturation pathway of bacteriophage T4 and T2. I. Morphology and kinetics of intracellular particles produced by mutants in the maturation genes. J Supramol Struct. 1977;7:135–161. [PubMed: 599933]
Lin H, Simon MN, Black LW. Purification and characterization of the small subunit of phage T4 terminase, gp16, required for DNA packaging. J Biol Chem. 1997;272:3495–3501. [PubMed: 9013596]
Leffers G, Rao VB. Biochemical characterization of an ATPase activity associated with the large DNA packaging unit gp17 from bacteriophage T4. J Biol Chem. 2000;275:37127–37136. [PubMed: 10967092]
Yang Q, Berton N, Manning MC. et al. Domain structure of gpNuI, a phage lambda DNA packaging protein. Biochemistry. 1999;38:14238–14247. [PubMed: 10571997]
Franklin J, Mosig G. Expression of the bacteriophage T4 DNA terminase genes 16 and 17 yields multiple proteins. Gene. 1996;177:179–189. [PubMed: 8921865]
Becker A, Gold M. Prediction of an ATP reactive center in the small subunit, gpNuI, of the phage lambda terminase enzyme. J Mol Biol. 1988;199:219–222. [PubMed: 2965248]
de BeerT, Fang J, Ortega M. et al. Insights into specific DNA recognition during the assembly of a viral genome packaging machine. Molecular Cell. 2002;9:981–991. [PubMed: 12049735]
Via A, Ferre F, Brannetti B. et al. Three-dimensional view of the surface motif associated with the P-loop structure: Cis and trans cases of convergent evolution. J Mol Biol. 2000;303:455–465. [PubMed: 11054283]
Guo P, Peterson C, Anderson D. The DNA packaging protein gp16 of bacteriophage f29 is a prohead- and DNA-gp3-dependent ATPase. J Mol Biol. 1987;197:229–236. [PubMed: 2960820]
Rao VB, Mitchell MS. The N-terminal ATPase site in the large terminase protein gp17 is critically required for DNA packaging in bacteriophage T4. J Mol Biol. 2001;314:401–411. [PubMed: 11846554]
Franklin J, Haseltine D, Davenport L. et al. The largest (70 kDa) product of the bacteriophage T4 DNA terminase gene 17 binds to single stranded DNA segments and digests them towards junctions with double-stranded DNA. J Mol Biol. 1998;277:541–557. [PubMed: 9533879]
Kuebler D, Rao VB. Functional analysis of the DNA-packaging/terminase protein gp17 from bacteriophage T4. J Mol Biol. 1998;281:803–814. [PubMed: 9719636]
Rentas F. Molecular genetic and biochemical analysis of a putative terminase cutting site in the large DNA packaging protein gp17 from bacteriophage T4. PhD thesis. Washington, DC: The Catholic University of America. 2001
Gerit JA. Mechanistic principles of enzyme catalyzed cleavage of phosphodiester bonds in nucleases 2nd edNY: Cold Spring Harbor Press,1993. 1–34.
Lin H, Rao VB, Black LW. Analysis of capsid portal protein and terminase functional domains: Interaction Sites Required for DNA Packaging in Bacteriophage T4. J Mol Biol. 1999;289:249–260. [PubMed: 10366503]
Malys N, Chang D-Y, Baumann RG. et al. A bipartite bacteriophage T4 SOC and HOC randomized peptide display library: Detection and analysis of phage T4 terminase (gp17) and late s factor (gp55) interaction. J Mol Biol. 2002;319:289–304. [PubMed: 12051907]
Zhang Z, Sumbilla C, Lewis D. et al. Mutational analysis of the peptide segment linking phosphorylation and Ca(2+)-binding domains in the sarcoplasmic reticulum Ca(2+)-ATPase. J Biol Chem. 1995;270:16283–16290. [PubMed: 7608196]
Louie D, Serwer P. Blunt-ended ligation can be used to produce DNA ladders with rung spacing as large as 0.17 Mb. Nucleic Acids Res. 1990;18:3090. [PMC free article: PMC330875] [PubMed: 2349126]
Bhattacharyya SP, Rao VB. A novel terminase activity associated with the DNA packaging protein gp17 of bacteriophage T4. Virology. 1993;196:34–44. [PubMed: 8356805]
Bhattacharyya SP, Rao VB. Structural analysis of DNA cleaved in vivo by bacteriophage T4 terminase. Gene. 1994;146:67–72. [PubMed: 8063105]
Baumann RG. The role of portal and terminase proteins in the mechanism of DNA packaging in the dsDNA bacteriophage T4. PhD thesis. Baltimore, Md: University of Maryland Med Sch. 2002
Hsiao CL, Black LW. Head morphogenesis of bacteriophage T4. III. The role of gene 20 in DNA packaging. Virology. 1978;91:26–38. [PubMed: 726265]
Hsiao CL, Black LW. DNA packaging and the pathway of bacteriophage T4 head assembly. Proc Natl Acad Sci USA. 1977;74:3652–3656. [PMC free article: PMC431676] [PubMed: 269421]
Driedonks RA, Engel A, ten Heggeler B. et al. Gene 20 product of bacteriophage T4 its purification and structure. J Mol Biol. 1981;152:641–662. [PubMed: 7334518]
Valpuesta JM, Carrascosa JL. Structure of viral connectors and their function in bacteriphage assembly and DNA packaging. Q Rev Biophys. 1994;27:107–155. [PubMed: 7984775]
Hsiao CL, Black LW. Head morphogenesis of bacteriophage T4. II. The role of gene 40 in initiating prehead assembly. Virology. 1978;91:15–25. [PubMed: 726259]
Michaud G, Zachary A, Rao VB. et al. Membrane associated assembly of a phage T4 DNA entrance vertex structure studied with expression vectors. J Mol Biol. 1989;209:667–681. [PubMed: 2685327]
Yap N, Rao VB. Novel mutants in the 5'-upstream region of the portal protein gene 20 overcome a gp40-dependent prohead assembly block in bacteriophage T4. J Mol Biol. 1996;263:539–550. [PubMed: 8918937]
Simpson AA, Tao Y, Leiman PG. et al. Structure of the bacteriophage Φ29 DNA packaging motor. Nature. 2000;408:745–750. [PMC free article: PMC4151180] [PubMed: 11130079]
Wu DG, Black LW. Gene amplification mechanism for the hyperproduction of T4 bacteriophage 17 and 18 proteins. J Mol Biol. 1987;195:769–783. [PubMed: 3498845]
Wu DG, Wu CH, Black LW. Reiterated gene amplifications at specific short homology sequences in phage T4 produce Hp17 mutants. J Mol Biol. 1991;218:705–721. [PubMed: 2023246]
Wu CHH, Lin H, Black LW. Bacteriophage T4 gene 17 amplification mutants: Evidence for initiation by the T4 terminase subunit gp16. J Mol Biol. 1995;247:523–528. [PubMed: 7723009]
Lin H, Black LW. DNA requirements in vivo for phage T4 packaging. Virology. 1998;242:118–127. [PubMed: 9501053]
Wu CHH, Black LW. Mutational analysis of the sequence-specific recombination box for amplification of gene 17 of bacteriophage T4. J Mol Biol. 1995;247:604–617. [PubMed: 7723018]
Black LW. DNA packaging and cutting by phage terminases: Control in phage T4 by a synaptic mechanism. Bioessays. 1995;17:1052–1060. [PubMed: 8634063]
Grossi GF, Macchiato MF, Gialanella G. Circular permutation analysis of phage T4 DNA by electron microscopy. Z Naturforsch. 1983;38:294–296. [PubMed: 6346725]
Carrascosa JL. Head maturation pathway of bacteriophage T4 and T2. IV. In vitro transformation of T4 head-related particles produced by mutants in gene 17 to capsid-like structures. J Virol. 1978;26:420–428. [PMC free article: PMC354079] [PubMed: 660719]
Steven AC, Greenstone H, Bauer AC. et al. The maturation-dependent conformational change of the major capsid protein of bacteriophage T4 involves a substantial change in secondary structure. Biochemistry. 1990;29:5556–5561. [PubMed: 2386785]
Rao VB, Black LW. Evidence that a phage T4 DNA packaging enzyme is a processed form of the major capsid gene product. Cell. 1985;185:565–578. [PubMed: 2932225]
Xue MQ, Black LW. Role of the major capsid protein of phage T4 in DNA packaging from structure-function and site-directed mutagenesis studies. J Struct Biol. 1990;104:75–83. [PubMed: 2150914]
Galisteo ML, King J. Conformational transitions in the protein lattice of phage P22 capsids. Biophys J. 1993;65:227–235. [PMC free article: PMC1225718] [PubMed: 8369433]
Steven AC. Conformational change–an alternative energy source? Exothermic phase transition in phage capsid maturation. Biophys J. 1993;65:5–6. [PMC free article: PMC1225687] [PubMed: 8369454]
Rao VB, Black LW. DNA packaging of bacteriophage T4 proheads in vitro: Evidence that prohead expansion is not coupled to DNA packaging. J Mol Biol. 1985;185:565–578. [PubMed: 4057255]
Jardine PJ, McCormick MC, Lutze-Wallace C. et al. The bacteriophage T4 DNA packaging apparatus targets the unexpanded prohead. J Mol Biol. 1998;284:647–659. [PubMed: 9826505]
Jardine PJ, Coombs DH. Capsid expansion follows the initiation of DNA packaging in bacteriophage T4. J Mol Biol. 1998;284:661–672. [PubMed: 9826506]
Hohn B. DNA sequences necessary for packaging of bacteriophage lambda DNA. Proc Natl Acad Sci USA. 1983;80:7456–7460. [PMC free article: PMC389970] [PubMed: 6324174]
Shibata H, Fujisawa H, Minagawa T. Characterization of the bacteriophage T3 DNA packaging reaction in vitro in a defined system. J Mol Biol. 1987;196:845–51. [PubMed: 3316664]
Son , Watson RH, Serwer P. The direction and rate of bacteriophage T7 DNA packaging in vitro. Virology. 1993;196:282–289. [PubMed: 8356800]
Steven AC, Bauer AC, Bisher ME. et al. The maturation-dependent conformational change of phage T4 capsid involves the translocation of specific epitopes between the inner and outer capsid surfaces. J Struct Biol. 1991;106:221–236. [PubMed: 1725126]
Mullaney J, Black LW. Activity of foreign proteins targeted within the bacteriophage T4 head and prohead: Implications for packaged DNA structure. J Mol Biol. 1998;283:913–929. [PubMed: 9799633]
Morita M, Tasaka M, Fujisawa H. DNA packaging ATPase of bacteriophage T3. Virology. 1993;193:748–752. [PubMed: 8460483]
Guo P, Peterson C, Anderson D. Initiation events in in-vitro packaging of bacteriophage phi 29 DNA-gp3. J Mol Biol. 1987;197:219–228. [PubMed: 3119862]
Duffy C, Feiss M. The large subunit of bacteriophage lambda's terminase plays a role in DNA translocation and packaging termination. J Mol Biol. 2002;316:547–561. [PubMed: 11866517]
Luftig RB, Wood WB, Okinaka R. Bacteriophage T4 head morphogenesis. On the nature of gene 49 defective heads and their role as intermediates. J Mol Biol. 1971;57:555–573. [PubMed: 5580436]
Kemper B, Brown DT. Function of gene 49 of bacteriophage T4. II. Analysis of intracellular development and the structure of very fast-sedimenting DNA. J Virol. 1976;18:1000–1015. [PMC free article: PMC354800] [PubMed: 775128]
Golz S, Kemper B. Association of holliday-structure resolving endonuclease VII with gp20 from the packaging machine of phage T4. J Mol Biol. 1999;285:1131–1144. [PubMed: 9918721]
Zachary A, Black LW. DNA ligase is required for encapsidation of bacteriophage T4 DNA. J Mol Biol. 1981;149:641–658. [PubMed: 7031271]
Zachary A, Black LW. Topoisomerase II and other DNA-delay and DNA-arrest mutations impair bactreriophage T4 DNA packaging in vivo and in vitro. J Virol. 1986;60:97–104. [PMC free article: PMC253906] [PubMed: 3018302]
Pearson RK, Fox MS. Effects of DNA heterologies on bacteriophage λ packaging. Genetics. 1988;118:5–12. [PMC free article: PMC1203266] [PubMed: 8608932]
Coren J, Pierce J, Sternberg N. Headful packaging revisited: The packaging of more than one DNA molecule into a bacteriophage P1 head. J Mol Biol. 1995;249:176–184. [PubMed: 7776370]
Leffers G, Rao V. A discontinuous headful packaging model for packaging of less than headful length DNA molecules by bacteriophage T4. J Mol Biol. 1996;258:839–850. [PubMed: 8637014]
Hang JQ, Tack BF, Feiss M. ATPase center of bacteriophage lambda terminase involved in post-cleavage stages of DNA packaging: Identification of ATP-interacting amino acids. J Mol Biol. 2000;302:777–795. [PubMed: 10993723]
Morita M, Tasaka M, Fujisawa H. Analysis of functional domains of the packaging proteins of bacteriophage T3 by site-directed mutagenesis. J Mol Biol. 1994;235:248–259. [PubMed: 8289246]
Koonin EV, Tatusov RL, Rudd KE. Sequence similarity analysis of Escherichia coli proteins: Functional and evolutionary implications. Proc Natl Acad Sci USA. 1995;92:11921–11925. [PMC free article: PMC40515] [PubMed: 8524875]
Baumann RG, Black LW. Isolation and characterization of T4 bacteriophage gp17 terminase: A large subunit multimer with enhanced ATPase activity. J Biol Chem. 2003;278:4618–4627. [PubMed: 12466275]