Substrate Flexibility of the Flavin‐Dependent Dihydropyrrole Oxidases PigB and HapB Involved in Antibiotic Prodigiosin Biosynthesis

Abstract In the biosynthesis of the tripyrrolic pigment prodigiosin, PigB is a predicted flavin‐dependent oxidase responsible for the formation of 2‐methyl‐3‐amylpyrrole (MAP) from a dihydropyrrole. To prove which dihydropyrrole is the true intermediate, both possibilities, 5‐methyl‐4‐pentyl‐3,4‐dihydro‐2H‐pyrrole (5 a, resulting from transamination of the aldehyde of 3‐acetyloctanal) and 2‐methyl‐3‐pentyl‐3,4‐dihydro‐2H‐pyrrole (6, resulting from transamination of the ketone), were synthesised. Only 5 a restored pigment production in a strain of Serratia sp. ATCC 39006 blocked earlier in MAP biosynthesis. PigB is membrane‐associated and inactive when its transmembrane domain was deleted, but HapB, its homologue in Hahella chejuensis, lacks the transmembrane domain and is active in solution. Two colourimetric assays for PigB and HapB were developed, and the HapB‐catalysed reaction was kinetically characterised. Ten analogues of 5 a were synthesised, varying in the C2 and C3 side chains, and tested as substrates of HapB in vitro and for restoration of pigment production in Serratia ΔpigD in vivo. All lengths of side chain tested at C3 were accepted, but only short side chains at C2 were accepted. The knowledge that 5 a is an intermediate in prodigiosin biosynthesis and the ease of synthesis of analogues of 5 a makes a range of prodigiosin analogues readily available by mutasynthesis.

. Alignment of PigB and HapB by EBI Clustal Omega. In orange is the conserved region predicted to form part of the FAD binding site. In red are two other highly conserved regions, Asp69 of HapB is in bold Figure S2. Left: homology model for HapB generated by the PHYRE website from the crystal structure of protoporphyrinogen oxidase from Myxococcus xanthus (PDB entry 2ive). The pose of the FAD is from overlay of residues 55-542 of the homology model (which lacks the FAD) with the 2ive crystal structure (which contains FAD). Right: detail of the binding site for FAD in the model, showing the hydrogen bonds from Asp69 to the 2'and 3'-OH groups of the adenosine part of FAD. Images generated by PyMol.   -3) or EcoRI/PstI (pHDB4) restriction enzymes for 2 h at 37 °C and ligated together with a compatibly digested pQE80L (pHDB1-2) or pQE80::oriT (pHDB3-4). The generated plasmids were transformed into E. coli DH5 and sequenced to confirm the absence of mutation.

Site-directed mutagenesis by overlap extension PCR
Asp69 of HapB was replaced by alanine by site-directed mutagenesis. PCR was performed using a DNA Engine® Peltier Thermal Cycler using PCR cycling parameters similar to those above. Two separate PCRs were carried out using oligonucleotide pairs (HB1 & HB7) and (HB2 & HB8). These PCR products were purified by agarose gel electrophoresis. A second PCR was then carried out using these two PCR products as primers. After purification of the PCR product of the correct size by agarose gel electrophoresis, it was digested with BamHI/SphI restriction enzymes for 2 h at 37 °C and ligated together with a compatibly digested pQE80-L vector to generated plasmid pHDB5. This was subsequently transformed into E. coli DH5. pHDB5 was sequenced by the DNA Sequencing Facility, Department of Biochemistry, University of Cambridge, to confirm the desired point mutation. The plasmid was transformed into protein expression strain E. coli BL21 (DE3) and the mutant HapB protein was overexpressed and purified by Ni-NTA affinity chromatography.
Grant et al. [3] BL21 (DE3) Studier et al. [4] C43 pQE80-L 6xHis fusion expression vector, Amp R Qiagen pQE80::oriT 6xHis fusion expression vector containing oriT gene for bacterial conjugation, Amp R Monson et al. [5]   Preparation of cell lysates: Cells transformed with suitable vector-constructs were grown at 37 °C on a shaker at 250 rpm to obtain an OD600 of 0.6. Cultures were then induced with 1 mM isopropyl-β-Dthiogalactopyranoside (IPTG) at 16 °C for 14-16 h. Aliquots were collected before and after the induction by IPTG to monitor the expression of protein and analyzed by Sodium Dodecyl Sulphate-Polyacrylamide Gel Electrophoresis (SDS-PAGE). Cells were harvested by centrifuging at 5000 rpm for 20 min at 4 °C. The cell pellet was resuspended in the lysis buffer including complete-mini-EDTA-free protease inhibitor cocktail (1 tablet per 10 ml lysis buffer) from Roche.
For protein purification from soluble fraction: The above lysis buffer was supplemented with 1% Triton-X100. The ice-water jacketed suspension was sonicated and then centrifuged at 10,000 rpm at 4 °C for 30 min. The clarified lysate was loaded onto a Ni-NTA (Ni 2+ -nitrilotriacetate) column (Qiagen) for affinity purification. The column was washed twice with wash buffer and the bound protein was eluted (1 ml fractions) with elution buffer and the fractions analysed by SDS-PAGE. The fractions containing purified protein were pooled and dialyzed in storage buffer.
For protein isolation as membrane fraction: Cell pellets resuspended in lysis buffer without detergents were subjected to lysis using the EmulsiFlex (Avestin) at 4 °C. The cell lysate was then ultracentrifuged at 28,000 rpm (Beckman Coulter -Optima™ L-100XP, rotor SW28) at 4 °C for 2 h to separate into three layers -the cell debris, membrane fraction and clarified lysate. The membrane fraction after dilution with 50% glycerol was used for SDS-PAGE analysis, activity determination and kinetic studies.
After growing the cell overnight at 30 °C at 250 rpm, pigment was extracted following the method described in the paper. The solvent was then removed under reduced pressure in a pre-weighed flask. The extracts were redissolved in CH3CN/H2O 1:1 at 10 mg/ml. When some solid was still visible after vortexing, the sample was filtered through cotton wool.
H2MAP: 200 ml of LB broth + sorbitol 0.25 M in a 2 L flask were inoculated with a 5 ml overnight culture of Serratia ΔpigB (or ΔpigD for negative control). The cells were cultured at 30 °C for 16 h. The cultures were centrifuged, and the cell pellets extracted with 100 ml of acetone/

Purification of Prodigiosin Analogues
Prodiginine samples were prepared from 1 L of culture of Serratia ΔpigD complemented with 5e or 5h.
The ethanolic extracts were sterile-filtered (0.22 µm pores) and concentrated under reduced pressure to dryness. The residue was resuspended in 1 ml of H2O/CH3CN and purified by HPLC (Agilent Zorbax SB-C18, flow: 3 ml/min, gradient 5 to 100 % CH3CN in water over 28 min then 100% for 2 min). Fractions containing the prodiginine were concentrated under reduced pressure and resuspended in acidified EtOH (50 µl

Substrate specificity assays
Agar feeding assays Figure S9: Serratia ΔpigB supplemented with H2MAP shows negligible formation of prodigiosin.
Figure S10: Results of agar plate feeding assay using compounds 5a-k.

Determination of the extinction coefficient for Ehrlich's assay
MAP was synthesized following the procedure described by Williamson et al. [2] Ehrlich's assay was performed on solution of MAP ranging between 0.1 and 0.8 mM, leading to the results shown in Fig. S8. As expected, a linear increase was obtained. The assay was performed in a 96 well microplate, with 50 µl of MAP solution, 50 µl of stop reagent and 100 µl of Ehrlich's reagent.