Supplementary Material for:

VISUALIZATION OF AN UNSTABLE COILED COIL FROM THE SCALLOP MYOSIN ROD

Yu Li*†‡, Jerry H. Brown*, Ludmilla Reshetnikova*, Antal Blazsek§, László Farkas§, László Nyitray§, and Carolyn Cohen*¶

*Rosenstiel Basic Medical Sciences Research Center and †Biophysics and Structural Biology Program, Brandeis University, Waltham, MA 02454-9110; §Department of Biochemistry, Eötvös Loránd University, Pázmány P. s. 1/c, 1117 Budapest, Hungary ‡Current address: Boston Biomedical Research Institute, 64 Grove Street, Watertown, MA 02472 ¶Corresponding author Email: . Phone: 781-736-2446, FAX: 781-736-2419

Supplemental Methods

Expression and purification

cDNA fragments encoding residues 250-281 of the GCN4 transcription factor (Saccharomyces cerevisiae) and residues 835- 885 of scallop myosin heavy chain (Argopecten irradians) were amplified by PCR from yeast genomic DNA, and from a previously described heavy chain clone1, respectively. The primers used for PCR were 5'-cggatccatgaaacaacttgaagacaag-3', 5'-acgagatcttagcgttcgccaactaatttc-3' (for the GCN4 leucine zipper), 5'-ggaattccatatgcctctcctcagcatcg-3' and 5'gcggatccgaagaggtcattcttc- 3' (for the myosin fragment). The two fragments were sequentially cloned into the NdeI and BamHI site of the expression vector pET15b. The chimeric construct, which contains an N-terminal hexahistidine tag, was expressed in the E. coli strain BL21(DE3)pLysS and purified on nickel-charged agarose resin (ProBond, Invitrogen) under denaturing conditions. The His-tag was removed by thrombin digestion in 150mM NaCl/ 1mM EDTA/ 3.5 mM CaCl2/ 50mM Tris-HCl, pH 8.0. The final sequence of "S2N51/GCN4" is GSHMPLLSIARQEEEMKEQLKQMDKMKEDLAKTERIKKELEEQNVTLLEQKNDLFGSMKQLEDKVEELLSKNYHLENEVARLKKLVGER. Starting from the N-terminus, this chimeric peptide consists of: Gly-Ser-His-Met, the scallop myosin S2 segment, a Gly-Ser linker, and the GCN4 leucine zipper (Fig. 1a). The gly-ser linker allows the myosin and GCN4 sequences to be in phase with respect to their heptad repeats. Before crystallization, S2N51/GCN4 was loaded on a MonoQ ion-exchange column, (in 20 mM Tris-HCl, pH 7.5, with a linear NaCl gradient from 20 to 200 mM) precipitated with ammonium sulfate (75% saturation), further purified on a Superdex 200 10/30 (Pharmacia) gel-filtration column, in 40 mM NaCl/ 2 mM NaN3/ 30 mM MOPS, pH 7.2, and then concentrated to 10mg/ml.

Crystallization and Data Collection

Two microliter (l) of precipitant solution [30-35% polyethylene glycol 2000 monomethyl ether (PEG 2000 MME, Fluka), and 5-10mM MgCl2] was added to 3 l of protein solution. The resultant mixture was equilibrated against a reservoir solution (15-20% PEG 2000 MME / 50 mM NaCl / 2 mM NaN3 / 44 mM MOPS, pH 6.9) by vapor diffusion at 16°C. From the spherulites that resulted, microseeds were introduced into new drops as above (but without MgCl2), and this process was then repeated with macroseeds. The orthorhombic crystals (P212121, a = 54.04 Å, b = 73.30 Å, c = 102.97 Å, 49% solvent) were harvested after 2 weeks and cryoprotected with 22.5% PEG 2000 MME / 25% glycerol / 50 mM NaCl / 2 mM NaN3 / 44 mM MOPS, pH 6.9. Data from a single crystal were collected at 100 K to 2.5-Å resolution with synchrotron radiation at the Cornell High Energy Synchrotron Source (beamline A1,  = 0.939 Å). Data were reduced with the software package HKL. (See Table 1 for data collection statistics.)

Structure Determination

We first attempted to solve the structure of the S2N51/GCN4 peptide by using the 32-residue-long high-resolution structure of GCN4 leucine zipper (Protein Data Bank ID code 2zta) as a search model for molecular replacement. Using AMoRe2, the solution with the best statistics (correlation coefficient = 0.281, Rcryst = 0.566, with data from 9.0 to 3.3 Å) corresponded to a model that turned out to have the correct orientation but an incorrect translational position. Each of the two chains of this model was then extended at the N-terminus by a 10-residue helical segment using the program O3). A rigid body refinement was then performed on the extended portion of the molecule, using CNS_SOLVE4 . This procedures was repeated 3 more times. After several rounds of manual rebuilding in O and of refinement (minimization and simulated annealing in CNS_SOLVE, with data to 2.5 Å), the crystallographic R factor fell to 0.446 (Rfree = 0.513). The agreement between the calculated and observed data, however, could not be improved using this solution. Using this partially “refined” 72-residue-long structure as a search model, we then performed molecular replacement again with AMoRe, and now found two molecules in the asymmetric unit (correlation coefficient = 0.544, Rcryst = 0.478 with data from 9.0 to 3.3 Å). This solution proved to be the correct one: after several additional rounds of refinement and rebuilding with data to 2.5 Å resolution, the final Rfree = 0.286 and the geometry of the structure is excellent (Table 1). Note that the N-terminal 12-15 residues of each chain are disordered and not included in the final model. Determination of -helical coiled coil structures by molecular replacement is often very difficult because of the elongated shape and repetitive nature of this motif. The procedure used here involving conventional crystallographic refinement prior to the completion of molecular replacement is unorthodox, and further studies will be required to determine its applicability to other coiled-coil structures and perhaps to globular proteins as well.

Supplemental references

  1. Nyitray, L., Goodwin, E.B., Szent-Györgyi, A.G. Nucleotide sequence of full-length cDNA for a scallop striated muscle myosin heavy chain. Nucleic Acids Res. 18, 7158 (1990).
  2. Navaza, J. & Saludjian, P. AMoRe: an automated molecular replacement program package. Methods Enzymol. 276, 581-594 (1997).
  3. Jones, T. A., Zou, L. Y., Cowan, S. W.& Kjeldgaard, M. Improved methods for building protein models in electron density maps and the location of errors in these models. Acta crystallogr. A 47, 110-119 (1991).
  4. Brünger, A. T., Adams, P. D., Clore, G. M., DeLano, W. L., Gros, P., Grosse-Kunstleve, R. W., Jiang, J.-S., Kuszewski, J., Nilges, M., Pannu, N. S., Read, R. J., Rice, L. M., Simonson, T. & Warren, G. L. Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr. D 54, 905-921 (1998).

Supplemental Table 1. Data collection and refinement statistics

Data Collection(∞2.5-Å resolution)

Measured reflections62,988

Unique reflections12,902

Completeness, %90.4 (75.7)*

Rmerge†0.062 (0.203)*

Refinement ¶(∞2.5-Å resolution)

No. of reflections12,880

Sigma cutoffnone

Rcryst‡/ Rfree§0.243/ 0.286

No. of protein/ water atoms2,499/ 83

rms bond lengths, Å/ angles, 0.006/1.012

______

*Values for the highest-resolution bin (2.50-2.60 Å) are shown in parentheses.

†Rmerge = hkl|I - <I>|/Ihkl.

‡Rcryst = hkl||Fobs| - |Fcalc||/hkl|Fobs|, where Fcalc and Fobs are calculated and observed structure factor amplitudes, respectively.

§Rfree is the same as Rcryst except that the summation is over 912 reflections randomly selected and excluded from the refinement.

¶ The main-chain Phi-Psi dihedral angles of all residues lie in the most-favored regions of the Ramachandran plot.

1