Investigation of the Unusual Carbon-Carbon Bond Formation in Cylindrocyclophane Biosynthesis

Investigation of the Unusual Carbon-Carbon Bond Formation in Cylindrocyclophane Biosynthesis

Nature constructs structurally diverse, bioactive molecules using enzymes. Many enzymes catalyze synthetically challenging reactions under mild, physiological conditions. Consequently, they have long been a source of inspiration for developing biomimetic organic syntheses and methods. In addition, enzymes are increasingly being used as biocatalysts in industry. Therefore, the discovery of enzymes that catalyze chemically intriguing transformations can positively impact synthesis in multiple ways. With the recent advances in next-generation DNA sequencing technologies, we are now able to access enormous amount of genomic sequencing data, which encodes a treasure chest of new enzymatic chemistry. The challenge now is to devise a method to efficiently identify chemically interesting enzymes from this vast pool of information.

One possible solution to this problem is to study the biosynthetic pathways of structurally unique natural products, which are predicted to involve novel enzymatic reactions. We aimed to discover new enzymes that catalyze intriguing chemical reactions through biosynthetic investigation guided by our knowledge in organic chemistry. The cylindrocyclophanes are a family of natural products that contain an unusual [7.7]paracyclophane core scaffold.¹ Based on the results of the previous feeding studies, cylindrocyclophane biosynthesis is predicted to involve an unusual C–C bond formation (Figure 1).² To discover the enzymes responsible for this chemistry, we studied the biosynthesis of the cylindrocyclophanes.

Figure 1. The structures of the cylindrocyclophanes. The predicted biosynthetic disconnection suggests that an unusual C–C bond formation is involved in their biosynthesis.

First, the candidate cylindrocyclophane biosynthetic (cyl) gene cluster was identified from the genomic sequence of the cylindrocyclophane producer, Cylindrospermum licheniforme ATCC 29412. We next formulated a biosynthetic hypothesis based on the cyl gene cluster annotation (Figure 2). In our original biosynthetic hypothesis, we predicted that cylindrocyclophane biosynthesis initiates with the activation of decanoic acid by the fatty acid activating enzymes, CylA and CylB, to form decanoyl-CylB. The activated decanoyl-CylB is then processed by the type I polyketide synthase (PKS) machinery, CylD-H. The nascent polyketide is released from the type I PKS assembly line by the type III PKS CylI to form the alkylresorcinol, which is the predicted monomeric unit of the cylindrocyclophanes.

Figure 2. The initial biosynthetic hypothesis for cylindrocyclophane assembly.

Based on our initial biosynthetic hypothesis, we biochemically characterized the functions of the fatty acid activating enzymes CylA/CylB and the type III PKS CylI. The in vitro activities of these three enzymes were consistent with our biosynthetic hypothesis, which validated the involvement of the cyl gene cluster in cylindrocyclophane production.³ In addition, we conducted feeding experiments using deuterium-labeled decanoic acid in the native producer to confirm that decanoic acid is a precursor to the cylindrocyclophanes. The incorporation of deuterium-labeled decanoic acid into the final cylindrocyclophane scaffold also indicated that cylindrocyclophane biosynthesis involves functionalization of the unactivated carbon center.³

Following our discovery and validation of the cyl gene cluster, we next focused on the investigation of the key C–C bond formation that results in the construction of the [7.7]paracyclophane scaffold. Through bioinformatics and biochemical characterizations of the enzymes encoded in the cyl gene cluster, we determined that cylindrocyclophane biosynthesis involv

(View PDFfor the rest of the abstract.)

Nature constructs structurally diverse, bioactive molecules using enzymes. Many enzymes catalyze synthetically challenging reactions under mild, physiological conditions. Consequently, they have long been a source of inspiration for developing biomimetic organic syntheses and methods. In addition, enzymes are increasingly being used as biocatalysts in industry. Therefore, the discovery of enzymes that catalyze chemically intriguing transformations can positively impact synthesis in multiple ways. With the recent advances in next-generation DNA sequencing technologies, we are now able to access enormous amount of genomic sequencing data, which encodes a treasure chest of new enzymatic chemistry. The challenge now is to devise a method to efficiently identify chemically interesting enzymes from this vast pool of information.

One possible solution to this problem is to study the biosynthetic pathways of structurally unique natural products, which are predicted to involve novel enzymatic reactions. We aimed to discover new enzymes that catalyze intriguing chemical reactions through biosynthetic investigation guided by our knowledge in organic chemistry. The cylindrocyclophanes are a family of natural products that contain an unusual [7.7]paracyclophane core scaffold.¹ Based on the results of the previous feeding studies, cylindrocyclophane biosynthesis is predicted to involve an unusual C–C bond formation (Figure 1).² To discover the enzymes responsible for this chemistry, we studied the biosynthesis of the cylindrocyclophanes.

Figure 1. The structures of the cylindrocyclophanes. The predicted biosynthetic disconnection suggests that an unusual C–C bond formation is involved in their biosynthesis.

First, the candidate cylindrocyclophane biosynthetic (cyl) gene cluster was identified from the genomic sequence of the cylindrocyclophane producer, Cylindrospermum licheniforme ATCC 29412. We next formulated a biosynthetic hypothesis based on the cyl gene cluster annotation (Figure 2). In our original biosynthetic hypothesis, we predicted that cylindrocyclophane biosynthesis initiates with the activation of decanoic acid by the fatty acid activating enzymes, CylA and CylB, to form decanoyl-CylB. The activated decanoyl-CylB is then processed by the type I polyketide synthase (PKS) machinery, CylD-H. The nascent polyketide is released from the type I PKS assembly line by the type III PKS CylI to form the alkylresorcinol, which is the predicted monomeric unit of the cylindrocyclophanes.

Figure 2. The initial biosynthetic hypothesis for cylindrocyclophane assembly.

Based on our initial biosynthetic hypothesis, we biochemically characterized the functions of the fatty acid activating enzymes CylA/CylB and the type III PKS CylI. The in vitro activities of these three enzymes were consistent with our biosynthetic hypothesis, which validated the involvement of the cyl gene cluster in cylindrocyclophane production.³ In addition, we conducted feeding experiments using deuterium-labeled decanoic acid in the native producer to confirm that decanoic acid is a precursor to the cylindrocyclophanes. The incorporation of deuterium-labeled decanoic acid into the final cylindrocyclophane scaffold also indicated that cylindrocyclophane biosynthesis involves functionalization of the unactivated carbon center.³

Following our discovery and validation of the cyl gene cluster, we next focused on the investigation of the key C–C bond formation that results in the construction of the [7.7]paracyclophane scaffold. Through bioinformatics and biochemical characterizations of the enzymes encoded in the cyl gene cluster, we determined that cylindrocyclophane biosynthesis involves a cryptic chlorination event. We revealed that CylC, an enzyme annotated as a “hypothetical protein”, is a new halogenating enzyme that catalyzes chlorination of the aliphatic carbon center of decanoyl-CylB (Figure 3). The chloride atom installed by CylC serves as a leaving group required for the eventual C–C bond formation in paracyclophane construction. Bioinformatic analysis revealed that CylC resembles ferritin-like di-metallo carboxylate enzymes. Further bioinformatic search showed that CylC homologs are widely distributed among cyanobacteria, and these halogenases are likely responsible for the halogenation of multiple known chlorinated natural products.⁴ We confirmed that CylC catalyzes chlorination of decanoyl-CylB in vivo and obtained the first biochemical evidence supporting the halogenase function of CylC.

Figure 3. Chlorination of decanoyl-CylB catalyzed by the novel halogenase CylC.

In addition to the new halogenase CylC, we discovered a new alkylating enzyme CylK that catalyzes a Friedel–Crafts-type C–C bond formation. CylK is a calcium-dependent enzyme that lacks primary sequence homology to known proteins. CylK catalyzes the final dimerization step in cylindrocyclophane biosynthesis to construct the [7.7]paracyclophane scaffold using chlorinated resorcinol substrates (Figure 4). Further biochemical characterizations revealed that CylK requires calcium for catalysis and that this enzyme has extremely high stereoselectivity toward its substrate. The activity of CylK confirmed that the biosynthesis of the cylindrocyclophanes proceeds through cryptic chlorination. To our knowledge, CylK is the first enzyme that has been demonstrated to use secondary alkyl chloride as an electrophile to alkylate nucleophilic aromatic rings. In addition, the expanded substrate scope of CylK indicates that this enzyme is a promising candidate for future bioengineering efforts to develop useful biocatalysts for Friedel–Crafts-type alkylation.

Figure 4. Friedel–Crafts-type alkylation catalyzed by CylK leads to the construction of the [7.7]paracyclophane scaffold in the cylindrocyclophanes.

Through our investigation of cylindrocyclophane biosynthesis, we succeeded in discovering a new halogenase, CylC, and a new alkylating enzyme, CylK. With the discovery of these two enzymes, we have completely elucidated the biosynthesis of the unusual [7.7]paracyclophane scaffold in the cylindrocyclophanes. Moving forward, we are interested in determining the structures and the mechanisms of these two novel enzymes. In addition, we will complete the substrate scope screen of CylK and explore the possibility of developing CylK into a useful biocatalyst through bioengineering efforts. Overall, our work demonstrated the usefulness of employing chemically guided approach to study the biosynthesis of structurally unique natural products and to efficiently identify novel enzymes that catalyze intriguing chemistry.

References:

1. Moore, B. S. et al. J. Am. Chem. Soc. 112, 4061-4063 (1990).

2. Bobzin, S. C.; Moore, R. E. Tetrahedron 49, 7615-7626 (1993).

3. Nakamura, H.; Hamer, H. A.; Sirasani, G.; Balskus, E. P. J. Am. Chem. Soc. 134, 18518-18521 (2012).

4. a) Ishida, K. et al. Tetrahedron 54, 13475-13484 (1998). b) Leao, P. N. et al. Angew. Chem. Int. Ed. 54, 11063-11067 (2015). c) Kleigrewe, K. et al. J. Nat. Prod. 78, 1671-1682 (2015).