The Skaggs Institute
for Chemical Biology

Scientific Report 2007

Structural and Functional Studies of the Severe Acute Respiratory Syndrome Coronavirus Proteome With Nuclear Magnetic Resonance Spectroscopy

K. Wüthrich, A. Chatterjee, M.A. Johnson, B. Pedrini, P. Serrano

In 2003, a major global outbreak of severe acute respiratory syndrome (SARS) was caused by the newly emerged SARS coronavirus (SARS-CoV). To date, no vaccine or effective treatment for SARS is available, and the reasons for the unusual severity of SARS relative to the milder infections caused by previously studied coronaviruses are unknown. To help establish a scientific basis for future rational development of SARS treatments, scientists at the Center for Functional and Structural Proteomics of SARS-CoV-Related Proteins (FSPS; http://visp.scripps.edu/SARS/default.aspx) are undertaking a systematic investigation of the structures and functions of the SARS-CoV proteome. Our participation in this project involves studies with nuclear magnetic resonance (NMR) spectroscopy.

The SARS-CoV genome is composed of a single plus-strand RNA molecule of about 30 kb. About two-thirds of the genome is devoted to encoding a replicase polyprotein. This single, long polypeptide chain is cleaved by viral proteases to release the mature nonstructural proteins, which then perform the enzymatic functions needed for viral replication. The genome also encodes structural proteins, which form part of the mature viral capsid.

Several of the replicase proteins of coronaviruses share only very low levels of sequence similarity with previously characterized proteins, and little is known about the functions of the replicase proteins during viral infection. Structural analysis of these proteins can therefore provide insight into biological function that may not be apparent when only the protein sequence is known.

Nonstructural Protein 3A

Nonstructural protein 3 (nsp3) of SARS-CoV is a large protein of about 1900 amino acids, within which 7 functional regions have been predicted. We are exploring individual domains of nsp3. In preliminary NMR screening, we identified several constructs containing globular protein folds, starting with nsp3a, which is the N-terminal domain of nsp3. We determined the solution structure and found that nsp3a adopts a ubiquitin-like α/β roll fold, with structural similarity to the Ras-associating family of ubiquitin-like proteins (Fig. 1). The structural similarity includes a conserved basic residue, arginine 23, which is located on the protein surface and is intimately involved in the interaction with Ras in some Ras-associating proteins. Thus, this protein may represent a viral factor that interferes with cell signaling pathways involving Ras, a situation that might cause alterations in the cellular environment that would favor viral replication.

Fig. 1. NMR solution structure of nsp3a(1–112). Left, The polypeptide backbone of a bundle of 20 energy-minimized CYANA conformers has been superimposed for minimal root-mean-square deviation of the backbone atoms of residues 20–108. The N-terminal segment of residues 1–19 is flexibly disordered. Right, Ribbon representation of one conformer from the bundle on the left. β-Strands are cyan; α-helices are red.

Further possible functions of this protein were suggested when we observed that RNA fragments copurify with nsp3a from Escherichia coli cell lysates. We characterized the RNA fragments by using mass spectrometry, NMR spectroscopy, and electrophoretic gel mobility shift assays and identified the short sequence adenosine-uridine-adenosine (AUA) as a key structural element. NMR chemical shift perturbation studies to identify the amino acid residues of nsp3a most strongly affected by RNA binding (Fig. 2) indicated that these residues correspond to 2 helical secondary structure elements that represent a feature unique to nsp3a, one that has not been observed in other ubiquitin-like proteins. It thus appears that nsp3a may have evolved a new function of the ubiquitin-like fold that involves RNA binding or RNA processing. The sequence AUA occurs several times in the 5′-untranslated region of the SARS-CoV genome, further indicating that this binding or processing activity may be related to subgenomic RNA synthesis, which is an essential part of CoV replication. Overall, these data provide a novel lead that perturbation of the nsp3a-RNA interaction may be a promising strategy for the design of therapeutic agents to treat SARS.

Fig. 2. Superposition of the [15N,1H]-correlation spectra of nsp3a(1–112) in the absence (blue) and presence (red) of a 4-fold excess of an exogenous single-stranded RNA sequence containing the trinucleotide segment AUA. The differences in corresponding peak positions indicate parts of the protein that are affected by the interaction with RNA and may form the RNA-binding site.

Nonstructural Protein 3C

The third domain from residues 366–722 of nsp3, termed nsp3c, is unique to SARS-CoV and does not occur in other known coronaviruses. One-dimensional ¹H NMR screening of constructs provided by the FSPS consortium showed that a construct spanning residues 451–651 contained a globular domain. We established purification protocols for this construct but found that it was chemically unstable, subject to proteolytic degradation after purification by nickel-affinity and size-exclusion chromatography. Edman degradation of the resulting 15.5-kD fragment indicated a high-frequency proteolytic cleavage site between residues 512 and 513. We cloned the fragment composed of residues 513–651 and found that it was amenable to NMR structure determination, with long-term stability of several weeks in aqueous solution. The solution structure revealed a mixed α/β fold with similarity to nucleic acid helicases and other nucleotide-binding proteins, suggesting a possible physiologic function in RNA unwinding during genome replication, a function that may be linked to those of other parts of nsp3, in particular, nsp3a.

Outlook

We are currently refining the analysis of the structure of nsp3c, including the spatial arrangement of potential active-site residues that may be responsible for nucleotide or RNA binding. NMR structure determinations of additional components of the SARS proteome are at various stages of completion. Combined with the data on other parts of the SARS-CoV proteome accrued within and outside of the FSPS, our research can be expected to contribute to a broad scientific foundation for the development of strategies for handling potential future infections caused by SARS-CoV.

Publications

Almeida, M.S., Johnson, M.A., Herrmann, T., Geralt, M., Wüthrich, K. Novel β-barrel fold in the nuclear magnetic resonance structure of the replicase nonstructural protein 1 from the severe acute respiratory syndrome coronavirus. J. Virol. 81:3151, 2007.

Chatterjee, A., Johnson, M.A., Serrano, P., Pedrini, B., Wüthrich, K. NMR assignment of the domain 513-651 from the SARS-CoV nonstructural protein nsp3. Biomol. NMR Assign. 1:191, 2007.

Serrano, P., Johnson, M.A., Almeida, M.S., Horst, R., Herrmann, T., Joseph, J.S., Neuman, B.W., Subramanian, V., Saikatendu, K.S., Buchmeier, M.J., Stevens, R.C., Kuhn, P., Wüthrich, K. Nuclear magnetic resonance structure of the N-terminal domain of nonstructural protein 3 from the severe acute respiratory syndrome coronavirus. J. Virol. 81:12049, 2007.