Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Sep 17.
Published in final edited form as: J Am Chem Soc. 2008 Aug 21;130(37):12236–12237. doi: 10.1021/ja8044162

The cis2,3-double bond of phoslactomycins is generated by a post-PKS tailoring enzyme

Nadaraj Palaniappan 1, Mamoun M Alhamadsheh 1, Kevin A Reynolds 1,*
PMCID: PMC2644895  NIHMSID: NIHMS90919  PMID: 18714992

Abstract

The antifungal phoslactomycins (PLM A-F), produced by Streptomyes sp HK803, are structurally unusual in that three of their four double bonds are in the cis form (Δ12,13, Δ14,15, Δ2,3). The PLM polyketide synthase (PKS) has the predicted dehydratase catalytic domain in modules 1,2 and 5 required for establishing two of these cis double bonds (Δ12,13Δ14,15), as well as the only trans Δ6,7double bond. By contrast, the formation of the cis Δ2,3 in the unsaturated lactone moiety of PLMs has presented an enigma because the predicted dehydratase domain in module 7 is absent. Herein, we have demonstrated that the plmT2 gene product, with no homology to PKS dehydratase domains, is required for efficient formation of the cis Δ2,3 alkene. A series of new PLM products in which the C3 hydroxyl group is retained, are made in plmT2 deletion mutants. In all of these cases, however, the hydroxyl group is esterified with malonic acid. These malonylated PLM products are converted to the corresponding cis Δ2,3 PLM products and acetic acid by a facile base-catalyzed decarboxylative elimination reaction. Complete or partial restoration of natural PLM production in a plmT2 deletion mutant can be accomplished by plasmid based expression of plmT2 or fos ORF4 (a homologous gene from the fostriecin biosynthetic gene cluster), respectively. The data indicate that dehydratase-independent pathways also function in establishment of unsaturated 6-membered lactone moieties in other PKS pathways, and provide the first biosynthetic insights into the possible routes by which unusual malonylated polyketide products are generated.


Modular polyketide synthases (PKSs) generate a vast array of structurally diverse natural products with important biological activities.1 They are responsible for formation of macrolides, the vast majority of which contain one or more double bonds.2 The majoritiy of other natural produces made by modular PKSs similarly contain some double bonds. In the majority of cases the alkenes in the polyketide product are in the trans form. The formation of these double bonds has been shown to be directly attributed to the presence of a ketoreductase-dehydratase (KR-DH) didomain within the appropriate module.3, 4 These didomains catalyze a 3-keto reduction and subsequent Δ2,3 elimination reaction with the PKS-tethered 3-ketoacyl polyketide intermediates.

The phoslactomycins (PLMs A-F Figure 1) and fostriecin belong to a class of phosphorylated polyketides that contain multiple double bonds in the cis form.5, 6 For the PLMs there are three double bonds in the cis form (Δ12,13, Δ14,15, Δ2,3) and one in the trans form (Δ6,7).7, 8

Figure 1.

Figure 1

Proposed biosynthetic roles of module 6 and 7 of the PLM PKS and structures of PLM products made by Streptomyces sp. HK803 and its NP derivatives.

The PLM biosynthetic gene cluster has been cloned and sequenced and shown to encode a modular PKS.9 Modules 1 and 2 contain the expected dehydratase DH-KR didomain required for formation of the conjugated diene, while module 5 contains a DH-KR domain likely responsible for formation of the trans Δ6,7 alkene. The KR-DH domains which generate a trans double bond do so via a D-3-hydroxyacyl intermediate (determined by the KR domain).2-4 The L-3-hydroxyacyl product is the speculated intermediate in KR-DH domains which generate cis double bonds.3, 5 A bioinformatic analyses of the KR domain in module 7 suggests it may generate a l-3-hydroxyacyl intermediate. However, there is no cognate DH domain (Figure 1). The DH domain is also absent in the respective terminal PKS modules involved in biosynthesis of fostriecin and leptomycin,10 related natural products with an unsaturated lactone moiety. We posited that a DH-independent post-PKS enzymatic catalyzed reaction might be involved in the formation of all of these cis Δ2,3 unsaturated lactone moieties. We have identified plmT2 (and its homolog ORF 4 from the fostriecin biosynthetic gene cluster) as being required for efficient formation of this alkene.

The predicted PlmT2 peptide sequence9 shows up to 50% sequence similarity to a series of a family of putative NAD dependent epimerase/dehydratases. We generate NP7, a plmT2 + plmS2 deletion mutant. It has been shown that a plmS2 deletion mutant (NP2) generates exclusively PLM B9 (Figure 2) and that PlmS2 catalyzes C18-hydroxylation of PLM B (subsequent acylation then provides the final PLM products, PLM A and PLM C-F.11 Thus a PLM B analog with a saturated lactone moiety and a C-3 hydroxyl substituent was one of the possible products generated by the NP7 mutant. Indeed, HPLC analyses of the fermentation broth of this mutant (Figure 2) revealed a major new and more hydrophilic PLM product, which was purified following standard protocols12 and shown to contain the predicted saturated lactone and C-3 hydroxyl substituent. However, spectroscopic analyses (1H- and 13C-NMR, and MS) revealed that the C-3 hydroxyl substituent was esterified with malonic acid. This new malonylated PLM B (M-PLM B, Figure 1), is unstable in mild basic conditions and rapidly undergoes an elimination reaction to provide PLM B and either malonic acid or acetic acid (acetic acid is observed by GC-MS analysis of the degradation products, but it remains to be determined at which stage the decarboxylation occurs). In contrast, saturated lactone compounds with a C-3 hydroxyl substituent have been generated both in vivo and in vitro by a modified form of DEBS PKS,13 and shown to be stable (there are no reports of formation of the corresponding unsaturated product). The low stability of M-PLM B is the most likely reason for the small levels of PLM B in the fermentation media of NP7 (Figure 2). PLM analogs or pathway intermediates in which the C-3 hydroxyl was not malonylated were not detected in NP7 fermentations, suggesting malonylation is an efficient process and may represent part of the normal biosynthetic process, facilitating the eventual formation of the unsaturated lactone. Thus one possible role for PlmT2 is catalysis of a decarboxylative-elimination reaction with a malonylated polyketide pathway intermediate (Figure 3). Elucidation of the specific role of PlmT2 in catalyzing formation of the unsaturated lactone remains undetermined, as does the question of which enzyme or catalytic domain is responsible for malonylation of the C-3 hydroxyl residue. Sequence analysis does not reveal any discrete AT proteins in the PLM gene cluster raising an intriguing possibility that AT in module 7 is responsible for malonylation of both the cognate ACP for the elongation process and the C-3 hydroxyl of the resulting extended PLM structure when it is attached to this ACP. We note that other polyketide natural products such as azalomycin F,14 malolactomycin A,15 shurimycins A and B16 contain malonyl esters and are almost certainly generated by modular PKSs. How these malonyl esters are formed has not been addressed, but it seems likely that there may be a similar malonylation process to that occurring in the PLM pathway. Production of M-PLM B indicates that modification steps (C-8 hydroxylation, phosphorylation of the C-9 hydroxyl and introduction of the C-25 amine functionality) of a PLM core skeleton can occur with the polyketide chain bearing a malonyl ester at the C-3 position. It has been shown that PlmS2 catalyzes the C-18 hydroxylation of PLM B.17 To test if PlmS2 can catalyze hydroxylation of the M-PLM B we generated an NP11 (ΔplmT2) mutant in the wild strain (in which plmS2 is present). In this mutant we observed formation of malonylated derivatives of PLM A, and PLM C-F (Figure 1), demonstrating that C-18 hydroxylation and subsequent O-18 esterification can occur with M-PLM B

Figure 2.

Figure 2

HPLC analyses of the fermentation broths of NP2 (ΔplmS2), NP7 (ΔplmT2 and ΔplmS2), NP7/pNS5 and NP7/pNS6 (PlmT2 and fos ORF4 complementation plasmids, respectively).

Figure 3.

Figure 3

One possible role for PlmT2 in catalyzing a decarboxylative elimination reaction of either M-PLM B or a pathway intermediate.

A set of complementation experiments were carried out in the NP7 strain, using expression plasmids for plmT2 (pNS5) and the plmT2 homolog fos ORF4 (pNS6). HPLC analyses (Figure 2) showed a complete restoration of PLM B with plmT2 complementation (NP7/pNS5 strain), and almost complete (95% PLM B + 5% M-PLM B) for the complementation with the fos ORF4 (NP7/pNS6 strain). These data supports the proposed role of PlmT2 in formation of the unsaturated lactone of the PLMs and indicates that a similar process of malonylation and decarboxylative elimination may occur in the fostriecin biosynthetic process.

The unsaturated lactone of cytostatin and fostriecin plays an important role in the potent and selective activity (IC50 of 1-3 nM) against protein phosphatase 2A (PP2A). This Michael acceptor is proposed to be required for formation of a covalent adduct with Cys269, unique to PP2A.18 In vitro assays have shown that various PLMs are selective but poorer inhibitors of PP2A (IC50 values 3-40 μM).19 Nonetheless, PLM A targets PP2A in human fibrosarcoma cells HT1080 cells 19 and Cys 269 on the catalytic residue of PP2A is essential for this interaction. Surprisingly, using a PP2A inhibitor activity studies under standard assay conditions 19 we observed that the M-PLM B was slightly more effective (IC50 =12.1 ± 1.3 μM) than PLM B (IC50 =36.7 ± 4.0 μM) against PP2A (there was no decomposition of M-PLM-B to PLM B under the assay conditions). A more detailed investigation of the biological activity of the M-PLM analogs is underway.

In summary, efficient formation of unsaturated lactone in the PLMs and other related compounds has been shown to be dependent upon an enzyme such as PlmT2, not a canonical DH domain in the modular PKS. Without PlmT2, new PLM analogs bearing a C-3 hydroxyl group esterified with malonic acid are generated. The M-PLM B, under mild basic conditions, undergoes rapid elimination to generate the cis Δ2,3 alkene of the PLMs in a process that may reflect that catalyzed by PlmT2. This catalytic process and formation of malonyl esters of polyketide products present new and as yet unanswered questions pertaining to processes catalyzed by modular PKSs.

Supplementary Material

1_si_001. Supporting Information Available.

Experimental procedures, spectroscopic data. This material is available free of charge via the Internet at http://pubs.acs.org.

Acknowledgments

Funding for this research was supported by National Institute of Health (A151629).

References

  • 1.Staunton J, Weissman KJ. Nat Prod Rep. 2001;18:380–416. doi: 10.1039/a909079g. [DOI] [PubMed] [Google Scholar]
  • 2.Wu J, Zaleski TJ, Valenzano C, Khosla C, Cane DE. J Am Chem Soc. 2005;127:17393–404. doi: 10.1021/ja055672+. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Caffrey P. Chembiochem. 2003;4:654–7. doi: 10.1002/cbic.200300581. [DOI] [PubMed] [Google Scholar]
  • 4.Reid R, Piagentini M, Rodriguez E, Ashley G, Viswanathan N, Carney J, Santi DV, Hutchinson CR, McDaniel R. Biochemistry. 2003;42:72–9. doi: 10.1021/bi0268706. [DOI] [PubMed] [Google Scholar]
  • 5.Alhamadsheh MM, Palaniappan N, Daschouduri S, Reynolds KA. J Am Chem Soc. 2007;129:1910–1. doi: 10.1021/ja068818t. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Stampwala SS, Bunge RH, Hurley TR, Willmer NE, Brankiewicz AJ, Steinman CE, Smitka TA, French JC. J Antibiot. 1983;36:1601–5. doi: 10.7164/antibiotics.36.1601. [DOI] [PubMed] [Google Scholar]
  • 7.Fushimi S, Furihata K, Seto H. J Antibiot (Tokyo) 1989;42:1026–36. doi: 10.7164/antibiotics.42.1026. [DOI] [PubMed] [Google Scholar]
  • 8.Fushimi S, Nishikawa S, Shimazu A, Seto H. J Antibiot (Tokyo) 1989;42:1019–25. doi: 10.7164/antibiotics.42.1019. [DOI] [PubMed] [Google Scholar]
  • 9.Palaniappan N, Kim BS, Sekiyama Y, Osada H, Reynolds KA. J Biol Chem. 2003;278:35552–7. doi: 10.1074/jbc.M305082200. [DOI] [PubMed] [Google Scholar]
  • 10.Hu Z, Reid R, Gramajo H. J Antibiot (Tokyo) 2005;58:625–33. doi: 10.1038/ja.2005.86. [DOI] [PubMed] [Google Scholar]
  • 11.Ghatge M, Palaniappan N, Das Choudhuri S, Reynolds K. J Ind Microbiol Biotechnol. 2006;33:589–99. doi: 10.1007/s10295-006-0116-1. [DOI] [PubMed] [Google Scholar]
  • 12.Choudhuri SD, Ayers S, Soine WH, Reynolds KA. J Antibiot (Tokyo) 2005;58:573–82. doi: 10.1038/ja.2005.78. [DOI] [PubMed] [Google Scholar]
  • 13.Kim BS, Cropp TA, Florova G, Lindsay Y, Sherman DH, Reynolds KA. Biochemistry. 2002;41:10827–33. doi: 10.1021/bi0256779. [DOI] [PubMed] [Google Scholar]
  • 14.Chandra A, Nair MG. J Antibiot (Tokyo) 1995;48:896–8. [PubMed] [Google Scholar]
  • 15.Kobinata K, Koshino H, Kusakabe H, Kobayashi Y, Yamaguchi I, Isono K, Osada H. J Antibiot (Tokyo) 1993;46:1912–5. doi: 10.7164/antibiotics.46.1912. [DOI] [PubMed] [Google Scholar]
  • 16.Kumazawa S, Asami Y, Awane K, Ohtani H, Fukuchi C, Mikawa T, Hayase T. J Antibiot (Tokyo) 1994;47:688–96. doi: 10.7164/antibiotics.47.688. [DOI] [PubMed] [Google Scholar]
  • 17.Ghatge MS, Reynolds KA. J Bacteriol. 2005;187:7970–7976. doi: 10.1128/JB.187.23.7970-7976.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Lawhorn BG, Boga SB, Wolkenberg SE, Colby DA, Gauss CM, Swingle MR, Amable L, Honkanen RE, Boger DL. J Am Chem Soc. 2006;128:16720–32. doi: 10.1021/ja066477d. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Teruya T, Simizu S, Kanoh N, Osada H. FEBS Lett. 2005;579:2463–8. doi: 10.1016/j.febslet.2005.03.049. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1_si_001. Supporting Information Available.

Experimental procedures, spectroscopic data. This material is available free of charge via the Internet at http://pubs.acs.org.

RESOURCES