¹H, ¹³C, and ¹⁵N NMR Assignments for the Helicase Interaction Domain of Staphylococcus aureus DnaG Primase

Abstract

The interaction between DnaG primase and DnaB helicase is essential for stimulating primer synthesis during bacterial DNA replication. The interaction occurs between the N-terminal domain of helicase and the C-terminal domain of primase. Here we present the ¹H, ¹³C, and ¹⁵N backbone and side-chain resonance assignments for the C-terminal helicase interaction domain of Staphylococcus aureus primase.

Biological context

Bacterial primase (DnaG) is a conserved and essential enzyme responsible for the synthesis of short RNA polymers during DNA replication (Frick and Richardson, 2001). The protein is composed of three domains: N-terminal domain responsible for DNA sequence-specific binding; the catalytic core responsible for RNA synthesis; and the C-terminal domain (CTD) responsible for its interaction with DnaB helicase (Bird, et al., 2000; Tougu and Marians, 1996). The bacterial primase sequence is conserved along its entire length of approximately 570 residues (Frick and Richardson, 2001; Pan and Wigley, 2000; Syson, et al., 2005). Conservation is highest in the N-terminal domain, intermediate in the catalytic core, and lowest in the CTD.

The solution structures of DnaG primase CTD show significant variability between Geobacillus stearothermophilus (PDB 1Z8S) and Escherichia coli (PDB 2HAJ) (Su, et al., 2006; Syson, et al., 2005). Generally, the DnaG primase CTD structure is composed of two sub-domains, an N-terminal six-helix bundle (sub-domain C1) essential for DnaB binding activity and a helical hairpin (sub-domain C2) that modulates the interaction with DnaB (Oakley, et al., 2005; Syson, et al., 2005). The two CTD solution structures share significant structure similarity in their C1 sub-domains, but there are sharp differences in their C2 sub-domains (Oakley, et al., 2005; Su, et al., 2006; Syson, et al., 2005). In E. coli primase, the CTD structure is composed of 7 helices, where the long, rigid helix 6 connects the two sub-domains (Su, et al., 2006). In contrast, the G. stearothermophilus structure has two helices connecting the two sub-domains, where a kink occurs at Pro556 (Syson, et al., 2005).

The biological significance of these structural differences may provide an explanation for the observed divergence in primase at the level of bacterial phyla. For instance, Firmicutes DnaB helicases have been shown to stimulate a range of Firmicutes primases, but not Proteobacterial primases, and vice versa (Koepsell, et al., 2006; Larson, et al., 2010). This appears to be a general principle since we recently demonstrated that a wide range of bacterial enzymes show structural and functional divergence at the level of phyla (Shortridge, et al., 2011). We undertook the determination of the solution structure of Staphylococcus aureus primase CTD because it belongs to the same phylum as G. stearothermophilus. We expect the S. aureus primase CTD structure to determine whether the bacterial primase CTD structure is also conserved within a phylum. Here we present the complete backbone and side chain assignments for the DnaG primase CTD of S. aureus to provide the basis for determining its solution structure.

Methods and Experiments

Expression and Purification of S. aureus DnaG Primase CTD

The amplicon for S. aureus primase CTD gene (143 aa) was generated by polymerase chain reaction from the genome reference strain in the American Tissue Culture Collection by Nature Technology, Inc., Lincoln, Nebraska. It was inserted into a vector that added 20 N-terminal residues (MGHNHNHNHNHNHNGGDDDD) comprising an HN metal affinity tag and a protease-sensitive DDDD sequence. After transformation, the resulting E. coli overexpression strain had a doubling time of 15 min. Since this was a “leaky” expression system, this fast growth rate indicated the protein was not toxic and actually beneficial.

The protein used in the NMR structure determination experiments was uniformly labeled with ¹³C and ¹⁵N by growing the overexpression strain in minimal media containing ¹³C-glucose and ¹⁵N-ammonia. The expression level was extremely high with a final yield of 229 mg from a 2-L preparation. After the cells were collected by centrifugation, resuspended, treated with lysozyme, and then re-centrifuged, the target protein was found entirely in the supernatant. The protein was affinity purified on a nickel-NTA column and eluted with 100 mM imidazole. A denaturing polyacrylamide gel showed the purified protein was primarily a single band of the expected size (19.6 kDa). Its concentration was determined by absorbance using the extinction coefficient, ε₂₈₀, of 14,080 M⁻¹ cm⁻¹, derived from its 11 tyrosines.

Nuclear Magnetic Resonance (NMR) spectroscopy

Samples of [U-¹³C,¹⁵N] DnaG primase CTD for backbone resonance assignment by NMR spectroscopy were prepared at a protein concentration of 1.2 mM in 95% H₂O/5% D₂O solution containing 25 mM potassium phosphate, 100 mM NaCl, 50 mM glutamine and 50 mM arginine at pH 6.6 (uncorrected) in a sealed Shigemi tube (Shigemi Inc., Allison Park, PA). Samples of [U-¹³C,¹⁵N] DnaG primase CTD for side-chain resonance assignment by NMR spectroscopy were prepared at a protein concentration of 1.5 mM in 99% D₂O solution containing 25 mM potassium phosphate, 100 mM NaCl, 50 mM glutamine and 50 mM arginine at pH 6.6 (uncorrected) in a sealed Shigemi tube (Shigemi Inc., Allison Park, PA). All NMR experiments used for the protein backbone assignments of DnaG primase CTD were collected at 298 K on a five-channel 600 MHz Bruker Avance spectrometer equipped with a 5 mm TXI probe. NMR experiments used for the protein side chain resonances were collected at the Rocky Mountain Regional 900 MHz NMR Facility on a four channel 900 MHz Varian INOVA spectrometer equipped with a 5 mm HCN probe. Assignments of the backbone and side chain resonances were obtained from the following spectra: ¹⁵N-HSQC, ¹³C-HSQC, HNCO, HNCA, CBCACONH, CBCANH, HNHA, HBHACONH, CCCONH, HCCCONH, H(CCH)-COSY, 3D ¹³C HSQC-edited NOESY and 3D ¹⁵N HSQC-edited NOESY (Ferentz and Wagner, 2000). The NMR data was processed using NMRpipe (Delaglio, et al., 1995) and analyzed using PIPP (Garrett, et al., 1991) and CcpNmr v.2.1.5 (Vranken, et al., 2005).

Assignments of Primase CTD and Data Deposition

The backbone resonance assignment was 93% complete with 133 amino acids of the 143 unambiguously assigned in the 2D ¹H-¹⁵N HSQC spectrum (Figure 1). Unassigned residues include M1-H13, D19, E470, H479, L480, M481, T500, R536, E537, E543, P551, and Y552. Primase CTD residues are numbered relative to the complete DnaG primase sequence, but the N-terminal purification tag is simply numbered from residue 1 to 20. Residues M1-H13 and D19 comprise the majority of the unassigned residues and are all found within the N-terminal purification tag. The remaining unassigned residues were located within highly solvent-exposed regions. Residues H479, L480, and M481 were in a turn region between helix 1 and 2. Residue T500 was in an unstructured loop region between helix 2 and helix 3, whereas residues E543, P551 and Y552 were in an unstructured loop region between helix 5 and helix 6. The only residues that appear to be localized within secondary structure were R536 and E537. They are the second and third residues of helix 5. An exhaustive analysis of the NMR data set was unable to yield an assignment for these two residues, suggesting the end of the helix may undergo partial unfolding that leads to chemical shift exchange broadening.

Fig. 1.

1 2D ¹H–¹⁵N HSQC spectrum of uniformly ¹³C, ¹⁵N labeled S. aureus primase CTD at 1.2 mM in 95% H₂O/5% D₂O at pH 6.6 (uncorrected). The spectrum was collected on a Bruker Avance 600 MHz spectrometer equipped with a 5 mm TXI probe. The backbone amide resonances are assigned. The side-chain amide groups of Asn and Gln are connected by horizontal lines.

Aliphatic side chain carbon chemical shift assignments were completed using the CCCONH experiment that correlated the preceding (i-1) residue to the following (i) backbone amide chemical shift. Aliphatic side chain proton chemical shifts were completed with the HCCH-COSY and HCCCONH experiments. Aromatic side chain assignments were completed using the 3D ¹³C-edited NOESY experiment. The statistics for resonance assignment include, 139/163 HN, 139/201 N, 139/163 Cα, 134/168 Hα, 128/141 Cβ, 148/181 Hβ, 85/92 Cγ, 89/160 Hγ, 49/64, Cδ, 49/64 Hδ, 19/31 Cε, 19/31 Hε, 4/7 C_ζ, 4/7 H_ζ, and 132/143 CO. Figure 1 shows an annotated 2D ¹H–¹⁵N HSQC spectrum of the S. aureus helicase interaction domain of primase.

The S. aureus primase CTD secondary structure was predicted using the ¹³C chemical shifts and a consensus chemical shift index (CSI) (Wishart and Sykes, 1994). CSI predicts an all α-helical protein with 8 helices (Figure 2). Helical structures in primase CTD include Helix 1 Arg468-Lys478, Helix 2 Met481-Asn489, Helix 3 Lys505-Phe514, Helix 4 Ser524-Tyr529, Helix 5 Asn533-Glu543, Helix 6 Ile556-Asn564, Helix 7 Ile571-Arg583, Helix 8 Val587-Lys602. The C1 sub-domain of primase CTD includes Helix 1–5 and the C2 sub-domain includes Helix 6–8. This is consistent with the S. aureus DnaG primase CTD homology modeling predicted from the Geobacillus stearothermophilus structure (Syson, et al., 2005).

Fig. 2.

The ¹³Cα and ¹³Cβ secondary structure chemical shifts are plotted as a function of sequence. The S. aureus primase CTD secondary structure is predicted based on the consensus chemical shift index (CSI) using the ¹³C chemical shift data (Wishart and Sykes, 1994). The secondary structure is shown above the chemical shift index where α–helices (bars) are numbered.

Of particular interest are the residues between the predicted helices 6 and 7 (residues K566-T570). This region is significantly different in G. stearothermophilus primase CTD compared to the E. coli primase CTD solution structures (PDB 1Z8S and 2HAJ, respectively) (Su, et al., 2006). In G. stearothermophilus, this region is a loop forming two distinct subdomains (C1, C2) of primase CTD. In E. coli, the region is a long and rigid α helix. For S. aureus, the consensus chemical shift index suggests this region is similar to the G. stearothermophilus structure with an extended loop starting at residue Gly567 (figure 2). The ¹³Cα and ¹³Cβ chemical shifts for residues in this region have chemical shift values that are similar to those found in random coil structures.

The sequence specific ¹H, ¹³C, and ¹⁵N resonance assignments for S. aureus primase CTD were deposited in the BioMagResBank under accession number 17512.