Herpes simplex virus type 1 helicase-primase: DNA binding and consequent protein oligomerization and primase activation

Abstract

The heterotrimeric helicase-primase complex of herpes simplex virus type I (HSV-1), consisting of UL5, UL8, and UL52, possesses 5' to 3' helicase, single-stranded DNA (ssDNA)-dependent ATPase, primase, and DNA binding activities. In this study we confirm that the UL5-UL8-UL52 complex has higher affinity for forked DNA than for ssDNA and fails to bind to fully annealed double-stranded DNA substrates. In addition, we show that a single-stranded overhang of greater than 6 nucleotides is required for efficient enzyme loading and unwinding. Electrophoretic mobility shift assays and surface plasmon resonance analysis provide additional quantitative information about how the UL5-UL8-UL52 complex associates with the replication fork. Although it has previously been reported that in the absence of DNA and nucleoside triphosphates the UL5-UL8-UL52 complex exists as a monomer in solution, we now present evidence that in the presence of forked DNA and AMP-PNP, higher-order complexes can form. Electrophoretic mobility shift assays reveal two discrete complexes with different mobilities only when helicase-primase is bound to DNA containing a single-stranded region, and surface plasmon resonance analysis confirms larger amounts of the complex bound to forked substrates than to single-overhang substrates. Furthermore, we show that primase activity exhibits a cooperative dependence on protein concentration while ATPase and helicase activities do not. Taken together, these data suggest that the primase activity of the helicase-primase requires formation of a dimer or higher-order structure while ATPase activity does not. Importantly, this provides a simple mechanism for generating a two-polymerase replisome at the replication fork.

FIG. 1.

Purified UL5-UL8-UL52 complex. The UL5-UL8-UL52 complex was purified from Sf9 insect cells that had been coinfected with recombinant baculoviruses encoding UL5, His-UL8, and UL52. The protein was purified on a HIS-Select nickel affinity column as described in Materials and Methods. Products were resolved by 10% SDS-PAGE subsequently stained with Coomassie blue. Molecular mass markers are shown in the left lane, and purified the helicase-primase complex is shown in the right lane.

FIG. 2.

A single-stranded region is required for efficient enzyme loading. (A) The sequences of the oligonucleotides used to prepare artificial substrates are shown. (B to G) DNA binding ability was examined by electrophoretic mobility shift assay (EMSA) between the helicase-primase complex (H/P) and ssDNA (B), fully annealed double-stranded DNA (C), forked DNA (D), a three-way junction substrate (E), a cruciform substrate (F), and a T-shaped substrate without any ssDNA region (G), respectively. A schematic of the substrate is indicated above each panel. The star (*) indicates the position of the ³²P label. Binding efficiency is shown below the lanes. Substrates were prepared as described in Materials and Methods by annealing different oligonucleotides, shown in panel A. Reaction mixtures contained 50 or 200 nM enzyme. A minus sign above the lanes indicates control reactions (no enzyme).

FIG. 3.

A single-stranded region is required for efficient unwinding. The unwinding activity of the helicase-primase complex (H/P)was assayed with different substrates: fully annealed double-stranded DNA (A), forked DNA (B), a three-way junction substrate (C), a cruciform substrate (D), or a T-shaped substrate without any ssDNA region (E). A schematic of the substrate is indicated above each panel. The asterisk (*) indicates the position of the ³²P label. Concentrations of the enzyme were 25, 50, 100, 200, and 400 nM. A minus sign above the lanes indicates control reactions (no enzyme).

FIG. 4.

The helicase-primase complex (HP)translocates along ssDNA primarily in a 5′ to 3′ direction. (A) Sequences of the substrates used. The labeled oligonucleotides are in italic. (B) Helicase assay using fork-15 and dsDNA. (C) Helicase assay using 5′-overhang substrates containing an extension of 3 nt, 6 nt, 9 nt, 12 nt, or 15 nt, as indicated above the lanes. (D) Helicase assay using 3′-overhang substrates containing an extension of 3 nt, 6 nt, 9 nt, 12 nt or 15 nt, as indicated above the lanes. Values for unwinding efficiency are shown below the lanes. A schematic of the substrate is shown above the panel. The asterisk (*) represents the location of the ³²P label. A minus sign above the lanes indicates control reactions (no enzyme). Reaction mixtures contained 200 nM protein.

FIG. 5.

Helicase-primase complex binding to forked and 5′- and 3′-overhang substrates. (A) Sequences of substrates used. Labeled nucleotides are in italics. The forked substrate contains a 30-nt duplex and two 30-nt single-stranded regions (underlined). The 5′- and 3′-overhang substrates contain a 30-nt duplex plus a 30-nt extension. (B) Electrophoretic mobility shift assays were performed with forked or overhang substrates. Reaction mixtures contained either 50 nM or 200 nM protein (H/P), as indicated above the lanes; control reactions are indicated as 0 nM.

FIG. 6.

HSV-1 helicase-primase complex binds to forked and 3′- and 5′-overhang substrates with comparable K_Ds as assessed by SPR. The DNA sequence and structure of substrates used for SPR were as shown in Fig. 5A. Sensorgrams for different concentrations of enzyme binding to various substrates are shown. Reactions were performed in the absence (top row) or presence (bottom row) of nonhydrolyzable ATP (AMP-PNP). Grey lines show experimental sensorgram data. Black lines show fitted curves obtained by global fitting using a 1:1 Langmuir binding model. A summary of the DNA binding parameters determined from the fitted curves is shown beneath each set of sensorgrams. k_a indicates the association rate, k_d indicates the dissociation rate, and K_D is calculated as k_d divided by k_a.

FIG. 7.

raPID plot of DNA binding parameters determined in Fig. 6. Association rates (k_a) are plotted on the x axis, and dissociation rates (k_d) are plotted on the y axis. Dissociation constants (K_D) are indicated on the diagonal axis. Open symbols indicate parameters measured in the absence of AMP-PNP. Closed symbols indicate parameters measured in the presence of AMP-PNP.

FIG. 8.

EMSA binding assay showing helicase-primase complex (H/P) bound to DNA at increasing protein concentrations. DNA binding of the H/P to the forked substrate described in the legend to Fig. 5 was examined by electrophoretic mobility shift assay at increasing protein concentrations from 0 to 200 nM.

FIG. 9.

Effects of varying the UL5-UL8-UL52 concentration on primase (A and C) and DNA-dependent ATPase (B and D) activity. Assays contained enzyme, 0.2 μM C₂₀GCC(C)₂₀, 500 μM [α-³²P]GTP (A), 1 mM [α-³²P]GTP (C), 500 μM [α-³²P]ATP (B), or 1 mM [α-³²P]ATP (D) and were performed as described in Materials and Methods. The line shows the fit with a K_D of 50 nM.

FIG. 10.

Model. (A) Scheme showing activities monomer or higher- order complexes of the UL5-UL8-UL52 heterotrimeric complex. Our model proposes that a single UL5-UL8-UL52 heterotrimeric complex lacks primase activity, but when UL5-UL8-UL52 forms higher-order complexes [(UL5-UL8-UL52)_n], these complexes exhibit primase activity. (B) Replication fork model. The HSV polymerase UL30 and its accessory protein UL42 are drawn as a spotted oval and a crescent, respectively. The UL5-UL8-UL52 helicase-primase ternary complex is depicted as a shaded oval for simplicity. The sawtooth line depicts the RNA primer, and the solid line depicts single-stranded DNA. In the top panel, low concentrations of the helicase-primase complex helicase activity result in unwinding of the duplex DNA but with little or no primase activity observed. At higher concentrations of helicase-primase complex (bottom panel), a dimer or higher-order multimer of the helicase-primase forms at the fork, and under these conditions, RNA primers are synthesized which can then be extended by HSV DNA Pol-UL42.