View PDF Version
 
DOI: 10.1039/D5OB00925A (Communication) Org. Biomol. Chem., 2025, Advance Article

Rapid construction of bisindole alkaloids caulersin and racemosin B from methyl 3-indolylpyruvate based on a biomimetic dimerization strategy†

Yutang Gao‡ a, Fei Yu‡a, Liang Chena and Sheng Liu*ab
aState Key Laboratory of Discovery and Utilization of Functional Components in Traditional Chinese Medicine, Guizhou Medical University, Guiyang 550014, China. E-mail: lsheng@126.com
bNatural Products Research Center of Guizhou Province, Guiyang, China

Received 4th June 2025 , Accepted 9th July 2025

First published on 10th July 2025

Abstract

Attempts to directly construct bisindole alkaloids caulerpin and caulersin resulted in the discovery of novel synthetic approaches to a specific bisindole framework with a central troponoid core and an unexpected product—racemosin B. The developed biomimetic domino sequences were triggered by a dimerization process from a shared precursor, methyl 3-indolylpyruvate. Thus, caulersin and racemosin B could be produced in 1 and 2 step processes.

Introduction

Bisindole structures are important scaffolds that widely exist in natural and synthetic bioactive compounds. The main feature of bisindole alkaloids is that the two indole molecules involved in the molecular structure are indirectly linked or directly polymerized.1 Caulerpin (1),2a,b caulersin (2)2c and racemosin B (3)2d are natural bisindoles that were all isolated from marine alga Caulerpa sp. (Fig. 1). Each of these bisindoles has an indole-methyl acrylate unit and a characteristic eight-, seven- or six-membered cyclic ring between the two indole rings.
image file: d5ob00925a-f1.tif
Fig. 1 Natural bisindole alkaloids containing 3-(1H-indol-3-yl)acrylate units (blue).

There are several synthetic routes to make these alkaloids that have been successfully established (Scheme 1).3–5 The key strategies developed for caulerpin and caulersin rely on a Knoevenagel condensation3,4b,c or a nucleophilic cyclization4a to construct cyclized eight- and seven-ring bridged bisindoles. Moreover, several synthetic approaches for the synthesis of racemosin B have reported, and the crucial steps of these processes involve a biomimetic deamination/aromatic cyclization sequence,5a Pd-catalyzed twofold oxidative cyclization,5b formal [1 + 2 + 3] annulation,5c π-extension of the indoles with acrolein5d and photochemical cyclization.5e


image file: d5ob00925a-s1.tif
Scheme 1 Previous synthetic studies for the synthesis of caulersin (2) and racemosin B (3) and the current study presented here.

Results and discussion

With our interest concerning the development of novel synthetic routes for bisindole alkaloids, we recently demonstrated a biomimetic one-step construction of lycogarubin C and lynamicin D by using methyl indolylpyruvate as a key synthon.6 Moreover, we further discovered an approach towards a specific carbazolelactam related to calothrixin B from methyl indolylpyruvate and indole.7 It was observed that caulerpin and caulersin shared very similar structural skeletons to these previously synthesized molecules. We hypothesized that both caulerpin and caulersin could be derived from a biosynthetic pathway that has a shared precursor methyl 3-indolylpyruvate 4a (Scheme 2). To be specific, it was suggested that a dimerization process involving an intermolecular Friedel–Crafts hydroxyalkylation/dehydration sequence between two molecules of 4a would generate intermediate I. Then, intramolecular cyclization occurs to convert intermediate I into caulerpin; a similar synthetic strategy has been successfully applied to construct bisindole fused cyclooctatetraenes using 3-substituted indole ketone.8 In another parallel direction, structure II could be formed from intermediate I through a Michael-type Friedel–Crafts alkylation, followed by an oxidative side chain cleavage that could lead to the formation of caulersin.
image file: d5ob00925a-s2.tif
Scheme 2 Proposed biogenesis approach for the synthesis of caulerpin (1) and caulersin (2).

To test this plan, we initially examined the cyclodimerization of 4a in the presence of several Brønsted acids in different solvents. To our disappointment, after extensive screening of these conditions (Table 1, entries 1–15), no products related to caulerpin or caulersin could be obtained. However, among these efforts, we unexpectedly found that racemosin B could be produced in 36% yield (Table 1, entry 12), when a SnCl4/HFIP (hexafluoroisopropanol) system7 was adopted in air. It was speculated that the domino transformation might be triggered by the proposed dimerization process and involved a 6π-electrocyclization/oxidative formyl formation9/oxidative deformylation10 sequence (Scheme 3). To investigate this proposal, the reaction was conducted under Ar and the desired conversion became very slow and racemosin B was obtained in 5% yield after 24 h. Moreover, two intermediates IV and V could also be detected successfully through high-resolution mass spectrometry. Encouragingly, we further discovered that a bisindole structure 5a with a central troponoid core could be formed in a moderate yield, when conc. HCl in MeOH was employed under an air atmosphere (Table 1, entry 16). Compound 5a represents the basic framework of caulersin and could act as a key synthetic precursor to produce caulersin. Therefore, we turned next attention to improving its yield by optimizing the reaction conditions. First, the volume ratio of conc. HCl to methanol was adjusted, and it was found that the yield could be enhanced to 64% when the ratio is 1[thin space (1/6-em)]:[thin space (1/6-em)]3 (Table 1, entries 17 and 18). When organic sulfonic acids such as MsOH (methanesulfonic acid), CSA (camphorsulfonic acid) and conc. sulfuric acid were employed, the yields decreased significantly (Table 1, entries 19–21). Furthermore, the adoption of several other solvents such as i-PrOH, THF, dioxane, DMSO and NMP (N-methylpyrrolidine) all gave negative results (Table 1, entries 22–26). The expected process was inhibited by an argon-protected environment (Table 1, entry 27), and the transformation was rather inefficient, producing 5a in 20% yield accompanied with 26% of compound II. Notably, the structure of II was further confirmed by NOESY correlation (Scheme 4). Moreover, when structure II was treated with the optimal conditions again, it could be converted to compound 5a in 86% yield. This observation indicated that compound II is an intermediate required for the formation of 5a.


image file: d5ob00925a-s3.tif
Scheme 3 Proposed reaction sequence for the synthesis of racemosin B.

image file: d5ob00925a-s4.tif
Scheme 4 Key NOESY correlation (arrow) of intermediate II and a control experiment.
Table 1 Optimization of the reaction conditionsa

image file: d5ob00925a-u1.tif
Entry Solvent Temperature (°C) Yield of 5a[thin space (1/6-em)]b (%)
a Reactions were performed using 4a (100 mg, 0.46 mmol) in solvent (2.0 mL) for 12–18 h in air.b Isolated yields.c Racemosin B was isolated in 36% yield.d Under an Ar atmosphere.
1 TFA 70 0
2 5.0 eq. TFA in HFIP 80 0
3 1.0 eq. HCOOH in DCE 80 0
4 1.5 eq. BF3·Et2O in MeCN 80 Trace
5 1.0 eq. TsOH in DCE 80 Trace
6 1.0 eq. TsOH in dioxane 80 Trace
7 1.2eq CSA in DCE 80 Trace
8 1.2eq CSA in DCE 80 Trace
9 1.5 eq. Cu(OTf)2 in DCE 65 Trace
10 1.0 eq. Fe(OTf)3 in HFIP 80 0
11 1.0 eq. Zn(OTf)2 in HFIP 80 0
12 1.5 eq. SnCl4 in HFIP 80 0c
13 5.0 eq. P2O5 in HFIP 80 0
14 3.0 eq. TfOH in n-Bu2O 30 Trace
15 0.2 eq. NHTf2 in toluene 80 0
16 Conc. HCl[thin space (1/6-em)]:[thin space (1/6-em)]MeOH (v/v = 1[thin space (1/6-em)]:[thin space (1/6-em)]2) 65 47
17 Conc. HCl[thin space (1/6-em)]:[thin space (1/6-em)]MeOH(v/v = 1[thin space (1/6-em)]:[thin space (1/6-em)]3) 65 64
18 Conc. HCl[thin space (1/6-em)]:[thin space (1/6-em)]MeOH (v/v = 1[thin space (1/6-em)]:[thin space (1/6-em)]4) 65 57
19 10 eq. CSA in MeOH 65 34
20 3.0 eq. MsOH in MeOH 65 33
21 98% H2SO4[thin space (1/6-em)]:[thin space (1/6-em)]MeOH (1[thin space (1/6-em)]:[thin space (1/6-em)]4) 65 Trace
22 Conc. HCl[thin space (1/6-em)]:[thin space (1/6-em)]i-PrOH (v/v = 1[thin space (1/6-em)]:[thin space (1/6-em)]3) 65 0
23 Conc. HCl[thin space (1/6-em)]:[thin space (1/6-em)]THF (v/v = 1[thin space (1/6-em)]:[thin space (1/6-em)]3) 65 0
24 Conc. HCl[thin space (1/6-em)]:[thin space (1/6-em)]1,4-dioxane (v/v = 1[thin space (1/6-em)]:[thin space (1/6-em)]3) 65 0
25 Conc. HCl[thin space (1/6-em)]:[thin space (1/6-em)]DMSO (v/v = 1[thin space (1/6-em)]:[thin space (1/6-em)]3) 65 0
26 Conc. HCl[thin space (1/6-em)]:[thin space (1/6-em)]NMP (v/v = 1[thin space (1/6-em)]:[thin space (1/6-em)]3) 65 0
27d Conc. HCl[thin space (1/6-em)]:[thin space (1/6-em)]MeOH (v/v = 1[thin space (1/6-em)]:[thin space (1/6-em)]3) 65 20

With a view to briefly explore the generality of this transformation, several indolylpyruvates 4b, 4c and 4f–i were prepared from tryptophan methyl esters with different substituents according to our previous biomimetic deamination procedure.6 Under the standard conditions, the presence of electron-donating groups (OMe, Me) or mild electron-withdrawing groups (Cl, F) on the indolylpyruvates did not significantly influence the efficiency (Scheme 5), and thus products 5b, 5c, 5f, 5g and 5i were produced in relatively normal yields (40–59%). Substrates bearing moderate electron withdrawing substituents, such as CO2Me, significantly blocked the transformation, and thus product 5h could not be obtained. Further, we envisioned that oximes 6b–j, which are easily prepared from in situ generated methyl 2-nitrosopropenoate11 and indole fragments, might also act as synthetic equivalents of indolylpyruvates. Compared to indolylpyruvate substrates, in most cases, the oxime substrates gave the desired compounds 5b–e, 5i and 5j in relatively lower yields (34–50%). Notably, products 5f and 5g could not be generated from the corresponding oxime substrates.


image file: d5ob00925a-s5.tif
Scheme 5 Substrate scope. Reactions were performed using 4 or 6 (0.46 mmol) in MeOH (1.5 mL) and conc. HCl (0.5 mL) for 12 h under an air atmosphere.

With compound 5a in hand, our final goal was to achieve a specific oxidation to directly complete the preparation of caulersin. A careful literature survey indicated that the one-step oxidation of 5a might be a tough task, since general routes that convert tropones to tropolones have been developed through two steps.12,13 Our optimization of the final conversion started with the use of various conventional oxidants, such as MnO2, KMnO4, PCC, DDQ, TBHP, IBX, K2S2O8, m-CPBA, CrO3, NaIO4, oxone and O3. Although most oxidants gave negative results, to our delight, excess NaIO4 in heated DMSO could uniquely provide the desired ketone in moderate yield (55%, Scheme 6).


image file: d5ob00925a-s6.tif
Scheme 6 Completion of the synthesis of caulersin.

Conclusion

In conclusion, we have investigated the proposed biomimetic dimerization of indolylpyruvate into the core ring system of caulerpin and caulersin. The results reported here partially vindicated our biosynthetic hypothesis, and this method features a simple system and good step economy. The framework of caulersin and several of its analogues could be delivered through a one-pot domino process, and thus caulersin could be produced in 2 steps in 35% overall yield. Moreover, an unexpected product racemosin B could be generated in 36% yield via a one-step procedure, a transformation that was thought to involve a 6π-electrocyclization/oxidative side-chain cleavage sequence. Further studies in this area are ongoing and will be reported in due course.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data underlying this study are available in the published article and its online ESI.†

Acknowledgements

The work was financially supported by Guizhou Provincial Key Technology R&D Program QKH-ZC[2023]YB209 and QKH-ZC[2022]YB192.

References

  1. (a) M. W. Xu, R. Peng, Q. Min, S. W. Hui, X. Chen, G. Yang and S. L. Qin, Bisindole natural products: a vital source for the development of new anticancer drugs, Eur. J. Med. Chem., 2022, 243, 114748 CrossRef CAS PubMed ; (b) F. Song, Y. Bian, J. Liu, Z. Li, L. Zhao, J. Fang, Y. Lai and M. Zhou, Indole alkaloids, synthetic dimers and hybrids with potential in vivo anticancer activity, Curr. Top. Med. Chem., 2021, 21, 377–403 CrossRef CAS PubMed ; (c) Y. Zhang and C. H. Hu, Anticancer activity of bisindole alkaloids derived from natural sources and synthetic bisindole hybrids, Arch. Pharm., 2020, 353, e2000092 CrossRef PubMed .
  2. (a) G. Aguilar-Santos, Caulerpin, a new red pigment from green algae of the genus Caulerpa, J. Chem. Soc. C, 1970, 6, 842–843 RSC ; (b) B. C. Maiti, R. H. Thomson and M. Mahendran, The structure of caulerpin, a pigment from Caulerpa algae, J. Chem. Res., Synop., 1978, 4, 126–127 Search PubMed ; (c) J. Y. Su, Y. Zhu, L. M. Zeng and X. H. Xu, A new bisindole from alga Caulerpa serrulata, J. Nat. Prod., 1997, 60, 1043–1044 CrossRef CAS ; (d) D. Q. Liu, S. C. Mao, H. Y. Zhang, X. Q. Yu, M. T. Feng, B. Wang, L. H. Feng and Y. W. Guo, Racemosins A and B, two novel bisindole alkaloids from the green alga Caulerpa racemosa, Fitoterapia, 2013, 91, 15–20 CrossRef CAS PubMed .
  3. (a) C. I. Canche Chay, R. Gomez Cansino, C. I. Espitia Pinzon, R. O. Torres-Ochoa and R. Martinez, Synthesis and anti-tuberculosis activity of the marine natural product caulerpin and its analogue, Mar. Drugs, 2014, 12, 1757–1772 CrossRef PubMed ; (b) H. B. Li, X. J. Liao, Y. G. Sun, R. Zhou, W. L. Long, L. Li, L. Q. Gu and S. H. Xu, An economical synthesis of caulerpin and evaluation of its new anticancer activities, ChemistrySelect, 2018, 3, 12406–12409 CrossRef CAS .
  4. (a) P. M. Fresneda, P. Molina and M. A. Saez, The first synthesis of the bis(indole) marine alkaloid caulersin, Synlett, 1999, 1651–1653 CrossRef CAS ; (b) N. Wahlstroem, B. Stensland and J. Bergman, Synthesis of the marine alkaloid caulersin, Tetrahedron, 2004, 60, 2147–2153 CrossRef ; (c) P. M. Fresneda, P. Molina and M. A. Saez, Synthesis of caulersin and its isomers by reaction of indole-2,3-dicarboxylic anhydrides with methyl indoleacetates, Tetrahedron Lett., 2006, 47, 5215–5218 CrossRef .
  5. (a) L. N. Liang, T. Y. Fan, T. Huang, C. Yan, M. Xu and S. Liu, A biomimetic method to synthesise indolo[3,2-a]carbazoles, Tetrahedron Lett., 2015, 56, 434–436 CrossRef CAS ; (b) C. Yang, K. Lin, L. Huang, W. D. Pan and S. Liu, Facile synthesis of indolo[3,2-a]carbazoles via Pd catalyzed two fold oxidative cyclization, Beilstein J. Org. Chem., 2016, 12, 2490–2494 CrossRef CAS PubMed ; (c) Y. Men, Z. Hu, J. Dong, X. Xu and B. Tang, Formal [1 + 2 + 3] annulation: domino access to carbazoles and indolocarbazole alkaloids, Org. Lett., 2018, 20, 5348 CrossRef CAS PubMed ; (d) Y. H. Li, J. Jin, W. B. Fan and D. G. Huang, π-Extension of indoles using acrolein linker: synthesis of indolo[3,2-a]carbazole-6-carbaldehydes and racemosin B, Org. Lett., 2023, 25, 8284–8289 CrossRef CAS PubMed ; (e) I. Wazir and R. Nagarajan, Synthesis of racemosin B and its analogues by a photochemical cyclization, Synlett, 2024, 1929–1931 CAS .
  6. W. X. Li and S. Liu, A biomimetic approach to lycogarubin C, lynamicin D and related analogues, Org. Biomol. Chem., 2024, 22, 2558–2561 RSC .
  7. Z. Z. Wang and S. Liu, A rapid approach towards a specific carbazolelactam system related to calothrixin B, Org. Biomol. Chem., 2024, 22, 2558–2561 RSC .
  8. L. Zhao, Z. H. Yan, S. Tang, Z. L. Wei and W. W. Liao, Brønsted acid-promoted cyclodimerization of indolyl ketones:construction of indole fused-oxabicyclo[3.3.1]nonane and -cyclooctatetraene ring systems, Org. Lett., 2021, 23, 166–171 CrossRef CAS PubMed .
  9. L. N. Liang, T. Y. Fan, T. Huang, C. Yan, M. Xu and S. Liu, A biomimetic method to synthesise indolo[3,2-a]carbazoles, Tetrahedron Lett., 2015, 56, 434–436 CrossRef CAS .
  10. Z. Z. Wang, H. H. Wang, M. Xu, L. P. Wang and S. Liu, Unexpected amine-triggered skeletal modification fascaplysin and its derivatives: rapid access to δ,γ-biscarbolines, Org. Biomol. Chem., 2025, 23, 2096–2100 RSC .
  11. (a) P. H. H. Hermkens, J. H. V. Maarseveen, C. G. Kruse and H. W. Scheeren, 1,3-Dipolarcycloaddition of nitrones with nitriles: scope and mechanistic study, Tetrahedron, 1988, 44, 6491–6504 CrossRef CAS ; (b) R. Pereira, R. Benedetti, S. Perez-Rodríguez, A. Nebbioso, J. García-Rodríguez, V. Carafa, M. Stuhldreier, M. Conte, F. Rodríguez-Barrios, H. G. Stunnenberg, H. Gronemeyer, L. Altucci and A. R. de Lera, Indole-derived psammaplin A analogues as epigenetic modulators with multiple inhibitory activities, J. Med. Chem., 2012, 55, 9467–9491 CrossRef CAS PubMed .
  12. (a) D. L. Boger and K. Takahash, Total Synthesis of granditropone, grandirubrine, imerubrine, and isoimerubrine, J. Am. Chem. Soc., 1995, 117, 12452–12459 CrossRef CAS ; (b) T. A. Miller, A. L. Bulman, C. D. Thompson, M. E. Garst and T. L. Macdonald, The synthesis and evaluation of functionalized estratropones: potent inhibitors of tubulin polymerization, Bioorg. Med. Chem. Lett., 1997, 7, 1851–1856 CrossRef CAS .
  13. M. Güney, A. Daştan and M. Balci, Chemistry of the benzotropone endoperoxides and their conversion into tropolone derivatives: unusual endoperoxide rearrangements, Helvetica, 2005, 88, 830–838 CrossRef .

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ob00925a
‡ These authors contributed equally to this work.
This journal is © The Royal Society of Chemistry 2025
Click here to see how this site uses Cookies. View our privacy policy here.