Abstract
The pyrroloquinoline alkaloid family of natural products that includes the immunosuppressant lymphostin has long been postulated to arise from tryptophan. We now report the molecular basis of lymphostin biosynthesis in three marine Salinispora species that maintain conserved biosynthetic gene clusters harboring a hybrid nonribosomal peptide synthetase-polyketide synthase central to lymphostin assembly. Through a series of experiments involving gene mutations, stable isotope profiling, and natural product discovery, we report the assembly line biosynthesis of lymphostin and nine new analogues that exhibit potent mTOR inhibitory activity.
Since the discovery of discorhabdin C in 1986,1 a group of cytotoxic natural products has emerged that contain a pyrroloquinoline core (Figure 1). Most members of this diverse family have been isolated from marine sponges2 and include the batzellines, damirones, and makaluvamines.3 More recently, compounds of this class have been obtained from terrestrial organisms including makaluvamine A from a myxomycete4 and the sanguinones and mycenarubins from mushrooms of the genus Mycena.5 Variations in the structural complexity of these compounds arise from substitutions about the pyrrolo[4,3,2-de]quinoline core, which is thought to be derived from tryptophan. However, despite the large number of compounds currently known, biosynthetic studies have been hampered due to the uncultivability and genetic intractability of the producer organisms.
Figure 1.
Structures of pyrrolo[4,3,2-de]quinoline natural products
To date, two natural products containing a pyrrolo[4,3,2-de]quinoline core have been isolated from bacteria, the myosin-targeting ammosamides from Streptomyces strain CNR-6986 and the immunosuppressant lymphostin (1) that was first isolated from Streptomyces sp. KY11783.7 Lymphostin has been shown to inhibit both lymphocyte kinase (IC50 0.05 µM) and phosphatidylinositol 3-kinase (IC50 0.001 µM).8 This biological activity, along with lymphostin’s tricyclic pyrroloquinoline skeleton, inspired a biomimetic total synthesis starting from tryptophan in 2004.9 We report here the molecular basis for lymphostin (1) biosynthesis in Salinispora spp. which involves a uniquely organized modular synthetase that through gene targeting provided the novel analogue lymphostinol (2). We also report eight new N-acyl derivatives including the potent mTOR inhibitor neolymphostin A (3).
Genome sequencing of the actinomycetes Salinispora tropica strain CNB-440 and Salinispora arenicola strain CNS-205 revealed that these marine bacteria have the genetic capacity to produce a wide variety of structurally and biologically diverse natural products.10 While the majority of their secondary metabolism genes are relatively unique, they share at least a dozen natural product gene clusters. Amongst these is the hydrid nonribosomal peptide synthetase-polyketide synthase (NRPS-PKS) locus lym (Stro3051–3057, Sare3277–3283) tentatively assigned to encode the biosynthesis of 1 (Figure 2A).10 The lym cluster is maintained in a third species, “Salinispora pacifica”, including strain DPJ-0019 where its gene organization is conserved (Table S1, Supporting Information [SI]).
Figure 2.
Organization of the lym gene cluster and proposed biosynthesis of lymphostin (1), lymphostinol (2), and related N-acyl derivatives 3–10. A. The putative lym gene cluster is colored blue (including genes lymA and lymB that were inactivated in this study) with adjacent open reading frames (ORFs) in gray that are associated with ribosomal peptide biosynthesis. Strop gene numbers are shown flanking the cluster. Each arrow represents the direction of transcription of an ORF. See the SI for the deduced functions of the ORFs and a comparison of the lym gene products (Tables S2 and S3, respectively). B. Neolymphostin A (3) biosynthesis was probed in “S. pacifica” strain DPJ-0019 by isotope labeling with 15N2-L-tryptophan, 15N-L-glutamine and 13C4-isobutyrate as shown. Abbreviations: A, adenylation; ACP, acyl carrier protein; AT, acyltransferase; KS, ketosynthase; MT, methyltransferase; NAT, N-acyltransferase; PCP, peptidyl carrier protein; R, reductase.
Bioinformatic analysis of the syntenic lym loci in S. tropica, S. arenicola and “S. pacifica” established the approximate physical boundaries of the gene cluster that includes 7 genes covering 14.8-kb (Figure 2A). The central lymA gene encodes a heptadomain synthetase that putatively functions to malonate extend a tryptophan-derived peptidyl carrier protein (PCP)-bound residue and reductively release an N-acetylated diketide product as an aldehyde from an acyl carrier protein (ACP)-bound intermediate (Figure 2B). Trans methylation by the LymB methyltransferase (MT) would preserve the oxidation state of the released aldehyde as a methoxy enolate characteristic of 1 as a product of an atypical hybrid NRPS-PKS. Further sequence analysis of the LymA adenylation (A) domain revealed that the invariant Asp residue, which is required to stabilize the α-amino group of an amino acid substrate,11 is replaced by Asn 226 that may activate an aryl acid substrate which may already contain the pyrrolo[4.3.2-de] core.
In order to explore this biosynthetic model of 1 assembly, we first established lymphostin production in Salinispora independently at UCSD and Pfizer. Standard growth conditions in A1 media that yielded products such as the salinosporamide proteasome inhibitors in S. tropica12 and the cyclomarin peptides in S. arenicola,13 however, did not yield 1. At UCSD, we explored different fermentation methods leading to a lymphostin production media that ultimately gave 1 in S. tropica and S. arenicola at ~0.4 mg/L and ~3 mg/L, respectively. We additionally observed a new derivative in S. arenicola with the molecular composition C15H14N4O3 as established by high-resolution electrospray ionization mass spectrometry (m/z [M+H]+ 299.1141 obs, 0.7 ppm error). Analysis of the 1H and 13C NMR spectra, with the aid of gradient-enhanced HMBC data, clearly indicated that lymphostinol (2) was a novel lymphostin derivative with a modified C-4 side chain in which the β-methoxy enone moiety of 1 was replaced with a β-hydroxy ketone residue. Key HMBC signals connected the C-10/C-11 ethyl alcohol residue to the C-9 carbonyl at 200.2 ppm, which was supported by the difference in MS data of a single carbon atom between 1 and 2.
In the Pfizer laboratory, 1 production conditions were first established with S. arenicola strain MOSO-0003 in 12 different media that ultimately gave a titer of ~20 mg/L in the medium M48-9. Subsequently, five of 11 different marine-invertebrate-derived strains of “S. pacifica” were found to produce 1. Although most strains, such as DPJ-0024, produced 1 alone, strain DPJ-0019 was found to produce a mixture of lymphostin analogues. To improve the fermentation profile and facilitate downstream processing of material, a simpler production medium was developed containing only soluble starch and yeast extract (YESS). For comparative HPLC analysis of 1 production in strains from all three species, S. tropica CNB-440, S. arenicola CNS-205, “S. pacifica” DPJ-0024 and DPJ-0019 were fermented in medium YESS at various scales, processed and analyzed in parallel (Figure 3).
Figure 3.
Comparative HPLC analysis (467 nm) of crude extracts from “S. pacifica” DPJ-0019, S. arenicola CNS-205, “S. pacifica” DPJ-0024, and S. tropica CNB-440 fermentations. Lymphostin (1), lymphostinol (2), neolymphostins A (3), B (4), C (5) & D (6), neolymphostinols A (7) B (8) C (9) & D (10) are noted. See Table S6 for titers.
Analysis of the comparative lym chemistry of the three strains revealed that S. tropica CNB-440 produced 1 alone, S. arenicola CNS-205 and “S. pacifica” DPJ-0024 produced 1, 2, 4, and 8, and “S. pacifica” DPJ-0019 uniquely produced neolymphostins 3, 5, and 6 and the corresponding neolymphostinols 7, 9, and 10 (Figure 3, Table S6). In the neolymphostins and neolymphostinols, the C-13 N-acetyl found in 1 and 2 is replaced by larger acyl groups, such as isobutyryl, propyl, sec-pentyryl, and isopentyryl (Figure 2B). All eight derivatives were isolated and fully characterized by HRMS and NMR analysis (Table S4). The most abundant member was neolymphostin A (3), whose 1H and 13C NMR spectra showed the presence of a 2-propyl amide moiety (δ3.06(1H, m)/δ34.7 and δ1.22(6H,d)/δ18.9) in place of the acetamide in 1. Lymphostin and the neolymphostins showed potent mTOR inhibition and cytoxicity (Table 1), whereas the neolymphostinols as a whole were 1000-fold less active than their neolymphostin counterparts (Table S5).
Table 1.
mTor Kinase inhibition and cytotoxicity of lymphostin and neolymphostins
Compound | mTOR (nM) | LNCap (nM) | MDA-468 (nM) |
---|---|---|---|
1 | 1.7 | 38 | 14 |
3 | 0.8 | 22 | 58 |
4 | 1.5 | 48 | 85 |
5 | 1.8 | 230 | 700 |
6 | 1.8 | 230 | 700 |
Buoyed by the characterization of lymphostin chemistry, we next interrogated the lym clusters by PCR-targeted gene replacement recently optimized for Salinispora genetics.13,14 Inactivation of the conserved lymA gene resulted in S. tropica and S. arenicola mutants devoid of 1 and 2 (Figure S1), thereby confirming the central role of the LymA synthetase. To further establish the biosynthetic relationship between 1 and 2, we inactivated the S. arenicola lymB MT. On the basis of biosynthetic precedence,15 we anticipated that the resulting mutant would accumulate an aldehyde intermediate that would lead to further carbonyl reduction by the LymA reductase in the absence of LymB (Figure 2B). Chemical analysis of the lymB::aprR mutant revealed the complete loss of 1 and increased production of 2, thereby confirming their biosynthetic linkage (Figure S1).
While the biosynthetic endgame is understood, we have yet to clarify the early stages of 1 biosynthesis that includes the formation of the pyrrolo[4,3,2-de]quinoline core structure. We established its biogenesis in “S. pacifica” strain DPJ-0019 from tryptophan through isotope experiments in which both nitrogen atoms were retained in the lymphostin products at N-1 and N-5 (Figure 2B). FTMS/MS analysis of 15N2-L-tryptophan-derived 3 revealed both mono (21%) and di (41%) 15N-labeling. The remaining nitrogen atoms in 3 were likewise shown to derive from glutamine, providing a possible role of the conserved amidotransferase encoded by Strop_3057 in adding the aryl amine groups.
The molecular basis for the larger diversity of analogues produced by “S. pacifica” strain DPJ-0019 likely relies on the different substrate discrimination of the unusual LymA N-acetyltransferase (NAT) domain. As expected in the case of 3, we confirmed that the N-isobutyryl group is derived from isobutyrate through 13C isotope labeling (Figure 2B). While NATs are common biosynthetic enzymes, its inclusion in a modular synthetase is to the best of our knowledge unprecedented. We are presently studying the molecular basis behind this difference in substrate specificity that offers an exciting opportunity to explore acyl group discrimination and to biosynthetically engineer new lymphostin and other pyrroloquinoline alkaloids.
In summary, we established the molecular basis of lymphostin biosynthesis in three Salinispora species by an unusual hybrid nonribosomal peptide-polyketide synthetase pathway that for the first time sheds light on the formation of pyrrolo[4,3,2-de]quinoline group alkaloids. Our work also resulted in the discovery of nine new lymphostin derivatives and the preliminary description of their potent mTOR inhibitory activity.
Supplementary Material
Acknowledgment
We gratefully acknowledge Dr. Yongxuan Su of the UCSD Mass Spectrometry Facility for MS assistance, Jan Kieleczawa and Tony Li at Pfizer’s BioCore facility (Cambridge, MA) for DNA sequencing of lym genes in DPJ-0019, Dr. Nobuhiro Fusetani at Hokkaido University for providing D. proliferum tissue from which DPJ-0019 was isolated, and Dr. Ker Yu for biological assays. This work was generously supported in part by a research grant from the NIH (GM085770) to P.R.J. and B.S.M., and a postdoctoral fellowship from the JSPS to A.M.
Footnotes
Supporting Information Available: FiguresS1–S4, Tables S1–S6, experimental methods, HRMS data, NMR spectra. This material is available free of charge via the Internet at http://pubs.acs.org.
References
- 1.Perry NB, Blunt JW, McCombs JD, Munro MHG. J. Org. Chem. 1986;51:5476–5478. [Google Scholar]
- 2.Antunes EM, Copp BR, Davies-Coleman MT, Samaai T. Nat. Prod. Rep. 2005;22:62–72. doi: 10.1039/b407299p. [DOI] [PubMed] [Google Scholar]
- 3.Sakemi S, Sun HH, Jefford CW, Bernardinelli G. Tetrahedron Lett. 1989;30:2517–2520. [Google Scholar]; Chang LC, Otero-Quintero S, Hooper JNA, Bewley CA. J. Nat. Prod. 2002;65:776–778. doi: 10.1021/np010581l. [DOI] [PubMed] [Google Scholar]; Stierle DB, Faulkner DJ. J. Nat. Prod. 1991;54:1131–1133. doi: 10.1021/np50076a039. [DOI] [PubMed] [Google Scholar]; Schmidt EW, Harper MK, Faulkner DJ. J. Nat. Prod. 1995;58:1861–1867. doi: 10.1021/np50126a008. [DOI] [PubMed] [Google Scholar]; Radisky DC, Radisky ES, Barrows LR, Copp BR, Kramer RA, Ireland CM. J. Am. Chem. Soc. 1993;115:1632–1638. [Google Scholar]
- 4.Ishibashi M, Iwasaki T, Imai S, Sakamoto S, Yamaguchi K, Ito A. J. Nat. Prod. 2000;64:108–110. doi: 10.1021/np000382m. [DOI] [PubMed] [Google Scholar]
- 5.Peters S, Spiteller P. J. Nat. Prod. 2007;70:1274–1277. doi: 10.1021/np070179s. [DOI] [PubMed] [Google Scholar]; Peters S, Spiteller P. Eur. J. Org. Chem. 2007:1571–1576. [Google Scholar]
- 6.Hughes CC, MacMillan JB, Gaudêncio SP, Fenical W, La Clair JJ. Angew. Chem. Int. Ed. 2009;48:728–732. doi: 10.1002/anie.200804107. [DOI] [PMC free article] [PubMed] [Google Scholar]; Hughes CC, MacMillan JB, Gaudêncio SP, Jensen PR, Fenical W. Angew. Chem. Int. Ed. 2009;48:725–727. doi: 10.1002/anie.200804890. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Nagata H, Ochiai K, Aotani Y, Ando K, Yoshida M, Takahashi I, Tamaoki T. J. Antibiot. 1997;50:537–542. doi: 10.7164/antibiotics.50.537. [DOI] [PubMed] [Google Scholar]; Aotani Y, Nagata H, Yoshida M. J. Antibiot. 1997;50:543–545. doi: 10.7164/antibiotics.50.543. [DOI] [PubMed] [Google Scholar]
- 8.Nagata H, Yano H, Sasaki K, Sato S, Nakanishi S, Takahashi I, Tamaoki T. Biosci. Biotechnol. Biochem. 2002;66:501–507. doi: 10.1271/bbb.66.501. [DOI] [PubMed] [Google Scholar]
- 9.Tatsuta K, Imamura K, Itoh S, Kasai S. Tetrahedron Lett. 2004;45:2847–2850. [Google Scholar]
- 10.Udwary DW, Zeigler L, Asolkar RN, Singan V, Lapidus A, Fenical W, Jensen PR, Moore BS. Proc. Natl. Acad. Sci. U.S.A. 2007;104:10376–10381. doi: 10.1073/pnas.0700962104. [DOI] [PMC free article] [PubMed] [Google Scholar]; Penn K, Jenkins C, Nett M, Udwary DW, Gontang EA, McGlinchey RP, Foster B, Lapidus A, Podell S, Allen EE, Moore BS, Jensen PR. ISME J. 2009;3:1193–1203. doi: 10.1038/ismej.2009.58. [DOI] [PMC free article] [PubMed] [Google Scholar]; Nett M, Ikeda H, Moore BS. Nat. Prod. Rep. 2009;26:1362–1384. doi: 10.1039/b817069j. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.May JJ, Kessler N, Marahiel MA, Stubbs MT. Proc. Natl. Acad. Sci. U.S.A. 2002;99:12120–12125. doi: 10.1073/pnas.182156699. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Gulder TAM, Moore BS. Angew. Chem. Int. Ed. 2010;49:9346–9367. doi: 10.1002/anie.201000728. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Schultz AW, Oh D-C, Carney JR, Williamson RT, Udwary DW, Jensen PR, Gould SJ, Fenical W, Moore BS. J. Am. Chem. Soc. 2008;130:4507–4516. doi: 10.1021/ja711188x. [DOI] [PubMed] [Google Scholar]
- 14.Eustáquio AS, Pojer F, Noel JP, Moore BS. Nat. Chem. Biol. 2008;4:69–74. doi: 10.1038/nchembio.2007.56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Du L, Lou L. Nat. Prod. Rep. 2010;27:255–278. doi: 10.1039/b912037h. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.