Introduction
Proteins in solution fluctuate between folded and unfolded conformations, which is often related to their functions and toxic aggregate formation.1-4 Outer surface protein A (OspA) is an immunogenic lipoprotein expressed by the spirocheteBorrelia (e.g. B. burgdorferi, B. garinii, and B. afzelii ), the etiologic agent of Lyme disease. It consists of 21 repeated β-strands and a short C-terminal α-helix. OspA also contains two globular domains, the N-terminal (β1–β7) and the C-terminal (β11–C-terminus) domains, as well as a single-layer β-sheet (central β-sheet, β8–β10) that connects the terminal domains.5,6
OspA binds to the tick receptor TROSPA when it enters the tick gut.7 Since the receptor-binding sites are occluded in the interior of the C-terminal domain,8 their exposure is believed to be required for receptor-binding. Koide et al employed native state hydrogen exchange nuclear magnetic resonance (NMR) spectroscopy to reveal that the C-terminal region, more specifically, the β9–C-terminus, including a portion of the single-layer β-sheet and the C-terminal domain, of this protein is less stable than the N-terminal domain. This study provided the first evidence of disordered intermediate conformations in the C-terminal regions of the protein.9-11 They also reported that the intermediate became stabilized as temperature increased.12,13 The temperature-stabilized intermediate was characterized using solution NMR,12 differential scanning calorimetry,12,13 and small angle X-ray scattering.11 Moreover, our group found that the pressure-stabilized intermediate was nearly fully disordered in the entire C-terminal region of the polypeptide chain, from β9 to the C-terminus. The formation of this intermediate was caused by exposure of a large internal cavity in the C-terminal domain.14 Recently, Makabe et al reported similar identity between a kinetic intermediate and the equilibrium intermediate.15 These results indicate that the pressure- and temperature-stabilized intermediates have similar structural characteristics to those existing under physiological conditions.
Transition into the intermediate in vivo may accelerate the binding of OspA to the tick receptor, as the intermediate exposes the receptor-binding sites (i.e. residues 236–237 and 242–244)8 , and can interact with a receptor distant from OspA.14 In addition, the intermediate may be important for efficient translocation of the protein through the outer membrane.16 More importantly, OspA and its C-terminal fragment have been included as vaccine candidates to prevent Borreliatransmission,17 some of which have been tested in Phase 3 clinical trials.18 A recombinant vaccine using B. burgdorferi OspA was previously licensed for Lyme disease (LYMErix, SmithKlineBeecham, Pittsburgh, PA, USA) in 1998, however, the manufacturer voluntarily withdrew the product from the market 3 years later.19 Although human vaccines for Lyme disease are not currently available, genetically modified OspA, which contain protective elements from two different OspA serotypes, continue to represent important vaccine candidates.18
Although previous studies have reported that the C-terminal half of the protein (β9–C-terminus) exhibits lower stability than the N-terminal half, our understanding of how the central β-sheet and C-terminal domain are disordered remains limited. Here, we further characterized the pressure- and temperature-stabilized intermediates of OspA using a paramagnetic relaxation enhancement (PRE)-assisted NMR.20-24 The PRE effect arises from dipole-dipole interactions between unpaired electrons of the paramagnetic agent and the nucleus of interest, and thus, spin relaxation contributions show r -6 dependence of distance between the paramagnetic center and nucleus. Accordingly, when the paramagnetic agent was conjugated to a portion of the central domain, we were able to collect information on the distance between the central β-sheet and the N- and C-terminal domains.
Materials and Methods
Sample preparation
A soluble form of OspA,25 consisting of residues 18–273 whereby the membrane-associated N-terminal region (residues 1–6) was truncated and the cysteine at residue 84 was substituted by serine, was used as the pseudo wild-type (C84S; WT*) form of OspA (BMRB Entry 4076). Uniformly 15N-labeled OspA WT* and three variants (D118C, E128C, and A140C) were produced by conventional E. coli expression in M9 medium with15NH4Cl as the sole nitrogen source. The protein was purified by a Ni2+-affinity column (Bio-Rad Laboratories, Hercules, CA) and a HiLoad 26/60 Superdex 75 prep grade column in the AKTA explore 10S (GE Healthcare Life Sciences, Pittsburgh, PA). To cleave the His-tag, the protease thrombin (GE Healthcare Life Sciences) was used. The cysteine variants were dissolved in 20 mM Tris-HCl buffer at pH 7.0, mixed withS -([1-oxyl-2,2,5,5-tetramethyl-2,5-dihydro-1H-pyrrol-3-yl]methyl) methanesulfonothioate (MTSL) (Toronto Research Chemicals, Ontario, Canada), and incubated for approximately 12 h in 20 mM Tris-HCl buffer at 277 K. The protein solution was filtered and concentrated using a Microcon (Merck Millipore, Billerica, MA, USA). The final protein solution was adjusted to a concentration of 0.3 mM for PRE experiments in 10 mM phosphate buffer (pH 5.9) containing 50 mM NaCl and 10%2H2O. The protein solution for temperature experiments contained 0.4 M guanidium chloride (GdmCl) in phosphate buffer. The spin-labeled MTSL was reduced with a 2-fold excess of ascorbic acid relative to the protein concentration.