Supporting Online Material

Full Text
Structure of the Extracellular Region of HER3 Reveals an Interdomain Tether
Hyun-Soo Cho and Daniel J. Leahy

Supporting Online Material

Materials and Methods

Expression, Purification, and Crystallization of sHER3. A DNA fragment encoding amino acids 1-621 of human HER3 (S1) was amplified by PCR from a plasmid containing the HER3 cDNA provided by A. Ullrich, subcloned into the pSGHV0 expression vector (S2), and sequenced. 100 Greek Letter Mug of this plasmid was electroporated along with 5 Greek Letter Mug each of plasmids encoding dihydrofolate reductase and neomycin resistance into to lec1 CHO cells (S3). Transfected cells were selected in medium containing geneticin and screened for expression by ELISA. Expression levels were increased by selection in methotrexate until a cell line expressing 3.6 mg/l was isolated. The sHER3 fusion protein was purified from serum-free medium by Ni-NTA and gel filtration chromatography and cleaved by TEV protease to release sHER3, which was then treated with endoglycosidase H and re-chromatographed on a gel filtration column. sHER3 was concentrated to 4 mg/ml, dialyzed into dH2O, and crystallized in hanging drops with 15-20% PEG 4000, 0.1 M Tris pH 9.0, and 0.15 M lithium sulfate as the precipitant. Crystals grow in space group C2 with cell dimensions a=236.25, b=49.61, c=190.86 Å, and Greek Letter Beta=125.54°.

Determination and Refinement of the sHER3 Structure. sHER3 crystals were transferred to well buffer containing 30% PEG 4000 prior to flash freezing in liquid nitrogen. Derivatized crystals were prepared by soaking 1-2 days in well buffer containing 1 mM mercurochrome. sHER3 crystals exhibited significant nonisomorphism, and comparison of diffraction data from multiple crystals was required to identify a suitable derivative. Diffraction data from one native and one derivatized crystal were collected with an R-Axis4 detector and a Rigaku RU-300HR X-ray generator. Diffraction data from one native and two mercurochrome derivatized crystals were collected at beamline X4A at the National Synchrotron Light Source (NSLS) at Brookhaven National Laboratory. Diffraction data were collected at 2 and 3 wavelengths for the derivatized crystals, which exhibited dimished diffraction with prolonged exposure to X-rays (Table 1). Initial phases were calculated using SOLVE (S4), which produced maps that allowed identification of domain positions and a two-fold noncrystallographic symmetry axis inclined 37.6° from the crystallographic two-fold axis. Two-fold symmetry averaging combined with density modification using DM (S5) produced interpretable maps, and an initial model for the molecule in the asymmetric unit that proved better ordered was built using the program O (S6). The second model was generated using the noncrystallographic symmetry axis. Initial rigid-body and positional refinements were performed with CNS (S7) using the native data collected at NSLS beamline X4A; TLS refinement and REFMAC5 were used for later refinement stages to model anisotropy in the diffraction data (S8). TLS tensors were refined for each of the four domains in both molecules in the asymmetric unit. Average isotropic B-values for domains I, II, III, and IV were 12.77 (70.30), 11.96 (92.46), 16.17 (58.63), and 13.86 (84.24) Å2, respectively, for one molecule and 18.84 (46.32), 16.74 (49.25), 15.56 (64.21), and 20.84 (46.39) Å2for the other. Values in parentheses are the average isotropic B-values prior to TLS refinement. Surface areas were calculated using GRASP (S9). The final model consists of residues 7-13 and 36-610 of one sHER3 molecule and residues 8-11 and 36-579 of another, 85 water molecules, 1 sulfate ion, and 16 N-acetylglucosamine moieties. 9 of 11 stereochemical parameters of the sHER3 structure analyzed by the program PROCHECK (S10) are better than the average for structures determined at 2.6 Å resolution with the remaining 2 within the normal range. The percentage of residues in the most favored regions of the Ramachandran plot (79.0% vs. 74.8% for comparable structures). An electron density map is shown in fig. S1.

Biochemical crosslinking experiments show that sHER3 forms dimers at concentrations (1 mg/ml) where sHER2 does not (fig. S2). A noncrystallographic dimer is present that buries 3880 Å2 of surface area. Calculation of a surface complementarity parameter for the dimer interface yields a relatively low value of .56 (S11). Analytical ultracentrifugation and gel filtration fail to detect sHER3 dimers at comparable concentration indicating a dimerization constant of >20 Greek Letter MuM. Even this relatively weak sHER3 dimer may be present in the cell membrane (S12), however, and may play a role in regulating HER3 function.

Figure Legends

Supplemental Figure 1. sHER3 Electron Density. A portion of domain I of sHER3 is shown with a composite simulated annealing omit map calculated with 2Fo-Fc coefficients and contoured at 2Greek Letter Sigma. At the bottom of this figure a conserved array of asparagine residues that are buried within the hydrophobic core of the beta-helix is seen.

Medium version | Full size version

Supplemental Figure 2. Chemical Crosslinking of sHER3. A 1 mg/ml solution of sHER3 was incubated for 30 minutes in the presence (+) or absence (-) of 10 mM Bis(Sulfosuccinimidyl)Suberate (Pierce). Samples were then electrophoresed on a denaturing polyacrylamide gel and stained with Coomassie Brilliant Blue. sHER2 remains monomeric under similar conditions.

Medium version | Full size version


S1. M. H. Kraus, W. Issing, T. Miki, N. C. Popescu, S. A. Aaronson, Proc Natl Acad Sci U S A86, 9193-7. (1989).

S2. D. J. Leahy, C. E. Dann, P. Longo, B. Perman, K. X. Ramyar, Protein Expr Purif20, 500-6. (2000).

S3. P. Stanley, V. Caillibot, L. Siminovitch, Cell6, 121-8 (1975).

S4. T. C. Terwilliger, J. Berendzen, Acta Crystallogr D Biol Crystallogr55, 849-61 (1999).

S5. K. Zhang, P. Main, Acta CrystallographicaA46, 377-381 (1990).

S6. T. Jones, J.-Y. Zou, S. Cowan, M. Kjeldgaard, Acta CrystA47, 110-119 (1991).

S7. A. T. Brunger, et al., Acta Crystallogr D Biol Crystallogr54, 905-21 (1998).

S8. M. D. Winn, M. N. Isupov, G. N. Murshudov, Acta Crystallogr D Biol Crystallogr57, 122-33. (2001).

S9. A. Nicholls, K. A. Sharp, B. Honig, Proteins11, 281-96 (1991).

S10. Laskowski R A, M. M. W, M. D. S, T. J. M, J. Appl. Cryst.26, 283-291 (1993).

S11. M. C. Lawrence, P. M. Colman, J Mol Biol234, 946-50. (1993).

S12. B. Grasberger, A. P. Minton, C. DeLisi, H. Metzger, Proc Natl Acad Sci U S A83, 6258- 62. (1986).