Bosch et al. 10.1073/pnas.0605301104. |
Fig. 6. Comparison between x-ray structure and in silico model of the PfAldo:TRAP complex. The Ca-carbons of the two PfAldo structures were superimposed and the resulting position of the TRAP-tail is indicated in magenta for the x-ray structure and in cyan for the in silico model (1). The modeling studies and our crystallographic investigations were carried out independently. There is good general agreement between our x-ray structures (Fig. 1B) and the results of the modeling studies (1). The direction of the TRAP peptides are the same in the two studies with an rms deviation of 2.85 Å for 11 main chain atoms. A major difference concerns the sidechain of the key residue TRAP-W605, which is in a different orientation, although the Ca-positions are only 2 Å apart. The largest distance is 7.3 Å between the Cz atom positions of the indole ring. The modeling program failed to predict the induced fit creating the hydrophobic binding pocket next to the two arginine sidechains R48 and R309, which is clearly visible in both x-ray structures obtained.
Fig. 7. Stereo view of superimposed DWN TRAP-tail positions of the 2.4 Å (magenta) and 2.7 Å (green) Pfaldo:TRAP-tail structures (see SI Table 1) after superposition of the PfAldo Ca-backbone alone.
Fig. 8. Conformational changes of P. falciparum aldolase upon TRAP-tail binding. (A) Surface representation of the PfAldo:TRAP-tail complex in light blue with the P. berghei C-terminal tripeptide DWN of the TRAP-tail shown as sticks in magenta. (B) Surface representation of unliganded PfAldo (2) in the same orientation as in A. The surface of the unliganded PfAldo structure 1A5C (2) is shown in gray. Superimposed is the TRAP-tail as observed in our new ligand bound structure as sticks in magenta. The closure of the binding site upon dissociation of the TRAP-tail peptide is evident.
Fig. 9. Model of the last four C-terminal residues of MIC2 in complex with T. gondii aldolase, based upon our PfAldo:TRAP-tail structure. The backbone in orange represents the P. falciparum aldolase structure adapted to match the sequence of T. gondii by side chain mutations in the program Coot (3). We have extended the Ca-backbone by following the indicated direction of the C-terminal COOH from our structure and added the terminal E769 of MIC2. All residues of the MIC2 are in allowed regions of a Ramachandran plot as defined by Laskowski et al. 1993 (4). The side chain of M766 of MIC2 is pointing toward the solvent, W767 of MIC2 is involved in the identical interactions as observed for W605 of the P. berghei TRAP-tail (see Fig. 2A), M768 of MIC2 makes favorable hydrophobic interactions to R148 after some minor side chain shifts, and E769 of MIC2 is interacting with bulk solvent.
1. Buscaglia, C. A., Hol, W. G., Nussenzweig, V. & Cardozo, T. (2007) Proteins 66, 528-537.
2. Kim, H., Certa, U., Dobeli, H., Jakob, P. & Hol, W. G. J. (1998) Biochemistry 37, 4388-4396.
3. Emsley, P. & Cowtan, K. (2004) Acta Crystallogr D 60, 2126-2132.
4. Laskowski, R. A., MacArthur, M. W., Moss, D. S. & Thornton, J. M. (1993) J Appl Crystallog 26, 283-291.
Table 1. Crystallographic data and refinement statistics of the P. falciparum aldolase in complex with the C-terminal hexapeptide of P. berghei TRAP
Data set name | 1 | 2 |
Space group | P212121 | P212121 |
Unit cell parameters, Å | 70.39 145.52 148.52 | 70.02 146.16 148.96 |
Wavelength, Å | 0.9796 | 0.9796 |
Resolution range, Å | 19.97 - 2.40 | 19.97 - 2.7 |
Number of reflections (all) | 390,235 | 142,816 |
Number of reflections (unique) | 54,265 | 40,459 |
Completeness (%) | 94.74 (77.16) | 98.1 (91.6) |
Redundancy | 3.3 (2.2) | 3.3 (2.3) |
Mean I/s(I) | 11.7 (2.2) | 12.4 (2.8) |
R sym on I | 0.05 (0.22) | 0.07 (0.55) |
Model and refinement statistics |
|
|
Resolution range, Å | 19.97-2.40 | 19.97-2.7 |
Number of reflections (total) | 54,265 | 40,459 |
Number of reflections (test) | 2,909 | 2,144 |
Stereochemical parameters |
|
|
Restraints (rms observed) |
|
|
Bond length, Å | 0.009 | 0.006 |
Bond angles, ° | 1.186 | 0.936 |
No. of residues | 1,439 | 1,445 |
No. of atoms | 11,591 | 11,448 |
No. of water molecules | 519 | 341 |
R work, % | 20.1 | 20.8 |
R free, % | 25.0 | 26.9 |
Wilson B, Å2 | 55.3 | 65.2 |
Ramachandran plot |
|
|
Favored, % | 88.6 | 87.7 |
Allowed, % | 10.2 | 11.0 |
General allowed, % | 0.8 | 1.0 |
Disallowed, % | 0.3 | 0.3 |
Number in parentheses refers to the highest resolution shell.
SI Materials and Methods
Protein Expression and Purification.
Expression and purification of recombinant PfAldo in E. coli was performed according to a previously reported procedure (1).After breaking the cells in two rounds with a French press at 13,000 PSI, the lysate was clarified by centrifugation at 4°C with 30,000 ´ g. The supernatant was incubated for 1 h at 4°C with preequilibrated NiNTA resin. The protein was eluted with a 200 mM imidazole gradient and fractions were subsequently checked for purity via SDS/PAGE. Fractions containing recombinant PfAldo were dialyzed against buffer A (0.1 M TEA, 100 mM NaCl, 1 mM EDTA, 1 mM PMSF, 2 mM DTT pH 7.0). The overnight dialyzed fractions were applied to an ion exchange chromatography column and eluted with a gradient of buffer A supplemented with 1 M NaCl. The protein was concentrated by centrifugation, and further purified by size exclusion chromatography on a Superdex 200 column at 4°C with buffer A and 300 mM NaCl at pH 8.0. The molecular mass for PfAldo was determined through analytical size exclusion chromatography to be 160 kDa, which is consistent with a tetramer. Peak fractions from the analytical size exclusion chromatography column were then pooled and dialyzed overnight in 10 mM Tris•HCl, 1 mM DTT pH 7.5. A final yield of 10 mg/l was concentrated to 10 mg/ml, flash frozen (2), and stored at -80°C for crystallization experiments.
Data Collection and Structure Determination.
Diffraction data for the two PfAldo-P. berghei TRAP-tail complex structures were collected at the ALS 8.2.2 and SSRL 9-2 beamlines at l= 0.9796 with 0.75°-1° rotation images and 10- to 20-s exposure time depending on the mosaicity of the crystals as determined by MOSFLM (3). Crystals of PfAldo in complex with P. falciparum TRAP were of insufficient quality for data collection and not further investigated.Data reduction was carried out with Wedger ELVES (4) as a front end for MOSFLM/SCALA (3) from the CCP4 package (5). The crystals belonged to the orthorhombic space group P212121 with four subunits per asymmetric unit resulting in a solvent content of 49.0%. MOLREP (6) and the coordinates from the unliganded PfAldo structure [PDB code A5C (1)] yielded a molecular replacement solution with one tetramer per asymmetric unit and an initial R-factor of 41.3% for the 2.4 Å dataset. Manual rebuilding was carried out using the program Coot (7) with simulated annealing composite omit maps generated by CNS (8). In two PfAldo subunits the C-terminal tail is visible, although represented by weak density for some of the side chains, and involved in crystal contacts with neighbouring molecules. TLS refinement (9) was performed with REFMAC5 (10) using the TLS groups determined by the TLSMD webserver (11). The 2.4-Å PfAldo structure was refined to a crystallographic Rwork of 20.1% and Rfree of 25.0% (8). The rms deviations of the final model from ideal geometry are 0.009Å for bond lengths and 1.186° for bond angles (12). The TRAP-tail residues D604-N606 were refined to 70% occupancy with addition of partial occupied water molecules nearby the W605. The 2.7 Å PfAldo structure was subsequently solved by molecular replacement using the 2.4 Å coordinates and then refined to a crystallographic Rwork of 20.8% and Rfree of 26.9% (See SI Table 1). The final models were analyzed with validation tools in Coot (7) as well as MOLPROBITY (13) and SFCHECK (14).
PfAldo Mutation and TRAP Binding.
Mutations on PfAldo were introduced by PCR and checked by DNA sequencing. Glutathione S-transferase (GST)-tagged PfAldo variants were expressed in E. coli, purified, and quantified as described (15). The GST-tag was removed by in-column treatment with activated Factor X (New England Biolabs).The effect of the mutations on TRAP binding was assayed as described (16), with minor modifications. Briefly, polystyrene ELISA microplates (Nunc) were coated with 1 nmol per well of TRAP 25mer peptide (15) in carbonate buffer 50 mM pH 9.6, blocked with 200 ml of ELISA buffer (Imidazole acetate 10 mM pH7.3, 50 mM KCl, 0.2% Tween-20) supplemented with 3% BSA, and probed with dilutions of different PfAldo mutants in the same buffer. Binding was revealed by the addition of a rabbit antibody against the C-terminal peptide of P. berghei aldolase (16) followed by HRP-conjugated anti-rabbit IgG (Amersham Pharmacia Biosciences). After five washings, plates were revealed with the addition of 100 ml of 3,3',5,5' tetramethylbenzidine (Pierce).
Peptide Inhibition Studies.
Inhibition studies were performed at 25°C using serial dilutions from 1 mM to 5 nM of P.berghei and P. falciparum TRAP-tail preincubated with 0.01 mM of freshly purified recombinant PfAldo. The decrease of NADH was continuously monitored spectroscopically after addition of the reaction mix containing 0.2 mM NADH, 50 mg/ml TIM, 50 mg/ml GDH, and 0.1 mM F1,6P. Details of the assay are described in Döbeli et al. (17). Linear regression plots were calculated with Kaleidagraph 4.0 (www.synergy.com).Figure Preparation.
Fig. 1B and SI Figs. 6-8 were prepared with Dino (18) and rendered with POVRAY (www.povray.org). All other figures were prepared with PyMol (19).Evidence for Different Ligand-Binding Modes by Different Subunits of the Same Fructose-1,6 Bisphosphate Aldolase Tetramer in the Literature.
Unequal occupancies of the same compound have been observed in the active site region of different aldolase subunits of the same tetramer. Examples are the following: (i) Several reports on "half of the sites reactivity" of aldolase have been published: Grazi and Trombetta (20); Grazi and Trombetta (21); Grazi et al. (22). (ii) In their crystallographic study on fructose 1,6-bisphosphate (F1,6P) binding to rabbit muscle aldolase, Choi et al. [(23); PDB Code 6ALD] report that "two aldolase subunits show clearer occupancy relative to the other two." In this structure, two subunits are occupied with F1,6P and the other two are empty. (iii) Blom and Sygusch [(24); PDB code 1ADO] describe a structure of aldolase where only two of the four subunits are occupied by the product dihydroxy-acetone phosphate (DHAP). The other two subunits are empty. (iv) The PDB contains a structure (PDB code 1FDJ; for which no accompanying paper is mentioned in the PDB) by Blom, White, and Sygusch, where, according to the coordinates, subunit A and B are occupied by the substrate F1,6P, and subunits C and D by DHAP.1. Kim H, Certa U, Dobeli H, Jakob P, Hol WGJ (1998) Biochemistry 37:4388-4396.
2. Deng JP, Davies DR, Wisedchaisri G, Wu MT, Hol WGJ, Mehlin C (2004) Acta Crystallogr D 60:203-204.
3. Leslie AGW (2006) Acta Crystallogr D 62:48-57.
4. Holton J, Alber T (2004) Proc Natl Acad Sci USA 101:1537-1542.
5. Bailey S (1994) Acta Crystallogr D 50:760-763.
6. Vagin A, Teplyakov A (1997) J Appl Crystallogr 30:1022-1025.
7. Emsley P, Cowtan K (2004) Acta Crystallogr D 60:2126-2132.
8. Brunger AT, Adams PD, Clore GM, Delano WL, Gros P, Grossekunstleve RW, Jiang JS, Kuszewski J, Nilges M, Pannu NS, et al. (1998) Acta Crystallogr D 54:905-921.
9. Winn MD, Isupov MN, Murshudov GN (2001) Acta Crystallogr D 57:122-33.
10. Murshudov GN, Vagin AA, Dodson EJ (1997) Acta Crystallogr D 53:240-255.
11. Painter J, Merritt EA (2005) Acta Crystallogr D 61:465-471.
12. Engh RA, Huber R (1991) Acta Crystallogr A 47:392-400.
13. Lovell SC, Davis IW, Adrendall WB, de Bakker PIW, Word JM, Prisant MG, Richardson JS, Richardson DC (2003) Proteins Struct Funct Genet 50:437-450.
14. Vaguine AA, Richelle J, Wodak SJ (1999) Acta Crystallogr D 55:191-205.
15. Buscaglia CA, Penesetti D, Tao MY, Nussenzweig V (2006) J Biol Chem 281:1324-1331.
16. Buscaglia CA, Coppens I, Hol WGJ, Nussenzweig V (2003) Mol Biol Cell 14:4947-4957.
17. Döbeli H, Trzeciak A, Gillessen D, Matile H, Srivastava IK, Perrin LH., Jakob PE, Certa U (1990) Mol Biochem Parasitol 41:259-268.
18. Philipsen A (2002) DINO: Visualizing Structural Biology, www.dino3d.org.
19. DeLano WL (2002) The PyMOL Molecular Graphics System (DeLano Scientific, Palo Alto, CA), www.pymol.org.
20. Grazi E, Trombetta G (1987) Int J Biochem 19:197-200.
21. Grazi E, Trombetta G (1984) Arch Biochem Biophys 233:595-602.
22. Grazi E., Trombetta G, Lanzara V (1983) Biochem Biophys Res Commun 110:578-83.
23. Choi KH, Mazurkie AS, Morris AJ, Utheza D, Tolan DR, Allen KN (1999) Biochemistry 38:12655-12664.
24. Blom N, Sygusch J (1997) Nature Structural Biology 4:36-39.