Biosynthesis of the enediyne antitumor antibiotic C-1027 involves a new branching point in chorismate metabolism -- Van Lanen et al. 105 (2): 494 Data Supplement - HTML Page - index.htslp -- Proceedings of the National Academy of Sciences

Van Lanen et al10.1073/pnas.0708750105XXYYYYY103.

Supporting Information

Files in this Data Supplement:

SI Figure 6
SI Figure 7
SI Table 1
SI Figure 8
SI Figure 9
SI Figure 10
SI Text

SI Figure 6A
SI Figure 6B

Fig. 6. Alignment of SgcD (accession no. AAL06664) with ASI from S. typhimirium (NP_460682) and S. marcescens (AAA57308). The filled triangle indicates amino acids involved in substrate, product, and Mg²⁺ binding or amino acids implicated in catalysis, and boxed residues indicate amino acids involved in Trp inhibition. The results, including the appropriate references, are summarized in SI Table 1.

SI Figure 7

Fig. 7. Model of SgcD with S. marcescens ASI (pdb 1I7QA). SgcD was submitted to ModWeb for comparative protein structure modeling (http://alto.compbio.ucsf.edu/modweb-cgi/main.cgi). The results were analyzed using Chimera from the University of San Francisco. (A) Overall predicted structure similarity between ASI and SgcD. Shown in purple is ASI, and SgcE is depicted in white. (B) Close-up view of the ASI active site comparing select amino acids involved in acid/base catalysis and Mg²⁺ coordination. Residues in red are from ASI, and those in blue are from SgcD. Numbering refers to the residues in ASI.

SI Figure 8

Fig. 8. Alignment of SgcG (European Molecular Biology Laboratory Nucleotide Sequence Database accession no. AAL06666) with Methanosarcina thermophila (AAC45465) and E. coli WrbA (NP_415524). The filled triangle indicates the four Cys residues involved in iron coordination.

SI Figure 9

Fig. 9. SDS/PAGE of SgcD (lane 1), molecular weight markers (lane 2), and SgcG (lane 3). The expected molecular masses are 58. 5 kDa for SgcD and 29.1 kDa for SgcG (an extra 5 kDa is attributed to the engineered His-tag).

SI Figure 10

Fig. 10. Flavin analysis of SgcG. (A) UV/Vis spectrum of the purified protein and comparison to authentic FAD. (B) HPLC determination of the flavin cofactor with authentic FMN (I), authentic FAD (II), extract of boiled SgcG (III), and extract of boiled SgcG spiked with authentic FAD.

Table 1. Comparison of SgcD with the proposed active site residues from S. marcescens ASI

Feature	Amino Acid
	S. marcescens	S. globisporus
	ASI	SgcD
Pyruvate coordination (1)	Tyr 449	Tyr 427
	Arg 469	Arg 447
	Gly 483*	Ser 461
Benzoate coordination (1)	Gly 328	Gly 313
Benzoate coordination (1)	Gly 485	Gly 463
Mg²⁺ coordination (1)	Glu 358	Glu 336
	Glu 361	Glu 339
	Glu 495	Glu 473
	Glu 498	Glu 476
General acid/base (1, 2)	Thr 329	Thr 314
General acid/base (1, 2)	His 398	His 376
Trp Inhibition^†	Glu 38	Ala 33
	Ser 39	Glu 34
	Met 292	Met 279
	Cys 464	Ala 442

*, Coordination occurs with the N from the polypeptide backbone. †, Reported using S. typhimirium ASI (3); for consistency, shown are the corresponding residues for S. marcescens ASI.

1. Spraggon G, Kim C, Nguyen-Huu X, Yee MC, Yanofsky C, Mills SE (2001) Proc Natl Acad Sci USA 98:6021-6026.

2. Morollo AA, Bauerle R (1993) Proc Natl Acad Sci USA 90:9983-9987.

3. Caligiuri MG, Bauerle R (1991) J Biol Chem 266:8328-8335

SI Text

Cofactor Analysis of SgcG.
SgcG was denatured by boiling for 3 min or by addition of TCA to 5%. After centrifugation, the clarified supernatent was loaded onto an Apollo C18 reverse phase column (4.6 ´250 mm, 5 mm; Grace Davison) equilibrated with 0.1% formic acid in 10% methanol (solvent D). A series of linear gradients were developed from solvent D to solvent D containing 70% methanol (solvent E) in the following manner: (beginning time and ending time with linear increase to percentage of solvent E): 0-2 min, 0% E; 2-32 min, 100% E; 32-35 min, 100% E; and 35-36 min, 0% E. The flow rate was kept constant at 1.0 ml/min, and elution was monitored at 260 nm.

Ferric iron content was determined using a previously described spectroscopic method (1). For this analysis, SgcG was purified aerobically or anaerobically, quantitated using Bradford protein dye binding assay (Bio-Rad) or a BCA protein assay kit (Pierce), and precipitated with heat, TCA treatment, or a Compat-Able protein assay kit (Pierce). In all cases, a 1:2 molar ratio of Fe²⁺:SgcG was detected. Attempts to reconsititute the iron-sulfur cluster following a described procedure (2) except excluding 2-mercaptoethanol failed to alter the ratio.

Enzymatic Preparation of ADIC.
ADIC was prepared enzyamtically in a reaction containing 100 mM Tris×HCl pH 8.0, 10 mM MgCl₂, 333 mM NH₄Cl, 0.5 mM DTT, 5 mM chorismate, and 1.0 mg/ml SgcD in 1.8 ml at 30°C. After 2 h, the reaction was terminated with ice cold TCA to 5% and the precipitated protein was removed by centrifugation. The clarified supernatent was loaded onto a NovaPak C18 reverse phase column (7.6 ´300 mm, 5 mm, Waters) equilibrated with 0.1% formic acid in 8% acetonitrile (sovent F). A series of linear gradients were developed from Solvent F to Solvent F containing 80% acetonitrile (solvent G) in the following manner: (beginning time and ending time with linear increase to percentage of solvent G): 0-5 min, 0% G; 5-20 min, 15% G; 20-30 min, 46% G; 30-34 min, 100%G, 34-40 100%G, and 40-44 min, 0% G. The flow rate was kept constant at 3.0 ml/min, and elution was monitored at 280 nm.

The peak corresponding to ADIC was collected, lyophilized, and resuspended in H₂O, and the concentration was determined by using e_{278 nm} = 11,500 M^-1×cm^-1 (3). ¹H NMR (CD₃OD, 400 MHz): d 7.37 (d, J = 4.5 Hz, 1H), 6.48 (m, 2H), 5.60 (d, J = 3.0 Hz, 1H), 5.08 (d, J = 2.7, 1H), 4.90 (m, 1H), 4.51 (d, J = 5.2, 1H). The data are consistent with the diaxial confirmation of ADIC as reported in refs. 4 and 5.

Enzymatic Preparation of OPA.
OPA was prepared enzyamtically in a reaction containing 100 mM Tris×HCl pH 8.0, 10 mM MgCl₂, 333 mM NH₄Cl, 0.5 mM DTT, 5 mM chorismate, 1.0 mg/ml SgcD and 0.7 mg/ml SgcG in 1.8 ml at 30°C. After 2 h, the reaction was terminated, and the protein was removed by ultra-filtration. The clarified supernatent was loaded onto a NovaPak C18 reverse phase column (7.6 ´300 mm, 5 mm) equilibrated with 10 mM ammonium acetate, 0.1% triethylamine in 8% acetonitrile (pH 6.0, sovent F). A series of linear gradients were developed in an identical manner as for ADIC preparation from solvent F to solvent F with 80% acetonitrile. The flow rate was kept constant at 3.0 ml/min, and elution was monitored at 316 nm. Properties and Structure of OPA.
OPA was purified as an off-white powder. The UV/Vis spectrum of OPA exhibited two peak absorption bands at l_max 218 and 335 nm in methanol. The extinction coefficient for OPA ( _{335 nm} = 3970 ±50 M^-1×cm^-1 in MeOH) was determined gravimetrically from three individual OPA preparations. HPLC-atmospheric pressure chemical ionization (APCI)-MS analysis yielded [M+H]⁺ and [MH]^- ions at m/z = 224.1 and 222.1, respectively, suggesting a molecular weight of 223.1. High resolution ESI-MS analysis of OPA afforded a [M+H]⁺ ion at m/z = 224.0541, suggesting a molecular formula of C₁₀H₉O₅N (calculated m/z = 224.0481) with seven degrees of unsaturation. ¹H NMR (CD3OD, 500 MHz): d 7.71 (d, J = 8.4 Hz, 1H), 7.04 (d, 8.0 Hz, 1H), 6.60 (m, 1H), 5.45 (s, 1H), 4.61 (s, 1H). ¹³C NMR (CD₃OD, 500 MHz): d 98.0 (C = C), 111.0 (ArC), 114.8 (ArC), 123.9 (ArC), 127.6 (ArC), 142.8 (ArC), 144.5 (ArC), 154.8 (C = C), 169.0 (C = O), 172.2 (C = O). The final assignments were aided by ¹H-COSY, HMQC, and gHMBC two-dimensional experiments.

1. Fish WW (1988) Methods Enzymol 158:357-364.

2. Latimer MT, Painter MH, Ferry JG (1996) J Biol Chem 271:24023-24028.

3. Morollo AA, Bauerle R (1993) Proc Natl Acad Sci USA 90:9983-9987.

4. Morollo AA, Finn MG, Bauerle R (1993) J Am Chem Soc 115:816-817.

5. Policastro PP, Au KG, Walsh CT, Berchtold GA (1984) J Am Chem Soc 106:2443-2444.