DNA packaging of bacteriophage T4 proheads in vitro. Evidence that prohead expansion is not coupled to DNA packaging

We developed a system for DNA packaging of isolated bacteriophage T4 proheads in vitro and studied the role of prohead expansion in DNA packaging. Biologically active proheads have been purified from a number of packaging-deficient mutant extracts. The cleaved mature prohead is the active structural precursor for the DNA packaging reaction. Packaging of proheads requires ATP, Mg2+ and spermidine, and is stimulated by polyethylene glycol and dextran. Predominantly expanded proheads (ELPs) are produced at 37 degrees C and predominantly unexpanded proheads (ESPs) are produced at 20 degrees C. Both the expanded and unexpanded proheads are active in DNA packaging in vitro. This is based on the observations that (1) both ESPs and ELPs purified by chromatography on DEAE-Sephacel showed DNA packaging activity; (2) apparently homogeneous ELPs prepared by treatment with sodium dodecyl sulfate (which dissociates ESPs) retained significant biological activity; (3) specific precipitation of ELPs with anti-hoc immunoglobulin G resulted in loss of DNA packaging activity; and (4) ESPs upon expansion in vitro to ELPs retained packaging activity. Therefore, contrary to the models that couple DNA packaging to head expansion, in T4 the expansion and packaging appear to be independent, since the already expanded DNA-free proheads can be packaged in vitro. We therefore propose that the unexpanded to expanded prohead transition has evolved to stabilize the capsid and to reorganize the prohead shell functionally from a core-interacting to a DNA-interacting inner surface.     

Maturation of the head of bacteriophage T4. I. DNA packaging events

Pulse-chase experiments in wild-type and mutant phage-infected cells provide evidence that the following particles called prohead I, II and III are successive precursors to the mature heads. The prohead I particles contain predominantly the precursor protein P23 and possibly P22 (mol. wt 31,000) and IP III (mol. wt 24,000) and have an s value of about 400 S. Concomitantly with the cleavage of most of P23 (mol. wt 55,000) to P231 (mol. wt 45,000), they are rapidly converted into prohead II particles which sediment with about 350 S. The prohead II particles contain, in addition to P231, the major constituents of the viral shella—a core consisting of proteins P22 and IP III. In cell lysates, prohead I and prohead II particles contain no DNA in a DNase-resistant form and are not bound to the replicative DNA. We cannot, however, positively rule out the possibility that these particles may have contained some DNA while in the cells.The prohead II particles are in turn converted into particles which sediment with about 550 S after DNase treatment (prohead III). During this conversion about 50% of normal DNA complement becomes packaged in a DNase-resistant form, and roughly 50% of the core proteins P22 and IP III are cleaved. In lysates the prohead III particles are attached to the replicative DNA. The prohead III particle appears to be the immediate precursor of the full mature head (1100 S). Cleavage of protein P22 to small polypeptides and conversion of IP III IP III1 are completed at this time. No precursor proteins are found in the full heads. Studies with various mutant phage showed that the prohead II to III conversion is blocked by mutations in genes 16 and 17 and that the conversion of the prohead III particles to the mature heads is blocked by mutations in gene 49. Cleavage of the head proteins, however, occurs normally in these mutant-infected cells. We conclude that the cleavage of the major component of the viral shell, P23, into P231 precedes the DNA packaging event, whereas cleavage of the core proteins P22 and IP III appears to be intimately linked to the DNA packaging event. Models relating the cleavage processes to DNA encapsulation are discussed.     

Evidence that a phage T4 DNA packaging enzyme is a processed form of the major capsid gene product

A phage T4 DNA packaging enzyme appears to arise as a processed form of the major T4 capsid structural protein gp23. The enzyme activity and antigen are missing from all head gene mutants that block the morphogenetic proteolytic processing reactions of the head proteins in vivo. The enzyme antigen can be formed in vitro by T4 (gp21) specific processing of gp23 containing extracts. Enzyme antigen is found in active processed proheads but not in full heads. The enzyme and the major capsid protein show immunological cross-reactivity, produce common peptides upon proteolysis, and share an assembly-conformation-dependent ATP binding site. The packaging enzyme and the mature capsid protein (gp23*) both appear to arise from processing of gp23, the former as a minor product of a specific gp23 structure in the prohead, acting in DNA packaging as a DNA-dependent ATPase, and a headful-dependent terminase.     

Double-stranded DNA bacteriophage prohead protease is homologous to herpesvirus protease

Double-stranded DNA bacteriophages and herpesviruses assemble their heads in a similar fashion; a pre-formed precursor called a prohead or procapsid undergoes a conformational transition to give rise to a mature head or capsid. A virus-encoded prohead or procapsid protease is often required in this maturation process. Through computational analysis, we infer homology between bacteriophage prohead proteases (MEROPS families U9 and U35) and herpesvirus protease (MEROPS family S21), and unify them into a procapsid protease superfamily. We also extend this superfamily to include an uncharacterized cluster of orthologs (COG3566) and many other phage or bacteria-encoded hypothetical proteins. On the basis of this homology and the herpesvirus protease structure and catalytic mechanism, we predict that bacteriophage prohead proteases adopt the herpesvirus protease fold and exploit a conserved Ser and His residue pair in catalysis. Our study provides further support for the proposed evolutionary link between dsDNA bacteriophages and herpesviruses.     

Initiation events in in-vitro packaging of bacteriophage phi 29 DNA-gp3

Initiation events in the packaging of bacteriophage phi 29 DNA-gp3 (DNA-gene product 3 complex) were studied in a completely defined in-vitro system that included purified proheads, DNA-gp3 and the DNA packaging protein gp16. In the sequential interactions, gp16 first bound to, and was modified by, the prohead. The prohead-gp16 complex then bound to DNA-gp3, resulting in a second modification of gp16 that permitted binding of ATP. DNA-gp3 aggregates were produced, and the hydrolysis of ATP accompanied DNA-gp3 packaging. Binding and hydrolysis of ATP by gp16 was both prohead- and DNA-gp3-dependent. Interruption of packaging by DNase I addition revealed filled heads but few particles containing partial lengths of DNA, suggesting that following a rate-limiting initiation, the translocation of DNA-gp3 into the prohead was much faster in the defined in-vitro system than in extracts.     

Portal-large terminase interactions of the bacteriophage T4 DNA packaging machine implicate a molecular lever mechanism for coupling ATPase to DNA translocation

DNA packaging by double-stranded DNA bacteriophages and herpesviruses is driven by a powerful molecular machine assembled at the portal vertex of the empty prohead. The phage T4 packaging machine consists of three components: dodecameric portal (gp20), pentameric large terminase motor (gp17), and 11- or 12-meric small terminase (gp16). These components dynamically interact and orchestrate a complex series of reactions to produce a DNA-filled head containing one viral genome per head. Here, we analyzed the interactions between the portal and motor proteins using a direct binding assay, mutagenesis, and structural analyses. Our results show that a portal binding site is located in the ATP hydrolysis-controlling subdomain II of gp17. Mutations at key residues of this site lead to temperature-sensitive or null phenotypes. A conserved helix-turn-helix (HLH) that is part of this site interacts with the portal. A recombinant HLH peptide competes with gp17 for portal binding and blocks DNA translocation. The helices apparently provide specificity to capture the cognate prohead, whereas the loop residues communicate the portal interaction to the ATPase center. These observations lead to a hypothesis in which a unique HLH-portal interaction in the symmetrically mismatched complex acts as a lever to position the arginine finger and trigger ATP hydrolysis. Transiently connecting the critical parts of the motor; subdomain I (ATP binding), subdomain II (controlling ATP hydrolysis), and C-domain (DNA movement), the portal-motor interactions might ensure tight coupling between ATP hydrolysis and DNA translocation.     

Mechanism of head assembly and DNA encapsulation in Salmonella phage P22. II. Morphogenetic pathway

We have identified and characterized structural intermediates in phage P22 assembly. Three classes of particles can be isolated from P22-infected cells: 500 S full heads or phage, 170 S empty heads, and 240 S “proheads”. One or more of these classes are missing from cells infected with mutants defective in the genes for phage head assembly. By determining the protein composition of all classes of particles from wild type and mutant-infected cells, and examining the time-course of particle assembly, we have been able to define many steps in the pathway of P22 morphogenesis.In pulse-chase experiments, the earliest structural intermediate we find is a 240 S prohead; it contains two major protein species, the products of genes 5 and 8. Gene 5 protein (p5) is the major phage coat protein. Gene 8 protein is not found in mature phage. The proheads contain, in addition, four minor protein species, PI, P16, P20 and PX. Similar prohead structures accumulate in lysates made with mutants of three genes, 1, 2 and 3, which accumulate uncut DNA. The second intermediate, which we identify indirectly, is a newly filled (with DNA) head that breaks down on isolation to 170 S empty heads. This form contains no P8, but does contain five of the six protein species of complete heads. Such structures accumulate in lysates made with mutants of two genes, 10 and 26.Experiments with a temperature-sensitive mutant in gene 3 show that proheads from such 3^? infected cells are convertible to mature phage in vivo, with concomitant loss of P8. The molecules of P8 are not cleaved during this process and the data suggest that they may be re-used to form further proheads.Detailed examination of 8^? lysates revealed aberrant aggregates of P5. Since P8 is required for phage morphogenesis, but is removed from proheads during DNA encapsulation, we have termed it a scaffolding protein, though it may have DNA encapsulation functions as well.All the experimental observations of this and the accompanying paper can be accounted for by an assembly pathway, in which the scaffolding protein P8 complexes with the major coat protein P5 to form a properly dimensioned prohead. With the function of the products of genes 1, 2 and 3, the prohead encapsulates and cuts a headful of DNA from the concatemer. Coupled with this process is the exit of the P8 molecules, which may then recycle to form further proheads. The newly filled heads are then stabilized by the action of P26 and gene 10 product to give complete phage heads.     

Structure and assembly of the capsid of bacteriophage P22.

Identification of the genes and proteins involved in phage P22 formation has permitted a detailed analysis of particle assembly, revealing some unexpected aspects. The polymerization of the major coat protein (gene 5 product) into an organized capsid is directed by a scaffolding protein (gene 8 product) which is absent from mature phage. The resulting capsid structure (prohead) is the precursor for DNA encapsidation. All of the scaffolding protein exits from the prohead in association with DNA packaging. These molecules then recycle, directing further rounds of prohead assembly. The structure of the prohead has been studied by electron microscopy of thin sections of phage infected cells, and by low angle X-ray scattering of concentrated particles. The results show that the prohead is a double shell structure, or a ball within a shell. The inner ball or shell is composed of the scaffolding protein while the outer shell is composed of coat protein. The conversion from prohead to mature capsid is associated with an expansion of the coat protein shell. It is possible that the scaffolding protein molecules exit through the capsid lattice. When DNA encapsidation within infected cells is blocked by mutation, scaffolding protein is trapped in proheads and cannot recycle. Under these conditions, the rate of synthesis of gp8 increases, so that normal proheads continue to form. These results suggest that free scaffolding protein negatively regulates its own further synthesis, providing a coupling between protein synthesis and protein assembly.     

A promiscuous DNA packaging machine from bacteriophage T4

Complex viruses are assembled from simple protein subunits by sequential and irreversible assembly. During genome packaging in bacteriophages, a powerful molecular motor assembles at the special portal vertex of an empty prohead to initiate packaging. The capsid expands after about 10%-25% of the genome is packaged. When the head is full, the motor cuts the concatemeric DNA and dissociates from the head. Conformational changes, particularly in the portal, are thought to drive these sequential transitions. We found that the phage T4 packaging machine is highly promiscuous, translocating DNA into finished phage heads as well as into proheads. Optical tweezers experiments show that single motors can force exogenous DNA into phage heads at the same rate as into proheads. Single molecule fluorescence measurements demonstrate that phage heads undergo repeated initiations, packaging multiple DNA molecules into the same head. These results suggest that the phage DNA packaging machine has unusual conformational plasticity, powering DNA into an apparently passive capsid receptacle, including the highly stable virus shell, until it is full. These features probably led to the evolution of viral genomes that fit capsid volume, a strikingly common phenomenon in double-stranded DNA viruses, and will potentially allow design of a novel class of nanocapsid delivery vehicles.     

Model for DNA packaging into bacteriophage T4 heads.

The mechanism of DNA packaging into bacteriophage T4 heads in vivo was investigated by glucosylation of hydroxymethylcytosine residues in a conditionally glucose-deficient host. Cytoplasmic DNA associated with partially packaged ts49 heads can be fully glucosylated, whereas DNA already packaged into these heads is shown to be resistant to glucosylation. After temperature shift and completion of arrested packaging into the reversible temperature-sensitive ts49 head, the structure of the DNA in the mature ts49 phage was investigated by restriction enzyme digestion, autoradiography, and other techniques. Such mature DNA appears to be fully glucosylated along part of its length and nonglucosylated on the remainder. Its structure suggests that the DNA is run into the head linearly and unidirectionally from one mature end and that there is little sequence specificity in that portion of the T4 DNA which first enters the capsid. This technique should be useful in investigation of the three-dimensional structure of first- and last-packaged DNA within the head; preliminary studies including autoradiography of osmotically shocked phage suggest that the DNA which first enters the head is deposited toward the center of the capsid and that the end of the DNA which first enters the head exits first upon injection. In conjunction with studies of the structure of condensed DNA, the positions and functions of T4 capsid proteins in DNA packaging, and the order of T4 packaging functions [Earnshaw and Harrison, Nature (London) 268:598-602, 1977; Hsiao and Black, Proc. Natl. Acad. Sci. U.S.A. 74:3652-3656, 1977; Müller-Salamin et al., J. Virol. 24:121-134, 1977; Richards et al., J. Mol. Biol. 78:255-259, 1973], the features described above suggest the following model: the first DNA end is fixed to the proximal apex of the head at p20 and the DNA is then pumped into the head enzymatically by proteins (p20 + p17) which induce torsion in the DNA molecule.     

Morphogenesis of bacteriophage phi 29 of Bacillus subtilis: preliminary isolation and characterization of intermediate particles of the assembly pathway.

Three classes of particles have been identified in restrictive phi 29 suppressor-sensitive (sus) mutant infections of Bacillus subtilis, including DNA-containing heads or phage, prohead, and empty heads. Pulse-chase labeling experiments indicate that the prohead, the first particle assembled in 14-infected cells, is converted to DNA-filled heads and phi 29. In addition to the proteins Hd, P10, and F found in mature phi 29, the prohead contains a "core" protein P7 that exits as the prohead matures and appears to recycle during subsequent rounds of prohead assembly. Prohead-like structures accumulate in UV-irradiated cells and are present in restrictive infections with sus mutants of cistrons 9 and 16. Empty heads are observed only when infection results in the formation of DNA-containing particles; this and other evidence indicates that the empty heads are probably not true intermediates. Phage phi 29 assembly apparently occurs by a single pathway in which neck and tail components interact to stabilize the completed DNA-containing head.     

An ATP hydrolysis sensor in the DNA packaging motor from bacteriophage T4 suggests an inchworm-type translocation mechanism

Tailed bacteriophages and large eukaryotic viruses employ powerful molecular motors to translocate dsDNA into a preassembled capsid shell. The phage T4 motor is composed of a dodecameric portal and small and large terminase subunits assembled at the special head-tail connector vertex of the prohead. The motor pumps DNA through the portal channel, utilizing ATP hydrolysis energy provided by an ATPase present in the large terminase subunit. We report that the ATPase motors of terminases, helicases, translocating restriction enzymes, and protein translocases possess a common coupling motif (C-motif). Mutations in the phage T4 terminase C-motif lead to loss of stimulated ATPase and DNA translocation activities. Surprisingly, the mutants can catalyze at least one ATP hydrolysis event but are unable to turn over and reset the motor. This is the first report of a catalytic block in translocating ATPase motor after ATP hydrolysis occurred. We suggest that the C-motif is an ATP hydrolysis sensor, linking product release to mechanical motion. A novel terminase-driven mechanism is proposed for translocation of dsDNA in viruses.     

DNA packaging motor assembly intermediate of bacteriophage phi29

Unraveling the structure and assembly of the DNA packaging ATPases of the tailed double-stranded DNA bacteriophages is integral to understanding the mechanism of DNA translocation. Here, the bacteriophage phi29 packaging ATPase gene product 16 (gp16) was overexpressed in soluble form in Bacillus subtilis (pSAC), purified to near homogeneity, and assembled to the phi29 precursor capsid (prohead) to produce a packaging motor intermediate that was fully active in in vitro DNA packaging. The formation of higher oligomers of the gp16 from monomers was concentration dependent and was characterized by analytical ultracentrifugation, gel filtration, and electron microscopy. The binding of multiple copies of gp16 to the prohead was dependent on the presence of an oligomer of 174- or 120-base prohead RNA (pRNA) fixed to the head-tail connector at the unique portal vertex of the prohead. The use of mutant pRNAs demonstrated that gp16 bound specifically to the A-helix of pRNA, and ribonuclease footprinting of gp16 on pRNA showed that gp16 protected the CC residues of the CCA bulge (residues 18-20) of the A-helix. The binding of gp16 to the prohead/pRNA to constitute the complete and active packaging motor was confirmed by cryo-electron microscopy three-dimensional reconstruction of the prohead/pRNA/gp16 complex. The complex was capable of supercoiling DNA-gp3 as observed previously for gp16 alone; therefore, the binding of gp16 to the prohead, rather than first to DNA-gp3, represents an alternative packaging motor assembly pathway.     

Virus DNA packaging: the strategy used by phage λ

Phage λ, like a number of other large DNA bacterio-phages and the herpesviruses, produces concatemeric DNA during DNA replication. The concatemeric DNA is processed to produce unit-length, virion DNA by cutting at specific sites along the concatemer. DNA cutting is coordinated with DNA packaging, the process of translocation of the cut DNA into the preformed capsid precursor, the prohead. A key player in the λ DNA packaging process is the phage-encoded enzyme terminase, which is involved in (i) recognition of the concatemeric λ DNA; (ii) initiation of packaging, which includes the introduction of staggered nicks at cosN to generate the cohesive ends of virion DNA and the binding of the prohead; (iii) DNA packaging, possibly including the ATP-driven DNA translocation; and (iv) following translocation, the cutting of the terminal cosN lo complete DNA packaging. To one side of cosN is the site cosB, which plays a role in the initiation of packaging; along with ATP, cosB stimulates the efficiency and adds fidelity to the endo-nuclease activity of terminase in cutting cosN. cosB is essential for the formation of a post-cleavage complex with terminase, complex I, that binds the prohead, forming a ternary assembly, complex II. Terminase interacts with cosN through its large subunit, gpA, and the small terminase subunit, gpNul, interacts with cosB. Packaging follows complex II formation. cosN is flanked on the other side by the site cosQ, which is needed for termination, but not initiation, of DNA packaging. cosQ is required for cutting of the second cosN, i.e. the cosN at which termination occurs. DNA packaging in λ has aspects that differ from other λ DNA transactions. Unlike the site-specific recombination system of λ, for DNA packaging the initial site-specific protein assemblage gives way to a mobile, translocating complete, and unlike the DNA replication system of λ, the same protein machinery is used for both initiation and translocation during λ DNA packaging.     

Synthesis of a trans-Acting Inhibitor of DNA Maturation by Prohead Mutants of Phage λ

Bacteriophage lambda with mutations in genes that control prohead assembly and other head precursors cannot mature their DNA. In this paper we present evidence that the failure of these phage mutants to mature DNA is a reflection of a mechanism that modulates terminase nicking activity during normal phage development. We have constructed plasmids that contain the lambda-cohesive end site (cos) and the genes that code for DNA terminase, the enzyme that matures DNA by cutting at cos. The DNA terminase genes are under control of a thermosensitive cI repressor. These plasmids lack most of the genes involved in prohead morphogenesis and other head precursors. However, when repression is lifted by destruction of the thermosensitive repressor, the terminase synthesized is able to cut almost 100% of the plasmids. Therefore, these plasmids can mature in the absence of proheads and other head gene products. The plasmids are also able to complement mutants of lambda deficient in terminase and DNA maturation. However, in these complementation experiments, if the phage carry mutations in prohead genes E or B, not only is phage DNA maturation blocked, but the plasmid also fails to mature. These experiments show that, in the absence of proheads, phage lambda produces a trans-acting inhibitor of maturation. The genetic determinant of this inhibitor maps in a region extending from the middle of gene B to the end of gene C. A model is proposed in which the nicking activity of DNA-bound terminase is inhibited by the trans-acting inhibitor. Prohead (and other factors) binding to this complex would release the block to allow DNA cleavage and packaging.     

Structure of the bacteriophage lambda cohesive end site. Genetic analysis of the site (cosN) at which nicks are introduced by terminase.

A collection of mutations affecting the site (cosN) at which the bacteriophage lambda DNA packaging enzyme, terminase, introduces nicks to generate mature lambda chromosomes has been studied. A good correlation was found for mutational effects on burst size, accumulation of unused proheads, packaging of DNA into heads and cos cutting by terminase in vitro, indicating that defective cosN cleavage by terminase is the molecular explanation for the phenotypic effects of the mutations. Although the base-pairs of cosN display partial twofold rotational symmetry, cosN was found to be asymmetric functionally. Certain mutations to the left side of the center of rotational symmetry have more pronounced phenotypic effects than rotationally symmetric mutations to the right. The cosN11G mutation has no phenotypic effects when present as a single mutation, but does affect DNA packaging and cosN cutting in the presence of the symmetrically disposed cosN2C mutation. Mutations that decrease cosN cleavage result in the accumulation of unexpanded proheads, indicating that prohead expansion depends on cosN cutting.     

Maturation of phage T7 involves structural modification of both shell and inner core components

The double-stranded DNA bacteriophages are good model systems to understand basic biological processes such as the macromolecular interactions that take place during the virus assembly and maturation, or the behavior of molecular motors that function during the DNA packaging process. Using cryoelectron microscopy and single-particle methodology, we have determined the structures of two phage T7 assemblies produced during its morphogenetic process, the DNA-free prohead and the mature virion. The first structure reveals a complex assembly in the interior of the capsid, which involves the scaffolding, and the core complex, which plays an important role in DNA packaging and is located in one of the phage vertices. The reconstruction of the mature virion reveals important changes in the shell, now much larger and thinner, the disappearance of the scaffolding structure, and important rearrangements of the core complex, which now protrudes the shell and interacts with the tail. Some of these changes must originate by the pressure exerted by the DNA in the interior of the head.     

Bacteriophage T4D Head Morphogenesis. VIII. DNA-Protein Associations in Intermediate Head Structures That Accumulate in Gene 49? Mutant-Infected Cells

We have utilized the gene 49⁻ mutant-infected cells of bacteriophage T4D to accumulate large numbers of nucleic acid-protein intermediate head structures. These heads were used as substrates for experiments in the investigations of the mechanism of DNA packaging. Specifically, we have examined: (i) the susceptibility of the DNA in these structures to digestion by a variety of nucleases after a series of increasing temperature pulses from 25 to 100°C, (ii) the physicochemical characteristics of the DNA inside these heads, and (iii) the mechanism by which proteins are displaced from the interior of the head after treatment with basic proteins. We isolated DNA from these gene 49⁻ heads by use of gradient centrifugation procedures. The DNA had a molecular weight of 8 × 10⁶ and a density of 1.697 ± 0.005 g/cm³, and it contained a short resistant fraction (SRF) which, when associated with the gene 49⁻ heads, exhibited AT-protected regions that were not susceptible to micrococcal nuclease digestion. Such a fraction may contain pieces which are important in the initial association of the DNA with the prohead. Exposure of the gene 49⁻ intermediate capsid structures to basic proteins, such as bovine trypsin inhibitor, lysozyme, and l-polylysine-70, caused a displacement of an amorphous-appearing structure which may be a complex of the gene 49⁻ DNA and interior components of the capsid (e.g., internal proteins, polyamines). Our general conclusion is that in the gene 49⁻ intermediate head structures which are only partly filled with DNA, this DNA is held inside the head by strong electrostatic linkages with interior polypeptides and polyamines.     

Analysis of the coliphage T5 DNA ejection process with free and liposome-associated TonA protein.

Outer membrane protein TonA, the receptor for coliphage T5, has been partially purified and incorporated into the phospholipid bilayer of liposomes. Adsorption of the phage to its receptor in either a free or liposome-associated form is fast and sufficient to trigger the ejection of encapsidated DNA. In both in vitro systems the exit of DNA from the phage capsid is a very slow process. Ejected DNA can partially accumulate inside the liposome aqueous compartment, but the transfer from the phage head to the liposome internal space is never complete, perhaps because the liposome volume is too small. The presence of polyamines or divalent cations (magnesium) or both in the incubation medium diminished the extent of DNA ejection, possibly by stabilizing DNA inside the head. DNA movement was slowed as the temperature was decreased from 37 to 18 degrees C. Furthermore, incubation at 4 degrees C totally prevented this DNA movement, even if a large part of the DNA had already exited the capsid.