Electrostatic Interactions Drive the Self-Assembly and the Transcription Activity of the Pseudomonas Phage ϕ6 Procapsid

Abstract

Assembly of an empty procapsid is a crucial step in the formation of many complex viruses. Here, we used the self-assembly system of the double-stranded RNA bacteriophage ϕ6 to study the role of electrostatic interactions in a scaffolding-independent procapsid assembly pathway. We demonstrate that ϕ6 procapsid assembly is sensitive to salt at both the nucleation and postnucleation steps. Furthermore, we observed that the salt sensitivity of ϕ6 procapsid-directed transcription is reversible.

TEXT

Electrostatic interactions are essential for the scaffolding protein-assisted procapsid (PC) assembly of double-stranded DNA (dsDNA) bacteriophages and allow subsequent scaffolding release during PC maturation (1, 2). Such assembly pathways are characteristically sensitive to high salinity (2–6). By contrast, the external scaffolding protein-mediated assembly of icosahedral single-stranded DNA (ssDNA) phages is insensitive to ionic conditions (7).

Pseudomonas phage ϕ6 utilizes a scaffolding-independent assembly pathway in which all protein components involved in PC assembly are also present in the mature virions. The empty PC packages the genomic ssRNA precursors and converts them into genomic dsRNAs (8, 9), a process associated with capsid expansion (10–12). The PC shell, composed of 120 copies of protein P1, harbors 20 potential binding sites for the polymerase P2 monomers, 12 for the packaging nucleoside triphosphatase P4 hexamers, and 20 for the assembly factor P7 trimers (10, 13). However, these sites are not fully occupied in the virions (13). P2 and P7 are located in the interior of the PC shell (14–16), while P4 resides on its surface (Fig. 1A) and is important for the nucleation of PC assembly (Fig. 1A) and its maturation, genomic packaging, and transcription (17–21).

FIG 1.

Assembly pathway for ϕ6 PCs and the electrostatic potential distributions in the main capsid proteins. (A) The ϕ6 PC is assembled from the major capsid proteins P1 (blue), the polymerase P2 (pink), the packaging NTPase P4 (yellow), and the assembly factor P7 (gray) (step I). The nucleation of the PC assembly is mediated by the P4 hexamer to form complexes composed of P1, P4, and P7 (step II). After nucleation, the particle assembly is rapidly completed via addition of the remaining building blocks (steps III and IV). P2 is incorporated into the particle during the late stage of assembly. Genome packaging into the compact empty PC (cross-section image in step IV) induces its expansion (surface presentation in step V). The PC shell, arranged on a T = 1 icosahedral lattice, is composed of 120 copies of P1 (step V). The two conformers of P1 are shown in light and dark blue. P4 hexamers are located on the PC surface, while P2 and P7 reside in its interior (step IV). (B) Interacting surfaces of the P4 hexamer and P1 pentamer. Shown are the electrostatic potential distribution in the P4 hexamer (PDB 4BLO) and the P1 pentamer (PDB 4K7H) (left) and a schematic presentation of the P1–P4 interaction at a five-fold vertex of the ϕ6 PC (right). The surfaces shown (left) are facing each other in the PC. The surfaces and electrostatic potentials were generated by using the UCSF Chimera 1.8.1 program (35). Positive, neutral, and negative potentials are shown in blue, white, and red, respectively. The potentials range from +256 mV (blue) to −256 mV (red).

High-resolution structures of P2 and P7 (22, 23), and more recently ϕ6 P1 and P4 (24, 25), are available. The interacting surfaces of the P1 pentamer and the P4 hexamer display complementary surface charge distributions (Fig. 1B), indicating that electrostatic interactions may be important in PC formation.

Electrostatic interactions drive ϕ6 PC self-assembly.

In order to evaluate the potential role of electrostatic interactions during ϕ6 PC self-assembly, purified P1, P2, P4, and P7 (18, 26–28) were combined in a molar ratio of 120:12:72:45 (0.1 mg/ml P1) under ϕ6 self-assembly conditions (13, 18) with 80 to 750 mM NaCl. Particle assembly was monitored by following changes in light scattering (Fig. 2A). The reaction products (PCs) were separated from unassembled proteins by rate zonal centrifugation in the presence of 150 mM NaCl, unless otherwise stated (Fig. 2B), and the protein contents of the gradient fractions were analyzed by SDS-polyacrylamide gel electrophoresis (SDS-PAGE). At 80 to 200 mM NaCl (Fig. 2A), the kinetics curves displayed an S shape, which is typical for nucleation-limited assembly (18, 29). At higher salinity levels, the assembly rate became progressively slower (Fig. 2A). This was also reflected in the intensity of the PC zones in sucrose gradients (Fig. 2B) and the protein quantities of the corresponding gradient fractions (Fig. 2C).

FIG 2.

The effect of ionic strength on ϕ6 PC self-assembly. The in vitro assembly reactions for ϕ6 PC were carried out in 80 to 750 mM NaCl, as indicated. (A) Increase in light scattering during the assembly reaction. (B) The reaction products were analyzed by using rate zonal centrifugation in a linear 10% to 30% (wt/vol) sucrose gradient. (C) The top and PC fractions were collected and analyzed by SDS-PAGE. Recombinant PCs [PC(R)] from an Escherichia coli expression system (9), and purified ϕ6 virions were added as protein markers. The PC proteins are indicated on the left. (D) The calculated relative copy numbers of P2, P4, and P7, normalized to P1, in the in vitro-assembled PC. The values were calculated based on three repetitions.

Most proteins remained unassembled at high NaCl concentrations (Fig. 2C, top fractions), while at 80 mM NaCl practically all proteins were assembled into PCs (Fig. 2C, PC zones). Approximately 80% of P1 was recovered as soluble PCs at 75 to 100 mM NaCl, based on results of 15 independent assembly reactions (data not shown). Lower salinity could not be achieved due to unavoidable carryover of NaCl from the protein preparations.

The observed salt effects (Fig. 2) imply that electrostatic interactions between capsid proteins are the major driving force in ϕ6 PC assembly. At low ionic strengths, the assembly reaction rapidly went to completion and the subunits were efficiently utilized. However, at elevated salinity levels, the potential complementary electrostatic protein surfaces of the interacting subunits, as depicted for the P1 pentamer and P4 hexamer in Fig. 1B, were masked, reducing the assembly rate and particle yields. It is likely that the observed effects (Fig. 2A to C) mainly reflect changes in P1–P4 interactions, as P1 and P4 are the minimal components needed for the nucleation and assembly of ϕ6 PC (18) and the only ones that significantly contribute to the particle yields and assembly kinetics (13). P7 is an assembly factor that stabilizes the initiation complex and increases the rate of assembly (18, 30). Consequently, the observed decrease in the initial assembly rate under elevated salinity (Fig. 2A) may also arise, at least partially, from impaired P1–P7 interactions. However, it is unlikely that the observed effects (Fig. 2A to C) reflect changes in P2 interactions with PC components, as P2 does not have any influence on assembly kinetics (13, 18).

Binding of minor proteins to the PC shell is compromised at elevated salinity levels.

The relative numbers of proteins in in vitro-assembled PCs were estimated from SDS-polyacrylamide gels as described previously (13). The P1:P2:P4:P7 ratio of the PCs that self-assembled at 80 mM NaCl (Fig. 2D) was similar to that observed with the virion (13). However, a clear decrease was observed in the relative amounts of P2, P4, and P7 when salinity was increased (Fig. 2D). These observations imply that electrostatic interactions also have a role during postnucleation processes, when P2 associates with P1 or P7, before closing of the P1 shell (Fig. 1A), and support the notion that the strength of interactions between P1 and P7 is likely reduced under elevated salinity.

Previous studies suggested that P4 hexamers are initially incorporated at each of the vertices of the P1 shell but can be subsequently dissociated from the surface of the particle (13). To further probe the interactions between the P4 hexamer and the P1 shell, P4 assembly on preformed, P4-deficient PCs was analyzed as described previously (21). The P4-deficient particles harbor approximately one P4 hexamer and display an expanded conformation and a slower sedimentation rate than the compact wild-type PC (19, 21). At 100 mM NaCl, approximately 7 P4 hexamers were assembled on the P4-deficient particles, and the particle sedimentation behavior resembled that of the wild type (Fig. 3A and B). Two distinct light-scattering zones were detected when the NaCl concentration was doubled, and the recovered particles contained approximately three P4 hexamers (Fig. 3A and B); No P4 reconstitution was observed at higher salinities. These results confirm that electrostatic interactions between P4 and P1 are important not only for the nucleation of assembly (Fig. 2) but also at later stages of assembly and maturation.

FIG 3.

The effect of ionic strength on P4 reconstitution reactions. The P4 assembly reaction on P4-deficient particles was carried out at 60 to 400 mM NaCl. P4-deficient particles without P4 or NaCl addition were used as a negative control. There was a carryover of 60 mM NaCl from the particle preparation mixture. (A) The reaction products were analyzed by rate zonal centrifugation in a linear 15% to 40% (wt/vol) sucrose gradient. PC(f) refers to particles that had a similar sedimentation velocity as the wild-type PC. PC(s) refers to particles that sedimented more slowly, similarly to the expanded P4-deficient particles (first lane). (B) The PC zones were collected and analyzed by SDS-PAGE. Recombinant PCs [PC(R)] from an E. coli expression system (9) and purified ϕ6 virions were used as protein markers. The PC proteins are indicated on the left, and the average copy numbers for P4 hexamers, relative to P1, are indicated at the bottom of the gel.

The transcription activity of the ϕ6 PC is sensitive to elevated salinity.

Combined in vitro ssRNA packaging, replication, and transcription reactions (31) (Fig. 4A) with self-assembled or recombinant PCs (9, 18) and genomic ssRNA molecules produced by T7 transcription (32) were performed to evaluate the effects of salinity on the enzymatic activities of the PCs (Fig. 4B and C). Optimized reaction conditions including 80 mM ammonium acetate were applied (31, 33). A clear decrease in transcription activity was detected at 75 mM NaCl (Fig. 4B; total salinity of 155 mM). However, the transcription activity was rescued when the reaction mixture was diluted and allowed to continue at a lower salinity (Fig. 4C). P4 is essential for the ϕ6 PC-based transcription reaction, and the addition of P4 hexamers to the transcription reaction mixture results in their association on the P4-deficient particles with a concomitant rescue of the transcription activity (21). Consequently, the observed reduction in transcription activity (Fig. 4B) could arise from the reduced affinity between P4 and the PC shell at the elevated ionic strength and the rescue of the transcription activity after dilution of the salt (Fig. 4C), via the reassociation of P4s on the particle. Although the interactions between P1 and P4 in the fully expanded genome containing particles catalyzing the transcription reaction are likely to be slightly different from those observed with the empty PC (Fig. 4A), the decrease in the transcription activity (Fig. 4B and C) and the reduction in P4 incorporation onto the P4-deficient particles (Fig. 3) were observed at similar total salinity levels.

FIG 4.

The effect of ionic strength on the transcription activity of ϕ6 PCs. (A) A schematic presentation of the packaging, replication, and transcription reactions catalyzed by the ϕ6 PC. The preassembled empty PC packages the three genomic ssRNA precursor s, m, and l segments, replicate them to dsRNA (minus-strand synthesis), and subsequently produce transcripts from the M and S segments (plus-strand synthesis). The compact PC shell undergoes progressive expansion to a spherical appearance during genome encapsidation and minus-strand synthesis. (B) Relative transcription activities of the in vitro-assembled PCs in 0 to 75 mM NaCl, based on the quantitation of the labeled ssRNAs after separation of the reaction products by agarose gel electrophoresis. The NaCl concentration in the sucrose gradients used for the particle isolation was twice as high as that used in the transcription reaction mixture. The data represent two independent experiments. The values were normalized to the particle amounts and by setting the highest observed value to 1.0. (C) Agarose gel analysis of transcription reactions catalyzed by recombinant PCs at 75 or 100 mM NaCl for 90 min followed by an additional 90-min incubation at lower ionic strength, as indicated. A control reaction was carried out in 6 mM NaCl for 90 min (first lane). The positions of double-stranded (L, M, and S) and single-stranded (m and s) RNA molecules are indicated on the left.

Conclusions.

Electrostatic interactions are important for the efficient and precise assembly of the ϕ6 PC and for its enzymatic activities. The optimal salinity conditions for ϕ6 assembly are similar to those described for the dsDNA bacteriophage P22 (3, 4), based on a scaffolding-mediated assembly pathway, and reflect the conditions of the host cytoplasm (34), indicating a strong adaptation to in vivo conditions.