Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Sep 21.
Published in final edited form as: Org Lett. 2018 Aug 30;20(18):5529–5532. doi: 10.1021/acs.orglett.8b01871

Biemamides A–E, Inhibitors of the TGF-β Pathway that Block the Epithelial to Mesenchymal Transition

Fan Zhang , Doug R Braun , Gene E Ananiev , F Michael Hoffmann ¥, I-Wei Tsai , Scott R Rajski , Tim S Bugni †,*
PMCID: PMC6207949  NIHMSID: NIHMS990990  PMID: 30160121

Abstract

Screening of a marine natural products library for inhibitors of TGF-β revealed five pyrimidinedione derivatives, biemamides A–E (15). The structures were determined by 2D NMR and HRMS experiments; absolute configurations were established by advanced Marfey’s analysis and ECD calculations. Biemamides A–E specifically inhibited in vitro TGF-β induced epithelial-to-mesenchymal transition in NMuMG cells. Additionally, using Caenorhabditis elegans, selected biemmamides were found to influence in vivo developmental processes related to body size regulation in a dose-dependent manner.

Graphical Abstract

graphic file with name nihms-990990-f0001.jpg


The transforming growth factor-β (TGF-β) superfamily of proteins consists of more than 30 members, including TGF-β, activins, inhibins, growth differentiation factors, and bone morphogenetic proteins (BMPs). It regulates many biological processes, such as cellular proliferation, differentiation, morphogenesis, matrix formation, extracellular proteolysis, angiogenesis, and inflammation.1 In mammals, three different TGF-β isoforms (TGF-β1, β2, and β3) encoded by different genes have been identified, and they exhibit similar effects in a variety of biological assays.2

TGF-β plays prominent roles in human disease, especially cancer,3 where it functions as a tumor suppressor in normal and early neoplastic cells. However, in established cancers, TGF-β is a central player in tumor growth, invasion and subsequent metastasis. TGF-β also coordinately upregulates the expression of many matrix proteins, and has been shown to play a causative role in the progression of many fibrotic disorders.4 Consequently, the development of TGF-β inhibitors has become a high priority; inhibitors promise to serve as useful biochemical tools as well as prospective therapeutics for fibrotic and oncological diseases.5 Indeed, several small molecule TGF-β inhibitors have been reported, and some have been advanced into clinical trials including Galunisertib (LY2157299 monohydrate),6 an oral small molecule inhibitor of the TGF-β receptor I kinase, and to date, the most advanced TGF-β signaling inhibitor under clinical development.

Inspired by recent advances in TGF-β and cancer biology, we, like many others, have incorporated a TGF-β modulating assay into screening strategies to identify new bioactive species from novel sources. Natural products have been, and remain, rich sources of clinically important compounds as reflected by the fact that 83% of small molecule anticancer drugs developed between 1981 and 2014 were either natural products or were inspired by them.7 We report here the discovery and characterization of five pyrimidinedione derivatives, biemamides A–E (15; Figure 1), from a marine-derived Streptomyces sp. These discoveries were enabled by a mink lung cell-based luciferase assay for TGF-β pathway inhibition; 15 were found to be potent TGF-β inhibitors.

Figure 1.

Figure 1.

Structures of 15.

In focusing on new natural product discoveries from marine-derived organisms, we have previously reported a highly productive metabolomics-based drug discovery platform. Microbial producer strain prioritization employs metabolomics and an LC/MS fractionation system to generate high-purity natural product libraries for high-throughput screening.8 In addition, low-temperature evaporative light scattering detectors (ELSD-LT) are generally used to quantitate high-purity natural product libraries in our approach.9 In this specific case, mink lung epithelial cells (MLEC) were stably transfected with an expression construct containing a truncated plasminogen activator inhibitor-1 (PAI-1) promoter fused to the firefly luciferase reported gene.10 Since PAI-1 is strongly induced by TGF-β, addition of TGF-β to the transfectants induces a dose-dependent increase in luciferase activity in the cell lysates. Hence, correlated luciferase activity to TGF-β pathway signaling represented a quantifiable means of monitoring TGF-β signaling efficiencies and the resultant mink cell line enabled a facile means of screening for small molecule TGF-β inhibitors. A total of 3840 fractions from 48 strains were screened, and, of these, 48 active fractions were identified. Dose-response TGF-β data and cytotoxicity assays using fibroblasts were employed to more rigorously evaluate the 48 initial hits. Three active wells (G8, G9, G10) from the strain WMMA2266 inhibited TGF-β signaling with IC50 values of 38.0, 45.7 and 49.5 ng/mL; none displayed significant cytotoxicity at the highest concentration 1 μg/mL. Chemical analysis of the secondary metabolites in the active wells/extract fractions of WMMA2266 led to the discovery of five pyrimidinedione derivatives, biemamides A–E (15). Subsequent application of the MLEC-based TGF-β inhibition assay revealed that purified 15 inhibited TGF-β with IC50s of 51.5, 18.7, 51.5, 101.7, and 83.2 nM, respectively, whereas the positive control SB431532 showed an IC50 value of 411 nM [see Supporting Information (SI)].

Notably, as we were identifying new TGF-β inhibitors 15, Goto and co-workers patented compounds 1 and 2, as antitumor agents from an alternative Streptomyces strain; no stereochemical assignments were publicized at the time.11 The 2D structures of WMMA2266-derived 1 and 2 were elucidated on the basis of NMR and MS data (Tables S1 and S2; SI) which were also in good agreement with Goto’s report.11

The absolute configurations of 1 and 2 were elucidated via application of Marfey’s advanced method.12 The acid hydrolysate of 1 was split into two equal portions and derivatized with L-FDLA and DL-FDLA, respectively. LCMS analysis of the L-FDLA and DL-FDLA products supported the assignments of D-(2,3-diaminopropanoyl)methylcarbamic acid (DAPMA) (Figure S51; SI).

The absolute configurational assignment of 1 was supported by comparing experimental and calculated electronic circular dichroism (ECD) spectra generated by time-dependent functional theory (TDDFT).13 A simplified structure 6 was used for R- and S-1 to limit conformational flexibility associated with the pendant lipid. A conformational analysis was performed using Spartan 14 software and MMFF molecular mechanics calculations. The MMFF conformational search was further optimized using TDDFT at the B3LYP/6–31G(d) basis set level, affording the eight lowest energy conformers.

The overall calculated ECD spectra of (R)-6 and (S)-6 were generated by Boltzmann distribution of the eight lowest energy conformers with 94.5%, 1.4%, 1.4%, 1.1%, 1.0%, 0.4%, 0.2% and 0.1%. The overall experimental CD spectra of 1 compared well only to that of the calculated ECD spectrum of (R)-6; both showed negative Cotton effects (CE) in the regions of 230–260 nm (Figure 2). Therefore, the absolute configuration of 1 was deduced to be 3R.

Figure 2.

Figure 2.

Experimental CD spectrum of compound 1 in CH3OH/CHCl3 (1:1) and calculated ECD spectra of two enantiomers, (3R)-6 and (3S)-6.

HRMS data indicated that both 3 and 4 possess the molecular formula C24H42N4O4. Additionally, 1H and 13C NMR data (Tables S3 and S4; SI) for 3 and 4 revealed a high degree of structural similarity to 1 and 2, although both 3 and 4 appeared to contain an olefin within each respective lipid chain. Both olefins were assigned as Z on the basis of 13C NMR chemical shifts and the realization that allylic carbons of linear olefins of the Z-isomers resonate at higher field than those of E-isomers (Z ~27 ppm; E ~32 ppm).14 Olefin positions for 3 and 4 were then determined on the basis of HRMS for semi-synthetic oxidatively cleaved variants of 3 and 4. Subjection of both 3 and 4 to NaIO4/OsO4 led, predictably, to aldehyde installation – both reactions afforded 7 (Figure S31; SI).15 HRMS supporting the molecular formula C17H28N4O5 for 7 along with all other spectral data revealed the Z olefin to reside between C-17 and C-16 of aldehyde precursors 3 and 4.

The elemental composition of 5 was established as C26H44N4O4 on the basis of HRESIMS and NMR data revealed that 5 possessed a core scaffold almost identical to that of 4. Clear however in spectra of both 4 and 5 was the presence of an additional olefinic signal within the lipid chain of 5 relative to 4. Analysis of 2D NMR (Table S5; SI), enabled us to identify the regiochemistry of both olefins within the pendant lipid of 5 and this assignment was readily validated by HRMS analyses of semi-synthetically oxidized 5. The 13C NMR chemical shifts of the allylic methylene carbons C-15 (δC 26.7), C-18 (δC 25.3), and C-21 (δC 26.7) suggested that the geometric conformation of the two olefins were Z.

Structural characterization of 15 was completed upon assigning stereochemistry to 25. On the strength of our C-3 assignment for 1 and the structural/biosynthetic relatedness of all species herein, we tentatively assigned 25 as 3R. Not surprisingly, this absolute configuration was confirmed on the basis of CD spectra and comparisons to data achieved for compound 1.

TGF-β plays a central role in processes related to tumor metastasis and consequently, its inhibition by small molecules has become a high profile pursuit in both industrial and academic settings. Epithelial to mesenchymal transitions (EMTs) are paramount to a variety of developmental processes and are largely orchestrated by TGF-β.16 Significantly, EMT dictates, to a large degree, the lethality of human cancers; primary tumor cells that have undergone EMT pose a tremendous threat by virtue of their capacity to trigger tumor metastasis. In addition, aberrant EMTs are a critical feature of other diseases, such as end-stage organ failure by fibrosis. Given this, and initial bioactivities noted for 15 in our early mink cell screens, we next sought to assess the ability of 15 to modulate EMTs via the agency of TGF-β attenuation.

The ability of biemamides A–E to modulate downstream TGF-β signaling (as reflected by EMT), was assessed using a NMuMG cell line in vitro. Notably, NMuMG is a non-transformed mouse mammary epithelial cell line that exhibits both a growth inhibitory and EMT response to TGF-β.19 In the absence of TGF-β, NMuMG cells form an epithelial shape; incubation with 50 pM TGF-β for 72 h induces a more fibroblast-like, spindle shaped morphology indicative of mesenchymal cells (Figure 3). As a positive control 1μM of SB431542,20 known as a TGF-β Type I receptor kinase inhibitor, blocked TGF-β-induced transition of the NMuMG epithelial cells into mesenchymal cells. Nine concentrations (160–0.625 nM) of 15 were tested for their ability to impact TGF-β signaling in NMuMG cells. The imaging results suggest that 15 effectively inhibited EMT induction; cells maintained their epithelial phenotype as reflected by data for 1 which is shown in Figure 3 below.

Figure 3.

Figure 3.

Effect of 1 on 50 pM TGF-β induced EMT in NMuMG cells (scale bars: 400 μm). Each experiment was repeated three times with similar results. (A) Untreated cells. (B) 50 pM TGF-β treated cells. (C) 160 nM compound 1 was incubated with 50 pM TGF-β treated cells for 72 h. (D) 1 μM SB431542 (positive control) was incubated with 50 pM TGF-β treated cells for 72 h.

Having established in vitro TGF-β inhibiting efficacy of 15, we next sought to evaluate possible in vivo activities. In C. elegans, two TGF-β-related pathways regulating dauer formation or body length have been identified, the Dauer pathway and the Sma/Mab pathway.22 In nutrient-rich environments, worms rapidly develop from embryo through four larval stages (L1–L4) to adult in 3.5 d (20 °C) in a process termed reproductive development. If exposed to nutrient limited conditions early in larval life, animals divert their development to a specialized third larval stage called the dauer diapause (L3d).23 C. elegans use TGF-β related signaling to regulate dauer formation. In addition to the dauer/TGF-β pathway, a second TGF-β related pathway Sma/Mab has been identified in C. elegans.24 Mutations in this pathway are characterized by small body size (Sma) and male tail defects (Mab). These two homologous pathways are genetically distinct, utilizing different ligands, type I receptors, and Smads. Only the type II receptor DAF-4 is shared between the two. Therefore, a TGF-β inhibitor may target the Dauer or Sma/Mab pathways individually or in tandem; specific pathway differentiation as a target of 15 was deemed early on as being beyond the scope of this report.

C. elegans were grown (in triplicate) in four concentrations of 15 ranging from 10 nM–10 μM in a liquid 96-well plate at room temperature; Escherichia coli served as a food source. After 3 d incubation at room temperature, all C. elegans were microscopically imaged and body length assessments were carried out (Figures 4, 5 and S45S49 in SI). C. elegans subjected only to solvent (control) had grown to adulthood, as evidenced by the onset of egg production as well as attainment of a typical adult size. Worms subjected to the highest concentration (10 μM) of 15 clearly displayed signs of late development and small body type individuals were easily detected. Unlike the in vitro system, no good TGF-β inhibitory positive controls were available for C. elegans studies. However, at reduced biemamide concentrations, a dose-dependent effect on development and body size regulation was clearly observed. These in vivo results suggest biemamides can inhibit TGF-β related pathways with effective concentrations spanning 100 nM and 1 μM and with likely IC50 values in the realm of ~300 nM. SB431542, which had been shown to inhibit TGF-β induced EMT in NMuMG cells in vitro, did not alter C. elegans size under three concentrations spanning 1 μM →100 μM (Figure S50; SI).

Figure 4.

Figure 4.

Alterations in C. elegans development by 15.

Figure 5.

Figure 5.

Phenotypes of C. elegans treated with 1. All panels are identically scaled [200 μm-see A red bar]. (A) Solvent only control. (B) 10 μM 1. (C) 1 μM 1. (D) 100 nM 1. (E) 10 nM 1.

In conclusion, potent TGF-β inhibitors have tremendous promise as therapeutics for pulmonary fibrosis and assorted cancers, especially those that become metastatic. In contrast to many known TGF-β inhibitors based on five-membered heterocyclic chemotypes (imidazoles, pyrazoles, or thiazoles), such as SB431542, SB505124,25 Galunisertib (LY2157299), LDN-193189,26 LY2109761,27 LY364947,28 GW788388,29 EW-7197,30 and SM-16,31 the natural product biemamides have a completely different scaffold. Highly significant is that these compounds potently suppress TGF-β signaling both in vitro and in vivo, suggesting that neither membrane permeability nor metabolic liabilities represent obvious future roadblocks to their continued development. Taken together, these findings highlight the importance of marine-derived microbes as sources of new natural products with biological activities likely to beneficially impact human health.

Supplementary Material

1

ACKNOWLEDGMENT

This work was supported by funding from the University of Wisconsin—Madison School of Pharmacy. This work was also funded by the NIH through the administration of NIGMS Grants R01GM104192 (to T.S.B.) We thank the Analytical Instrumentation Center at the University of Wisconsin—Madison for the facilities to acquire spectroscopic data and the UW Small Molecule Screening Facility which is supported by University of Wisconsin Carbone Cancer Center Support Grant P30 CA014520. This study made use of the National Magnetic Resonance Facility at Madison (NMRFAM), which is supported by NIH Grant P41GM103399.

ASSOCIATED CONTENT

Supporting Information

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.8b01871.

General information for all experimental procedures, characterization data, biological activity assays (including 48 initial hit screening data), and NMR spectra for compounds 15 (PDF).

The authors declare no competing financial interest.

REFERENCES

  • 1.(a) Heldin CH; Miyazono K; ten Dijke P Nature 1997, 390, 465–471. [DOI] [PubMed] [Google Scholar]; (b) Massague J Annu. Rev. Biochem 1998, 67, 753–791. [DOI] [PubMed] [Google Scholar]
  • 2.Schiller M; Javelaud D; Mauviel A J. Dermatol. Sci 2004, 35, 83–92. [DOI] [PubMed] [Google Scholar]
  • 3.Siegel PM; Massague J Nat. Rev. Cancer 2003, 3, 807–821. [DOI] [PubMed] [Google Scholar]
  • 4.Pohlers D; Brenmoehl J; Loffler I; Muller CK; Leipner C; Schultze-Mosgau S; Stallmach A; Kinne RW; Wolf G Biochim. Biophys. Acta 2009, 1792, 746–756. [DOI] [PubMed] [Google Scholar]
  • 5.Nacif M; Shaker O J. Cancer Therapy 2014, 5, 735–747. [Google Scholar]
  • 6.de Gramont A; Faivre S; Raymond E Oncoimmunology 2017, 6, e1257453. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Newman DJ; Cragg GM J. Nat. Prod 2016, 79, 629–661. [DOI] [PubMed] [Google Scholar]
  • 8.Zhang F; Barns K; Hoffmann FM; Braun DR; Andes DR; Bugni TS J. Nat. Prod 2017, 80, 2551–2555. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Adnani N; Michel CR; Bugni TS J. Nat. Prod 2012, 75, 802–806. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Abe M; Harpel JG; Metz CN; Nunes I; Loskutoff DJ; Rifkin DB Anal. Biochem 1994, 216, 276–284. [DOI] [PubMed] [Google Scholar]
  • 11.Kawahara T; Shinke K; Kagatani N; Suzuki S; Nishio M; Goto H Jpn. Kokai Tokkyo Koho (2017), JP 2017214304 A 20171207. [Google Scholar]
  • 12.Harada K; Fujii K; Mayumi T; Hibino Y; Suzuki M Tetrahedron Lett 1995, 36, 1515–1518. [Google Scholar]
  • 13.(a) Furche F; Ahlrichs R; Wachsmann C; Weber E; Sobanski A; Vogtle F; Grimme S J. Am. Chem. Soc 2000, 122, 1717–1724. [Google Scholar]; (b) Ding Y; Li XC; Ferreira D J. Org. Chem 2007, 72, 9010–9017. [DOI] [PubMed] [Google Scholar]
  • 14.(a) Carpita A; Quirici MG; Veracini CA Tetrahedron 1982, 38, 639–644. [Google Scholar]; (b) Mizutani K; Fukunaga Y; Tanaka O; Takasugi N; Saruwatari Y; Fuwa T; Yamauchi T; Wang J; Jia M; Li F; Ling Y Chem. Pharm. Bull 1988, 36, 2362–2365. [Google Scholar]
  • 15.Rappo R; Allen DS Jr; Lemieux U; Johnson WS J. Org. Chem 1956, 21, 478–479. [Google Scholar]
  • 16.Thiery JP Curr. Opin. Cell Biol 2003, 15, 740–746. [DOI] [PubMed] [Google Scholar]
  • 17.Zhang J; Tian X; Xing J J. Clin. Med 2016, 5, 41. [Google Scholar]
  • 18.Miettinen PJ; Ebner R; Lopez AR; Derynck R J. Cell Biol 1994, 127, 2021–2036. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Bhowmick NA; Ghiassi M; Bakin A; Aakre M; Lundquist CA; Engel ME; Arteaga CL; Moses HL Mol. Biol. Cell 2001, 12, 27–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Inman GJ; Nicolas FJ; Callahan JF; Harling JD; Gaster LM; Reith A D; Laping NJ; Hill CS Mol. Pharmacol 2002, 62, 65–74. [DOI] [PubMed] [Google Scholar]
  • 21.Kenyon C Cell 2005, 120, 449–460. [DOI] [PubMed] [Google Scholar]
  • 22.Savage-Dunn C TGF-β signaling (September 9, 2005), WormBook, ed. The C. elegans Research Community, WormBook, doi/10.1895/wormbook.1.22.1 , 10.1895/wormbook.1.22.1http://www.wormbook.org, http://www.wormbook.org. [DOI] [Google Scholar]
  • 23.Cassada RC; Russell R Dev. Biol 1975, 46, 326–342. [DOI] [PubMed] [Google Scholar]
  • 24.Padgett RW; Das P; Krishna S Bioessays 1998, 20, 382–390. [DOI] [PubMed] [Google Scholar]
  • 25.Byfield SD; Major C; Laping NJ; Roberts AB Mol. Phamacol 2004, 65, 744–752. [DOI] [PubMed] [Google Scholar]
  • 26.Sanvitale CE; Kerr G; Chaikuad A; Ramel M; Mohedas AH; Reichert S; Wang Y; Triffitt JT; Cuny GD; Yu PB; Hill CS; Bullock AN PLoS One 2013, 8, e62721. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Melisi D; Ishiyama S; Sclabas GM; Fleming JB; Xia Q; Tortora G; Abbruzzese JL; Chiao PL Mol. Cancer Ther 2008, 7, 829–840. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Sawyer JS; Anderson BD; Beight DW; Campbell RM; Jones ML; Herron DK; Lampe JW; McCowan JR; McMillen WT; Mort N; Parsons S; Smith EC; Vieth M; Weir LC; Yan L; Zhang F; Yingling JM J. Med. Chem 2003, 46, 3953–3956. [DOI] [PubMed] [Google Scholar]
  • 29.Petersen M; Thorikay M; Deckers M; van Dinther M; Grygielko ET; Gellibert F; de Gouville AC; Huet S; ten Dijke P; Laping NJ Kidney Int 2008, 73, 705–715. [DOI] [PubMed] [Google Scholar]
  • 30.Jin CH; Krishnaish M; Sreenu D; Subrahmanyam VB; Rao KS; Lee HJ; Park S; Park H; Lee K; Sheen YY; Kim D J. Med. Chem 2014, 57, 4213–4238. [DOI] [PubMed] [Google Scholar]
  • 31.Fu K; Corbley MJ; Sun L; Friedman JE; Shan F; Papadatos JL; Costa D; Lutterodt F; Sweigard H; Bowes S; Choi M; Boriack-Sjodin PA; Arduini RM; Sun D; Newman MN; Zhang X; Mead JN; Chuaqui CE; Cheung HK; Zhang X; Cornebise M; Carter MB; Josiah S; Singh J; Lee WC; Gill A; Ling LE Arter. Thromb. Vasc. Biol 2008, 28, 665–671. [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

RESOURCES