Yennawar et al. 10.1073/pnas.0605979103. |
Supporting Table 1
Supporting Figure 5
Supporting Figure 6
Supporting Figure 7
Supporting Figure 8
Supporting Figure 9
Supporting Text
Supporting Figure 5
Fig. 5. Model of glycosyl residues attached to N10 of EXPB1. (A) Electron density map (2Fo - Fc difference) of the N-glycosylation at N10 contoured at 0.9 s level. N10 is colored green, and sugar residues are peach-colored. The figure was generated by using the program PYMOL. (B) Schematic of the linkages between the sugar residues.
Fig. 6.
Two-dimensional topology diagram of EXPB1 domain 1. Numbers indicate residue numbers; b -strands are numbered I-VI.Fig. 7.
Structure-based sequence alignment of EXPB1 and the catalytic domain of Humicola Cel45 (PDB ID code 4ENG), showing regions where the peptide backbones and secondary structures overlap. Regions where the peptide backbones are well aligned are indicated with capital letters. b-Strands and a-helices in the two structures are indicated by arrows and double lines.Fig. 8.
Comparison of wall extension activity of heat-inactivated wall samples upon addition (arrows) of 20 mg of EXPB1, 30 mg of Mytilus edulis MeCel45 (1, 2), 20 mg of Trichoderma TrCel45 (3), 50 mg of Trichoderma swollenin (4), or 37 mg of EXPA. (A) Wheat coleoptilies. (B) Cucumber hypocotyls. Arrows indicates time when protein was added. Shown are representative traces from three to five replicates. Methods are described in ref. 5.1. Xu B, Janson JC, Sellos D (2001) Eur J Biochem 268:3718-3727.
2. Xu B, Hellman U, Ersson B, Janson JC (2000) Eur J Biochem 267:4970-4977.
3. Saloheimo A, Henrissat B, Hoffren AM, Teleman O, Penttila M (1994) Mol Microbiol 13:219-228.
4. Saloheimo M, Paloheimo M, Hakola S, Pere J, Swanson B, Nyyssonen E, Bhatia A, Ward M, Penttila M (2002) Eur J Biochem 269:4202-4211.
5. Cosgrove DJ, Bedinger P, Durachko DM (1997) Proc Natl Acad Sci USA 94:6559-6564.
Fig. 9.
Two-dimensional topology diagram of D2 of EXPB1, showing topology of the b -strands, using the naming system of De Marino et al. (1).1. De Marino S, Morelli MA, Fraternali F, Tamborini E, Musco G, Vrtala S, Dolecek C, Arosio P, Valenta R, Pastore A (1999) Struct Folding Des 7:943-952.
Table 1. Refinement statistics and final model statistics for EXPB1
Resolution, Å | 2.75 |
Cell | |
a , Å | 113.408 |
b, Å | 44.512 |
c, Å | 69.467 |
b, ° | 124.64 |
Space group | C 2 |
Final Rfree, % | 29.07 |
Final R, % | 23.32 |
No. of reflections | 6,640 (working)/367(test) |
No. of atoms | 1,969 |
B(iso) 2 of protein atoms, Å | 55.5 |
Main chain | 53.4 |
Side chain | 57.6 |
rmsd in bond lengths, Å | 0.008 |
rmsd in bond angles, ° | 1.68 |
Solvent content, % | 53.9 |
Ramachandran plot, % | |
Most favored region, % | 91.9 |
Generously allowed, % | 6.6 |
Structure Solution and Refinement.
After several cycles of rigid body refinement the maps still looked noisy. To improve this, density modification was performed by using the program CNS (1). Solvent density modification and density truncation features were used. The resulting maps gradually helped in modeling regions of the missing N- terminal residues and the loop between residues 29 and 38. The side chains that were different in Phl p 1 compared to EXPB1 could also be corrected, and the four extra residues at the C terminus could be located. The polysaccharide covalently linked to Asn-10 was modeled as shown in Fig. 2. Only group B -factor refinement was used, given that the resolution of the data was 2.75 Å. After several iterations of modeling using the program O (2) and density modification and refinement using the program CNS, the R-factors converged at 23.32% and 29.07%, respectively. At the very end of the refinement a total of nine water molecules could be located. As reported by PROCHECK (3), all residues lie in or close to the allowed regions of the Ramachandran plot. The first three residues at the N terminus are disordered and are not part of the model.Comparison with Phl p 1 (PDB ID code 1N10).
Compared to the 2.9-Å structure of 1N10, which has a dimer in the asymmetric unit, EXPB1, with a monomer in the asymmetric unit, is solved to a better resolution (2.75 Å). The loop consisting of residues 29-38 is not resolved in 1N10 but has good electron density in EXPB1. This is an important loop because it contains D37, a potential candidate for the catalytic base. The first 15 residues at the N-terminal extension are oriented entirely differently in the two structures (leading to successful molecular replacement when omitted). The N-terminal strand in 1N10 extends out away from the protein and interacts with a second monomer. Because the recombinantly produced Phl p 1 used to solve the 1N10 structure was not native protein, the processing of the N-terminal extension appears to be atypical (the hydroxylation of prolines is lacking, and glycosylation pattern at N10 is probably different and was resolved to only one GlcNac).When the Ca carbon atoms of both D1 and D2 are superimposed for the two proteins, the rmsd is 1.84 Å. Superposition of the D1s alone (excluding the first 15 residues at the N terminus) reveals a good overlap (rmsd of 0.88 Å) whereas the D2s overlap poorly (rmsd of 1.82 Å). Comparison of the overlapping structures shows that the Ca of W194 (D2), which is a crucial part of the putative binding groove, is displaced by 4 Å in 1N10 and its side chain is rotated and displaced by almost 12 Å. However, the tryptophan ring continues to stay in the same plane as the other residues at the base of the groove, and hence a sugar could still bind in a fashion similar to that for EXPB1.
1. Brunger AT, Adams PD, Clore GM, DeLano WL, Gros P, Grosse-Kunstleve RW, Jiang JS, Kuszewski J, Nilges M, Pannu NS, et al. (1998) Acta Crystallogr D 54:905-921.
2. Jones TA, Zou JY, Cowan SW, Kjeldgaard M (1991) Acta Crystallogr A 47:110-119.
3. Laskowski RA, Macarthur MW, Moss DS, Thornton JM (1993) J Appl Crystallogr 26:283-291.