Polyketide genes in the marine sponge Plakortis simplex: a new group of mono-modular type I polyketide synthases from sponge symbionts

Summary Sponge symbionts are a largely unexplored source of new and unusual metabolic pathways. Insights into the distribution and function of metabolic genes of sponge symbionts are crucial to dissect and exploit their biotechnological potential. Screening of the metagenome of the marine sponge Plakortis simplex led to the discovery of the swf family, a new group of mono-modular type I polyketide synthase/fatty acid synthase (PKS/FAS) specifically associated with sponge symbionts. Two different examples of the swf cluster were present in the metagenome of P. simplex. A third example of the cluster is present in the previously sequenced genome of a poribacterium from the sponge Aplysina aerophoba but was formerly considered orthologous to the wcb/rkp cluster. The swf cluster was also found in six additional species of sponges. Therefore, the swf cluster represents the second group of mono-modular PKS, after the supA family, to be widespread in marine sponges. The putative swf operon consists of swfA (type I PKS/FAS), swfB (reductase and sulphotransferase domains) and swfC (radical S-adenosylmethionine, or radical SAM). Activation of the acyl carrier protein (ACP) domain of the SwfA protein to its holo-form by co-expression with Svp is the first functional proof of swf type genes in marine sponges. However, the precise biosynthetic role of the swf clusters remains unknown.


DNA isolation and library construction
To ~40 mg of frozen sponge (in RNA later) 700 µL of lysis buffer I (200 mM Tris-Cl, 50 mM EDTA, 1.4 M NaCl, 2% CTAB, 0.5% PVP, all in milliQ®-H 2 O) were added and incubated at 37 °C for 1 h in a thermomixer (1,400 rpm). After addition of 2.8 µL β-mercaptoethanol, 70 µL 10% SDS, 2 µL RNase A (100 mg/mL), and 40 µL proteinase K (10 mg/mL) the tube was incubated at 55 °C for a further hour in a thermomixer (1,400 rpm). At this time, the microcentrifuge tube was spun 4 min at 5,000 rpm. The clear middle phase was transferred to a new microcentrifuge tube containing 750 µL CHCl 3 and centrifuged 10 min at 15,000 rpm. After repetition of the CHCl 3 wash, the supernatant was transferred to a new microcentrifuge tube containing 750 µL of 70% aqueous isopropyl alcohol containing 10% (v/v) 3M NaOAc (pH 5.5) at room temperature. The precipitated DNA was spun down at top speed for 20 min, washed with ice-cold ethanol, dried and dissolved in ~60 µL elution buffer (10 mM Tris-Cl, pH 8.5). The amount of one tube was enough for PCR screening, for library construction the protocol was upscaled to ~500 mg of frozen sponge.
Metagenomic libraries of P. simplex were constructed using the CopyControl Fosmid library production kit (Epicentre Biotechnologies) according to the manufacturer's protocol.
(Introduced restriction sites are underlined and Stop codon is in bold.) The 234 bp long PCR product was subcloned via T/A-cloning in pBluescript II SK(+) (Stratagene) to yield pGS21, and cut out from the vector with BamHI and HindIII. The fragment was then ligated into two different vectors: (i) a pHIS8 vector with an additional gene for the PPTase Svp inserted into its NotI site under the control of the same promoter as the ACP (i.e., pHIS8-Svp, to yield pGS30), and (ii) a pHIS8 vector without this gene (i.e., pHIS8, to yield pGS34). The history of pHIS8-Svp has been described in the literature (Jez et al., 2000;Izumikawa et al., 2006). Proteins from plasmids pGS30 and pGS34 were expressed as N-terminal His8-tag fusion proteins and purified over Ni-NTA columns. To do so, E. coli BL21-CodonPlus®(DE3)-RIPL electrocompetent cells (Stratagene) were transformed with the expression plasmid and spread out on LB agar plates containing 50 µg/mL kanamycin (Kan50) and grown o/n at 37 °C. From this plate, 4 clones were picked to prepare o/n cultures in LB + Kan50. 5 mL of this o/n cultures were used to inoculate 200 mL LB medium without antibiotics. The cultures were incubated at 37 °C for 2-3 h shaking with 250 rpm (OD 600 ~0.6-0.8). After cooling to 16 °C, the cultures were induced with 1 mM IPTG and incubated for further 24 h at 16 °C. After harvesting by centrifugation, the cell pellets were dissolved in 3 mL lysis buffer (50 mM Tris-Cl, 500 mM NaCl, 10 mM MgCl 2 , pH 8.0) and frozen o/n. After thawing on ice and ultrasonication, the lysates were obtained by centrifugation for 20 min at 15,000 rpm in a table top centrifuge at 4 °C.

Cloning and in vitro expression of the swf cluster in E. coli
A region upstream the POR_0546 homolog (22091-22720,~1000 bp upstream supE permease) was amplified by PCR from the fosmid pPS2D9 using primers: 5'-AAA GGA TCC tga cca cgc cct cgt gct ca-3' (forward) and 5'-AAA GAA TTC tgg acc act tcc ggc aac tac-3' (reverse), ligated into pBluescript II SK(+) via T/A cloning to yield pGS15, and cut out from the vector with BamHI and EcoRI. Another region at the end of the swf cluster (i.e. swfC 36006-37813) was amplified from pPS2D9 with the primers 5'-AAA GAA TTC tcg aac gcc ctc ctc atc tac c-3' (forward) and 5'-AAA AAG CTT gcc cac tac gtg ctg cat cgg-3' (reverse), ligated into pBluescript II SK(+) via T/A cloning to yield pGS29, and cut out from the vector with EcoRI and HindIII. Both inserts were ligated in a three-point ligation into the BamHI/HindIII sites of pHIS8-Svp. The resulting plasmid (pGS27) was linearized with EcoRI and dephosphorylated. Homologous recombination was performed with this linear fragment (Kan R ) and the fosmid pPS2D9 (Cam R ) as a donor in the host strain E. coli BW25113 (Datsenko and Wanner, 2000) with the red helper plasmid pKD46 (Amp R ). Transformants were selected on LB + Kan50 agar plates and contained the whole swf cluster in the vector pHIS8-Svp (i.e. the plasmid pGS38). The plasmid was introduced by electroporation into E. coli BAP1 (Pfeifer et al., 2001), which is a derivative of E. coli BL21-(DE3) including a gene for the 4'phosphopantheteinyltransferase Sfp (Lambalot et al., 1996). The strains were grown in baffled 500 mL Erlenmeyer flasks in 100 mL MMGAGTr medium supplemented with 50 µg mL -1 kanamycin. MMGAGTr was prepared as follows: 200 mL M9 salts (64 g Na 2 HPO 4 x 7 H 2 O, 15 g KH 2 PO 4 , 2.5 g NaCl, 5.0 g NH 4 Cl, ad 1000 mL milliQ H 2 O), 500 µL of 2000X trace element solution (1.2 g FeCl 3 x 6 H 2 O, 1.4 g MnSO 4 , 1.6 g CuSO 4 , and 500 mL milliQ H 2 O), 2 mL 1 M MgSO 4 , 100 µL 1 M CaCl 2 , 10 g glucose, 5 g L-glutamic acid, ad 1000 mL milliQ H 2 O. Four clones of transformants were picked from the LB plates to inoculate 2.5 mL MMGAGTr medium in one test tube for overnight cultures (37 °C, 250 rpm) in order to inoculate the 100 mL medium in EM-flasks next day. After 2-3 h of shaking (250 rpm) at 37 °C the T7 promoter was induced with 0.1 mM IPTG, and 5-10 nM cobalamin (vitamin B12) was added as a co-factor of SwfC. The cells and the culture broths were harvested separately after 3 h of further growth at 30 °C, 250 rpm. The cell pellet was dissolved in 1 mL of nanopure water, frozen o/n, and after thawing on ice and ultrasonication, the lysates were subjected to extraction with 4 mL of MeOH. Culture broths were freeze dried o/n and extracted with 5 mL of MeOH/H 2 O (8:2). All the experiments were performed in triplicates.

LC-HR-ESI-MS analyses
The experiments were performed using a Thermo LTQ Orbitrap XL high-resolution ESI mass spectrometer coupled to an Agilent model 1100 LC, which included a solvent reservoir, in-line degasser, binary pump, and refrigerated autosampler. A 2.6 μm Kinetex C18 column (50 × 2.1 mm), maintained at room temperature, was used. It was eluted at 150 μL min −1 with H 2 O and CH 3 CN, using as a gradient elution 70-95% CH 3 CN over 23 min and hold 13 min. Crude extracts from transformants and their culture broths were filtered and injected (5 μL) without any further workup. Both positive-ion and negative-ion mass spectra were recorded in separate HPLC runs.  The KS domain of SwfA shows features that are not shared by any other KS domains, including WcbR/RkpA. These are, for instance, the triple "D" at position 8-10 aa, the "EATAVN" motif (aa 250-255) and 8 additional aminoacids after aa 275, and the NGHCVVR motif at the end (aa 413-419). In addition, the second "Q" in the DPQQR motif is either "I" or "V", which is similar to animal FAS I and not to PKSs (all cis-AT and WcbR/RkpA sequences possess an intact DPQQR motif), and the motif HGTGT of cis-AT and WcbR/RkpA sequences is changed to HATGT, with the A being unique among all KS sequences. Figure S4. The proteins apo-SwfAACP and holo-SwfAACP were expressed as N-terminal His8-tag fusion proteins and purified over Ni-NTA columns. Elution fractions (500 µL each) were obtained by increasing the imidazole concentration in the lysis buffer from 0 to 300 mM stepwise. SDS-PAGE gel was loaded with resuspended cell pellet (5 µL, lane P), column flow through, containing unbound proteins (12 µL, lane FT), washing fractions eluted with lysis buffer (20 µL each, lanes W1 and W2), and fractions eluted with increasing (50-300 mM) imidazole concentrations (20 µL each, lanes 50,100,150, 200, 250, and 300). All the fractions were previously mixed with 3X SDS loading buffer. Lane M is a pre-stained protein marker (Roti ® -Mark STANDARD). S11 Figure S5. Analysis of the fine structure of the pseudomolecular ion peak of holo-SwfA ACP at m/z 1776.7, showing the simultaneous presence of the a 6+ ion of the monomer and a 12+ ion of the dimer. (a) Theoretical spectrum of the 6+ ion of holo-SwfA ACP radical, molecular formula C 460 H 730 N 136 O 146 PS 4 6+ , calculated using the program IsoPro 3.0; (b) Theoretical spectrum of the 12+ ion of holo-SwfA ACP dimer, molecular formula C 920 H 1460 N 272 O 292 P2S 8 12+ . The ions of the monomer and of the dimer have the same average m/z, but a different charge and a different isotope pattern; (c) Sum of the theoretical spectra of the monomer and dimer. (d) Experimental spectrum. Figure S6. PCR detection of swf genes in seven different "high microbial abundance" marine sponges. Primers SWF_ATF and SWF_ATR designed on SwfA AT were used. Abbreviations: Ac, Aiolochroia crassa; Af, Aplysina fulva; Ps, Plakortis simplex; Sa, Smenospongia aurea; M, DNA size marker; If, Ircinia felix; Ts, Theonella swinhoei; Xm, Xestospongia muta; N, negative control.