The role of the C-domain of bacteriophage T4 gene 32 protein in ssDNA binding and dsDNA helix-destabilization: Kinetic, single-molecule, and cross-linking studies.

Kiran Pant,Matthew A Malinowski,Aye T Win,Richard L Karpel,Christopher Pabst,Brian Anderson,Mark C Williams,Hendrik Perdana,Yaakov Koby Levy

doi:10.1371/journal.pone.0194357

Kiran Pant, Matthew A Malinowski + Show 7 more

Open Access

https://doi.org/10.1371/journal.pone.0194357

Copy DOI

Abstract

The model single-stranded DNA binding protein of bacteriophage T4, gene 32 protein (gp32) has well-established roles in DNA replication, recombination, and repair. gp32 is a single-chain polypeptide consisting of three domains. Based on thermodynamics and kinetics measurements, we have proposed that gp32 can undergo a conformational change where the acidic C-terminal domain binds internally to or near the single-stranded (ss) DNA binding surface in the core (central) domain, blocking ssDNA interaction. To test this model, we have employed a variety of experimental approaches and gp32 variants to characterize this conformational change. Utilizing stopped-flow methods, the association kinetics of wild type and truncated forms of gp32 with ssDNA were measured. When the C-domain is present, the log-log plot of k vs. [NaCl] shows a positive slope, whereas when it is absent (*I protein), there is little rate change with salt concentration, as expected for this model.A gp32 variant lacking residues 292–296 within the C-domain, ΔPR201, displays kinetic properties intermediate between gp32 and *I. The single molecule force-induced DNA helix-destabilizing activitiesas well as the single- and double-stranded DNA affinities of ΔPR201 and gp32 truncated at residue 295 also fall between full-length protein and *I. Finally, chemical cross-linking of recombinant C-domain and gp32 lacking both N- and C-terminal domains is inhibited by increasing concentrations of a short single-stranded oligonucleotide, and the salt dependence of cross-linking mirrors that expected for the model. Taken together, these results provide the first evidence in support of this model that have been obtained through structural probes.

Highlights

Lohman and Kowalczykowski found that the association kinetics of full-length gene 32 protein with single-stranded DNAs were characterized by two first-order decays.[33]Preliminary experiments [33] indicated that the ÃI truncate has this property, and we observed biexponential decays with both full-length and ÃI protein
We have provided a variety of evidence for a conformational change involving the C-terminal domain of gene 32 protein when it binds to single-stranded DNA
Under low salt (“strong”) binding conditions, where the rate determining step of the association kinetics is the formation of a non-cooperatively bound protein-DNA complex, the removal of the C-domain greatly alters the salt dependence of the kinetics

Summary

Introduction

The C-domain creates a “kinetic block” to the protein’s thermodynamically-predicted ability to lower the thermal melting temperature (Tm) of natural double-stranded DNAs.[4, 8, 9, 21, 22] The temperature range of these experiments was necessarily limited by the temperature at which the protein denatures (50–55 ̊C). Single molecule DNA stretching experiments, where the mechanically-induced helix!coil transition is observed at temperatures well below the protein denaturation point,[23,24,25,26,27] have provided new insights into the nature of the kinetic barrier.[28,29,30,31,32] These studies indicate that the critical factor governing the helix-destabilizing activity is the magnitude of the protein’s (non-cooperative) affinity for double-strandedDNA, which is significantly higher for the ÃI truncate (protein minus the Cdomain) relative to intact protein

Methods

Results

Conclusion