This Article Abstract Full Text Services Email this article to a colleague Alert me to new issues of the journal Request Copyright Permission Citing Articles Citing Articles via CrossRef

McDonough et al. 10.1073/pnas.0601283103.

Supporting Information

Files in this Data Supplement:

Supporting Figure 6
Supporting Figure 7
Supporting Figure 8
Supporting Figure 9
Supporting Figure 10
Supporting Figure 11
Supporting Figure 12
Supporting Text
Supporting Table 1

Fig. 6. Stereoview of the C-a trace of PHD2_cat (black) superimposed with DAOCS (gray) (PDB ID code 1E5I; ref. 1). View of the crystal structure of FIH (PDB ID code 1H2N; ref. 2) showing how its C terminus is involved in formation of a homodimer by using interlocking helices (within box). The homodimer has been observed in solution by native gel electrophoresis, gel filtration, and native MS. The C-terminal helices of FIH are involved in productive substrate binding by forming a hydrophobic pocket at the oligomerization interface, in which Leu-795 of the HIF1-a substrate binds (2, 3).

1. Lee, H. J., Lloyd, M. D., Harlos, K., Clifton, I. J., Baldwin, J. E. & Schofield, C. J. (2001) J. Mol. Biol. 308, 937-948.

2. Elkins, J. M., Hewitson, K. S., McNeill, L. A., Seibel, J. F., Schlemminger, I., Pugh, C. W., Ratcliffe, P. J. & Schofield, C. J. (2003) J. Biol. Chem. 278, 1802-1806.

3. Lancaster, D. E., McNeill, L. A., McDonough, M. A., Aplin, R. T., Hewitson, K. S., Pugh, C. W., Ratcliffe, P. J. & Schofield, C. J. (2004) Biochem. J. 429-427.

Fig. 7. PHD2 active site contacts with compound A. Contacts are indicated by dashed lines between atoms, and distances are in angstroms.

Fig. 8. (a) The effect of sample cone voltage (SC) in electrospray ionization mass spectrometric (ESI-MS) analysis of compound A binding to PHD2_181-426 (1eq/1eq). Peaks at 28,107 Da (A) and 28,480 Da (B) represent the binary complex PHD2_181-426(Fe^II) (or E) and the ternary complex PHD2_181-426(Fe^II)-compound A (or EI), respectively. The molecular mass difference between these two peaks (B-A) is 373 Da, which is in close agreement with the calculated mass of compound A (error » 0.3%). Note that there is no evidence for covalent modification of the inhibitor. (b) Deconvoluted ESI-MS spectrum of an equimolar mixture of E, 2OG, and compound A. Peak at 28,109 Da (A) represents the binary complex E; peaks at 28,256 Da (B) and 28,487 Da (C) represent the ternary complexes E-2OG and E-compound A, respectively. Note that there is no evidence for the formation of the quaternary complex E-2OG-compound A, indicating that compound A competes with 2OG for binding to E.

Fig. 9. Sequence alignment of the catalytic domain of PHD homologues: mPHD2 gi|32129512 Mus musculus; rPHD2 gi|28274768 Rattus norvegicus; hPHD2 gi|13489073 Homo sapiens; gi|24459908 Takifugu rubripes; mosquito gi|21296789 Anopheles gambiae str. PEST; mPHD1 gi|18182546 Mus musculus; rPHD1 gi|28631169 Rattus norvegicus; hPHD1 gi|14547148 Homo sapiens; mPHD3 gi|14547243 Mus musculus; rPHD3 SM-20 gi|469478 Rattus norvegicus; hPHD3 gi|11545787 Homo sapiens; fatiga A gi|7296772 Drosophila melanogaster; fatiga C gi|23170432 Drosophila melanogaster; fatiga B gi|20151779 Drosophila melanogaster; gi|39593339 Caenorhabditis briggsae; gi|5923812 EGL-9 Caenorhabditis elegans; gi|23027653 Microbulbifer degradans; gi|22998686 Magnetococcus sp.; MC-1 gi|27361757 Vibrio vulnificus; gi|28806413 Vibrio parahaemolyticus; gi|9658386 Vibrio cholerae O1 biovar eltor; gi|24349818 Shewanella oneidensis; gi|46187776 Pseudomonas syringae; gi|28855663 Pseudomonas syringae pv. tomato; gi|24986951 Pseudomonas putida; gi|23060943 Pseudomonas fluorescens; gi|9946156 Pseudomonas aeruginosa; gi|32043481 Pseudomonas aeruginosa; gi|23105771 Azotobacter vinelandii; gi|34102493 Chromobacterium violaceum; gi|39576174 Bdellovibrio bacteriovorus; HD100 gi|23135877 Cytophaga hutchinsonii; P4H1 gi|28828253 Dictyostelium discoideum. Sequence alignment was derived from 3D-PSSM structure prediction server (1).

1. Kelley, L. A., MacCullum, R. M & Sternberg, M. J. (2000) J. Mol. Biol. 299, 499-520.

Fig. 10. Sequence alignment of the hydroxylase domains of closely related PHDs from human, mouse, rat, fruit-fly, and mosquito. The key to descriptors and accession numbers is the following: hPHD2 gi|13489073 Homo sapiens; mPHD2 gi|32129512 Mus musculus; rPHD2 gi|28274768 Rattus norvegicus; mPHD1 gi|18182546 Mus musculus; hPHD1 gi|14547148 Homo sapiens; mPHD3 gi|14547243 Mus musculus; rPHD3 SM-20 gi|469478 Rattus norvegicus; hPHD3 gi|11545787 Homo sapiens; fatiga B gi|20151779 Drosophila melanogaster; fatiga C gi|23170432 Drosophila melanogaster; fatiga A gi|7296772 Drosophila melanogaster; mosquito gi|21296789 Anopheles gambiae str. PEST. Sequence alignment performed using CLUSTALW in default mode (1).

1. Chenna, R., Sugawara, H., Koike, T., Lopez, R., Gibson, T. J., Higgins, D. G. & Thompson, J. D. (2003) Nucleic Acids Res. 31, 3497-3500.

Fig. 11. Structural similarities between the catalytic domains of the three human PHDs. The surface of PHD2_cat is colored by its sequence homology to the human isozymes PHD1 and PHD3. Dark blue represents total conservation of residues, light blue indicates semiconserved residues, and white indicates nonconserved. The yellow arrows point through highly conserved grooves on the proteins surface that are of the appropriate size and shape for a peptide substrate and thus indicate a potential substrate-binding site.

Fig. 12. Harker sections of the anomalous difference Patterson of the PHD2_181-426 SAD data (a) and the predicted Patterson (b) based on the seven heavy atom sites located by SOLVE displayed in the xy plane, z = 0.5; m = 2:y, –x + y,1/2; m = 1:2*x,2*y,1/2 and calculated by using a |DF|/|sDF| cutoff of 0.25 and is contoured beginning at 3s with 0.5s intervals. The dominating iron and iodine peaks are >20s over the weaker sulfur peaks.

*R_sym = ∑|I-<I>|/∑I, where I is the intensity of an individual measurement and <I> is the average intensity from multiple observations.

Supporting Text

Construct Design, Expression, and Purification. The PHD2_181-426 expression construct (prolyl hydroxylase, PHD)was designed based on the results from a 3D-PSSM (1) search, which indicated the catalytic domain of PHD2 as having structural similarity to deacetoxycephalosporin C synthase (DAOCS) (2) and implied that the beginning of its C-terminal prolyl hydroxylase domain started at approximately residue 181. PHD2_181-426 was cloned into the pET-28a(+) vector (Novagen), incorporating an N-terminal hexa-histidine tag. The construct was transformed into BL21(DE3) Escherichia coli cells, and protein expression was induced with 0.5 mM isopropyl-b-D-thiogalactosidase for 3-4 h at 37°C. Cells were harvested and lysed by sonication in 0.1 M MES (pH 5.8), and soluble protein (»5% total soluble extract; ref. 3) was purified by cation exchange chromatography. Overnight thrombin digestion to remove the histidine tag was followed by further purification by using size exclusion chromatography. Proteins were buffer exchanged into 20 mM Tris×HCl (pH 7.5) and concentrated to 15-30 mg/ml. Purity to >95% was verified by SDS/PAGE and electrospray ionization MS(3). The PHD2_181-417 construct was cloned with a C-terminal hexa-histidine tag into the pET-28a(+) expression vector. The construct was transformed into BL21 Escherichia coli cells (Amersham Pharmacia Biosciences), and protein expression was induced for 6 h at 37°C. Harvested cells were resuspended in buffer and lysed by a microfluidizer. Cleared lysate was applied to Ni-NTA resin and incubated for 2 h. Protein was eluted with 250 mM imidazole, concentrated to 15 mg/ml, and applied to a size exclusion column (Superdex 200, Amersham Pharmacia) equilibrated with buffer. Purified protein was typically analyzed by liquid chromatography-MS, and aliquots were snap frozen in liquid nitrogen.

Soft-Ionization Mass Spectrometry Analysis of PHD2_181-426. PHD2_181-426 was desalted by using Microcon YM-10 (cut off = 10,000 Da) centrifugal filters (Millipore, Bedford MA) in 15 mM ammonium acetate (pH 7.5). Seven dilution/concentration steps were performed at 4°C and 14,000 ´ g. The stock solution was diluted in the same buffer to a final concentration of 100 mM.

Compound A was dissolved in 100% DMSO to a concentration of 100 mM then diluted to 100 mM in 15 mM ammonium acetate (pH 7.5). A 15 mM equimolar sample of inhibitor and enzyme in 15 mM ammonium acetate at pH 7.5 was prepared, and a 5-ml aliquot was placed in a 96-well plate and analyzed by electrospray mass spectrometry by using a NanoMate (Advion Biosciences, Hethersett Norwich Norfolk, U.K.) linked to a Q-TOF mass spectrometer (Q-TOF micro, Micromass, Altrincham, U.K.) with a spray voltage of 1.70 kV ± 0.2 kV and a delivery pressure of 0.25 psi (1 psi = 6.89 kPa). Calibration and sample acquisitions were performed in the positive ion mode in the range of 500–5,000 m/z. Sample cone voltage was varied between 10 and 200 V with a source temperature of 40°C and with acquisition/scan times of 10 s/1 s. The pressure at the interface between the atmospheric source and the high vacuum region was fixed at 6.60 mbar. Data were processed by using MASSLYNX 4.0 (Waters).

Crystallization Details. PHD2_181-426 crystallization samples were prepared by adding a stock solution of 100 mM compound A in 100% DMSO directly to protein at 15-30 mg/ml in 100 mM Tris×HCl (pH 7.5) to a final compound A concentration of 4 mM and DMSO concentration of 2%. Hanging and sitting drop (1:1 well:protein) crystallization experiments were performed in an anaerobic glove box (Belle Technologies, Dorset, U.K.) at 18°C by using the following well conditions: 50 mM MES (pH 6.5)/1.5-2.0 M ammonium sulfate/0.5-3.5% dioxane/1 mM iron(II) sulfate. Crystal growth initiated within 24 h and continued for up to 5 days. PHD2_181-417-(His)₆ protein was concentrated to »15 mg/ml by using a Vivaspin concentrator (Sartorius) followed by the addition of 10-fold molar excess of the inhibitor dissolved in 100% DMSO. Using the hanging drop vapor diffusion method, Hampton HT Crystal and Index Screens (Hampton Research, Aliso Viejo, CA) were set up aerobically at room temperature with most crystals growing within 24 h. Initial crystallization screen hits were further optimized to 28% polyethylene glycol 8000/0.2 M ammonium sulfate/100 mM MES, pH 6.4. Using this crystallization condition, crystals were allowed to grow for 24 h and then flash frozen in a 30% ethylene glycol cryo-solution (3:7 ethylene glycol:well) by plunging into liquid N₂ for data collection. These crystals were used to collect native data up to 1.7 Å. A mercury derivative was prepared by adding 1 ml of 100 mM HgAc₂ to crystals in a 5-ml drop of well solution and soaked overnight. Mercury derivative data were collected to 2.6 Å by using the same cryo-conditions used for native crystals. The selenomethionine-derivatized PHD2_181-417-(His)₆ crystals were obtained by using 0.2 M lithium sulfate/0.1 M Tris×HCl (pH 8.5)/30% polyethylene glycol 4000 and flash frozen in a 30% ethylene glycol cryo-solution (3:7 ethylene glycol:well) by plunging into liquid nitrogen for data collection. All data were collected at 100 K.

Data Collection and Processing. The ability to solve structures based on anomalous differences by using a single, highly redundant, data set collected on a rotating anode x-ray generator at a home institution is attractive relative to the collection of heavy atom data sets in an attempt to find one that is isomorphous with native data and has heavy atom(s) bound or to arranging the time and expense of traveling to or shipping crystals to a synchrotron with tunable wavelength for multiwavelength anomalous dispersion (MAD) data collection. For PHD2_181-426, its high symmetry space group (P6₃) together with only one molecule in the asymmetric unit and two atoms that have relatively strong anomalous differences at CuKa, Df'' for Fe and I at 8 keV is 6.91 and 3.18, respectively, made it an ideal situation for evaluating the use of in-house single-wavelength anomalous diffraction (SAD) data collection. In addition, collecting a highly accurate and redundant data were made possible by using a four-circle goniometer, microfocus optics on a rotating x-ray generator, and a charge-coupled device (CCD) detector. MAD x-ray diffraction data and the high-resolution 1.7 Å native data were collected at 100 K beamline 5.0.2 at the Advanced Light Source, Berkeley, CA. Mercury derivative x-ray diffraction data were collected in-house on an R-AXIS IV++ detector mounted on a FR-E SuperBright rotating copper anode generator operating at 45 kV and 45 mA, equipped with VariMax optics (Rigaku/MSC, The Woodlands, TX). Cryo-cooling was carried out by using the X-Stream system (Rigaku/MSC). The data were processed with HKL2000 (HKL Research, Charlottesville, VA). The space group was determined to be P6₃ with one molecule in the asymmetric unit (Table 1).

Structure Solution and Refinement. Parameter and topology files for compound A were generated by using PRODRG (4) for refinement of the PHD2_181-426 structure in CNS (5). Iterative cycles of model building in COOT (6) and slowcool-simulated annealing refinement by using the maximum-liklehood function and bulk-solvent modeling in CNS proceeded until the R_cryst/R_free no longer converged while decreasing. PROCHECK (7) was used to monitor geometric quality of the model between refinement cycles and pick out poorly modeled areas needing attention. Water molecules were added to peaks >1.8s in 2F_o – F_c electron density maps that were within hydrogen bonding distance to protein with reasonable hydrogen bonding geometry until R_cryst/R_free no longer converged while decreasing (21.67/28.97). Crystallographic weighting factors were optimized for the final rounds of refinement.

Parameter and topology files for CNX2000.12 (5) for compound A were generated by using QUANTA2000.2 (Accelrys, Inc., San Diego) for refinement of the PHD2_181-417 structure. Iterative cycles of model building in QUANTA2000.2 and slowcool-simulated annealing refinement by using the maximum-liklehood function and bulk-solvent modeling in CNX2000.12 were carried out until the R_cryst/R_free no longer converged while decreasing. The geometric quality of the structure was monitored by using the Protein Health module of QUANTA2000.2 during the iterative cycles of model building and refinement. Water molecules were picked to peaks >2.0s in F_o – F_c electron density maps based on a reasonable hydrogen bonding distance criteria and then manually checked for final acceptance before proceeding to the refinement (R_cryst/R_free of 0.226/0.256) with a data set collected to 1.7 Å. The final round of refinement with the optimized crystallographic weighting factors yielded R_cryst/R_free of 0.216/0.253. The coordinates have been deposited in the Protein Data Bank as 2G1M and 2G19. Cavity depth (Fig. 4) was assessed by using SYBYL 7.0 (Tripos, Inc., St. Louis)

Comparative Homology Modeling. Homology models of the prolyl-hydroxylase domains of human PHD1 and human PHD3 were built by using the sequence alignment in Fig. 10 and the PHD2_cat structure as a template in the automodel feature of MODELLERv8.1 (8).

1. Kelley, L. A., MacCallum, R. M. & Sternberg, M. J. (2000) J. Mol. Biol. 299, 499-520.

2. Valegard, K., van Scheltinga, A. C. T., Lloyd, M. D., Hara, T., Ramaswamy, S., Perrakis, A., Thompson, A., Lee, H. J., Baldwin, J. E., Schofield, C. J., et al. (1998) Nature 394, 805-809.

3. McNeill, L. A., Flashman, E., Buck, M. R. G., Hewitson, K. S., Clifton, I. J., Jeschke, G., Claridge, T. D. W., Ehrismann, D., Oldham, N. J. & Schofield, C. J. (2005) Mol. BioSys. 4, 312-324.

4. Schuttelkopf, A. W. & van Aalten, D. M. (2004) Acta Crystallogr. D 60, 1355-1363.

5. Brunger, 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., et al. (1998) Acta Crystallogr. D 54, 905-921.

6. Emsley, P. & Cowtan, K. (2004) Acta Crystallogr. D 60, 2126-2132.

7. Laskowski, R. A., MacArthur, M. W., Moss, D. S. & Thornton, J. M. (1993) J. Appl. Crystallogr. 26, 283-291.

8. Sali, A. & Blundell, T. L. (1993) J. Mol. Biol. 234, 779-815.

This Article Abstract Full Text Services Email this article to a colleague Alert me to new issues of the journal Request Copyright Permission Citing Articles Citing Articles via CrossRef