Heterologous Expression of Cryptomaldamide in a Cyanobacterial Host

Abstract

Filamentous marine cyanobacteria make a variety of bioactive molecules that are produced by polyketide synthases, nonribosomal peptide synthetases, and hybrid pathways that are encoded by large biosynthetic gene clusters. These cyanobacterial natural products represent potential drug leads; however, thorough pharmacological investigations have been impeded by the limited quantity of compound that is typically available from the native organisms. Additionally, investigations of the biosynthetic gene clusters and enzymatic pathways have been difficult due to the inability to conduct genetic manipulations in the native producers. Here we report a set of genetic tools for the heterologous expression of biosynthetic gene clusters in the cyanobacteria Synechococcus elongatus PCC 7942 and Anabaena (Nostoc) PCC 7120. To facilitate the transfer of gene clusters in both strains, we engineered a strain of Anabaena that contains S. elongatus homologous sequences for chromosomal recombination at a neutral site and devised a CRISPR-based strategy to efficiently obtain segregated double recombinant clones of Anabaena. These genetic tools were used to express the large 28.7 kb cryptomaldamide biosynthetic gene cluster from the marine cyanobacterium Moorena (Moorea) producens JHB in both model strains. S. elongatus did not produce cryptomaldamide; however, high-titer production of cryptomaldamide was obtained in Anabaena. The methods developed in this study will facilitate the heterologous expression of biosynthetic gene clusters isolated from marine cyanobacteria and complex metagenomic samples.

Keywords: cryptomaldamide; cyanobacteria; heterologous expression; natural products.

Figure 1.. Biosynthesis of cryptomaldamide by M. producens.

(A) Micrograph of M. producens. (B) Cryptomaldamide structure. (C) Putative biosynthetic gene cluster for the biosynthesis of cryptomaldamide (cpm) in M. producens. The cpmA gene encodes for an amidinotransferase that is believed to initiate the pathway through transfer of an amidino group from arginine to serine to produce an amidino-serine residue. The cpmB gene encodes a PKS/NRPS megasynthase.

Figure 2.. Strategy for cloning the cryptomaldamide BGC into S. elongatus.

The cryptomaldamide BGC was amplified by PCR from genomic DNA as 3 overlapping fragments covering 28,095 bp starting 408 nucleotides upstream of the cpmA start codon to 47 nucleotides downstream of a multi-antimicrobial extrusion protein (MATE) efflux family protein gene. The first and last PCR products carried 40 nucleotides, pink and green dashes, that overlap with the ends, pink and green segments, of the linearized S. elongatus TAR cloning vector pAM5571. The 3 PCR products and pAM5571 were assembled in S. cerevisiae by recombination. Yeast clones containing plasmids carrying the entire BGC were identified by PCR. Positive plasmids were then transformed into E. coli and further verified by restriction digests with NcoI. Finally, positive plasmid clones prepared from E. coli were transformed into S. elongatus. Red arrows, forward primers; blue arrows, reverse primers.

Figure 3.

Cloning vectors and Anabaena chromosome engineering. (A) TAR cloning vector that comprises the following modules: (1) yeast components ARSH/CEN6 for replication and trp1 encoding tryptophan synthetase and ura3 encoding orotidine-5-phosphate decarboxylase (ODCase) for selection or counter selection in yeast, (2) the pBR322 origin of replication and the antibiotic resistance genes for kanamycin (aphI) and gentamycin (aac(3)IV) for replication and selection in E. coli, (3) the origin of transfer (RK2oriT) of the RK2 plasmid for conjugation into other microorganisms including cyanobacteria, and (4) S. elongatus NS2 homologous sequences for recombination into neutral site 2 (NS2). (B) Linear map of the plasmid constructed to engineer the Anabaena chromosome to harbor the S. elongatus NS2. (C) CRISPR vector that includes the cpf1 gene and CRISPR array from pSL2680, a Sp/Sm resistance aadA gene, and a modified RSF1010 replicon carrying an RK2 origin of transfer and a high-copy-number E. coli origin of replication.

Figure 4.. Detection of cryptomaldamide.

(A) Representative LC-MS Total Ion Chromatograms (TICs) of the extracts obtained from the native M. producens strain and heterologous hosts Anabaena and S. elongatus carrying the cryptomaldamide BGC. Cryptomaldamide peaks are indicated with asterisks (*). (B) Representative tandem mass spectrum indicating m/z fragmentations of cryptomaldamide isolated from a culture of Anabaena that expresses the cryptomaldamide BGC.

Figure 5.. Production of cryptomaldamide in Anabaena.

(A) Total cryptomaldamide concentrations (bars) at 3 time points (1, 6, and 20 days) obtained for cultures (cell biomass + growth medium) of two independent Anabaena clones (AMC2564 and AMC2565) that contain the cryptomaldamide BGC, and cell density (dots, OD₇₅₀) over a 20-day time course. 2-L cultures were grown in 2.8-L Fernbach flasks in a 3% CO₂ atmosphere. The OD₇₅₀ values are the mean values ± standard deviations for the 2 cultures. OD₇₅₀ was measured every day for the first 6 days and every 2 days for the rest of the time course. (B) Amount of cryptomaldamide in the biomass and the medium normalized to cell biomass dry weight. The cells and medium were collected on days 1, 6, and 20. (C) Concentration of cryptomaldamide in 50-mL cultures of AMC2565 grown in 125-mL flasks in a 3% CO₂ atmosphere with and without 5 mM arginine and grown in air without supplemental CO₂. Cultures were grown until the cells reached an OD₇₅₀ of approximately 1.5. Cryptomaldamide concentrations were normalized to cell density by dividing by the culture OD₇₅₀. The experiment was carried out in triplicate for each condition, and cryptomaldamide concentrations are shown as mean values ± standard deviations.