Modified peptides and organic metabolites of cyanobacterial origin with antiplasmodial properties

Abstract

As etiological agents of malaria disease, Plasmodium spp. parasites are responsible for one of the most severe global health problems occurring in tropical regions of the world. This work involved compiling marine cyanobacteria metabolites reported in the scientific literature that exhibit antiplasmodial activity. Out of the 111 compounds mined and 106 tested, two showed antiplasmodial activity at very low concentrations, with IC₅₀ at 0.1 and 1.5 nM (peptides: dolastatin 10 and lyngbyabellin A, 1.9% of total tested). Examples of chemical derivatives generated from natural cyanobacterial compounds to enhance antiplasmodial activity and Plasmodium selectivity can be found in successful findings from nostocarboline, eudistomin, and carmaphycin derivatives, while bastimolide derivatives have not yet been found. Overall, 57% of the reviewed compounds are peptides with modified residues producing interesting active moieties, such as α- and β-epoxyketone in camaphycins. The remaining compounds belong to diverse chemical groups such as alkaloids, macrolides, polycyclic compounds, and halogenated compounds. The Dolastatin 10 and lyngbyabellin A, compounds with antiplasmodial high activity, are cytoskeletal disruptors with different protein targets.

Graphical abstract

1. Introduction

The phylum Cyanobacteriota (also known as blue-green algae or Cyanophyceae Class) is ancestral photosynthetic prokaryotes and among the earliest known living organisms on the Earth. They are responsible for the increase in atmospheric oxygen that occurred 2.49 billion years ago, although fossil evidence has suggested their existence at least 1.9 billion years ago. They have adapted to diverse ecosystems and habitats (e.g., from glaciers to continental freshwater environments and oceans) and engage in specialized symbiotic relationships with a remarkable variety of environments and hosts (e.g., Cnidaria, Mollusca, Porifera, and other phyla) (Demoulin et al., 2019; Mutalipassi et al., 2021). Cyanobacteria are a rich and peculiar source of novel natural bioactive compounds produced by their secondary metabolism, which has diversified through adaptations to colonize distinct habitats and niches. These challenges generate a variety of enzymatic pathways resulting in original compounds or modifications (reviewed by Dixit and Suseela, 2013; Salvador-Reyes and Luesch, 2015; Shah et al., 2017; Khalifa et al., 2021). The discovery of marine cyanobacteria chemo-biodiversity offers the possibility to find new drugs. According to recent assessments of marine cyanobacteria chemo-biodiversity, it comprises more than 450 unique compounds from species of the genera Lyngbya, Oscillatoria, and Symploca (Anjum et al., 2017). Another study categorized 91 compounds of eight chemical groups (lipopeptides, polyketides, peptolide, depsipeptides, peptides, protein, polysaccharide, and alkaloids) from species of 14 genera and 28 compounds with diverse biological activities (e.g., antibiotic, anticancer, antifungal, antiviral, anthelmintic, antimicrobial, anti-parasitic, and immunosuppressive) from a fraction of the >800 cyanobacterial compounds in a database (Sweeney-Jones et al., 2020; Khalifa et al., 2021). This diversity of compounds and biological activities has led researchers to manage specific topics such anti-parasitic, to find new compounds with specific activities. Therefore, this review focuses on cyanobacterial compounds with antiplasmodial activity.

Global healthcare systems confront a diversity of existing diseases (e.g., malaria complex diseases, trypanosomiasis diseases, trichomoniasis, giardiasis, and amebiasis) caused by known protozoan parasites, as well as new diseases caused by protozoans through zoonoses (i.e., Microsporidia diseases and Naegleria fowler) for which we have limited or no effective drugs (Estrella-Parra et al., 2022). Recently, the world was affected by a new coronavirus disease 2019 (COVID-19) infectious disease caused by the severe acute respiratory syndrome coronavirus 2 (SARS-COV2), which originated in bats. The ensuing pandemic serves as a clear example of how an unexpected new disease can emerge and cause a pandemic for which we have no medicines (Konda et al., 2020). In this context, the emergence of new protozoan parasitic diseases, new disease varieties, or novel drug resistance among existing diseases is also possible. Therefore, new natural compounds contribute to the design of chemical libraries to finding new drugs (anti-parasitic and anti-infective/chemoprevention therapies) or new lead compounds, through chemical synthetic methods. The malaria complex disease, which mainly affects developing countries in tropical and subtropical regions, is a neglected tropical disease that requires constant attention and access to new drugs (Salas-Sarduy et al., 2013; Kheang et al., 2022). The primary goal of our review is to discuss the variety of marine-isolated cyanobacterial chemo-biodiversity with antiplasmodial activity.

2. Malaria complex diseases

Malaria and paludism (neglected tropical diseases) are endemic versions of the same tropical and subtropical complex disease (now named malaria only) caused by protozoan parasites in the phylum Apicomplexa (Alveolata Class) from the genus Plasmodium. Malaria is caused by Plasmodium falciparum and Plasmodium vivax; however, five species of parasites are recognized as part of the malaria complex diseases, including P. falciparum (causes severe malaria), P. vivax (the most geographically widespread), P. malariae, P. knowlesi, and P. ovale (P. o. curtisi and P. o. wallikeri), all of which are transmitted by mosquitoes of the Anopheles genus. These parasites can infect humans and animals and are transmitted via the insect vector by infected mosquito bites. Nearly 30 years ago, the malaria complex disease killed an estimated 1 to 2 million people annually (Murphy and Oldfield, 1996). However, the number of fatalities decreased steadily over time to 897 000 in 2000 and 568 000 in 2019, increased by 10% in 2020 (to 625 000), and declined slightly to 619 000 in 2021 (World Health Organization, 2022). The causes of this decline can be explained by simple reasons such as better preventive actions including the fumigation of mosquitoes (insecticide-treated mosquito net strategies) and the use of seasonal malaria chemoprevention strategies using drugs such as chloroquine phosphate (with reported resistance), doxycycline, mefloquine, and proguanil, as well as combination drug therapies for curative treatments (e.g., artemether-lumefantrine, artesunate-mefloquine, artesunate-pyronaridine tetraphosphate, and artemisinin-based combination therapies), and other treatments such as quinine sulfate and primaquine phosphate (World Health Organization, 2019a; 2019b). Although Plasmodium mortality rates decreased between 2000 and 2019, the disease continues to affect large human populations in Haiti, Venezuela, and Central African countries. However, in recent years, it has been observed that treatment with combination therapies consistently fails at a rate of approximately 10% in the disease's hotspot in Africa. Although many countries of the world maintain regions with a high endemic prevalence—such as Brazil, Nicaragua, Colombia, and Peru in South America, as well as India, Pakistan, Myanmar, Papua New Guinea, the Philippines, and Vietnam in Asia—there are countries with no public data available. Furthermore, one problem with malaria treatment is the increased drug resistance and side effects of toxicity among at-risk patients (Kulkeaw, 2021).

3. Search for the compounds, targets, and functional mechanisms

A search of the scientific literature was performed in the PubMed server (https://pubmed.ncbi.nlm.nih.gov/) using Boolean operators, as follows. The first search terms “(plasmodium OR malaria) AND (cyanobacteria)” produced 70 results and the second search terms “(antimalarial) AND (cyanobacteria)” produced 55 (total: 125). During the title/abstract revision of the collected research articles, 65 initial research articles were selected. A third search used the terms “(plasmodium OR malaria) AND (marine drugs)” and produced 120 results. Of these, only 19 were new and were also included in the final set of 84 research articles. From this final set, various articles were discarded when the terms were contained but the articles are on topics not related to our topic; when the articles focused on molecules extracted from other marine organisms (e.g., sponges; while we are aware that compounds could be produced by the symbiotic cyanobacteria from sponges host, the precise sources have not been well established); when the antimalarial tests were performed as a rapid test with a partially purified extract fraction without an identification of a specific molecule. The final number of research articles with purified compounds tested with IC₅₀ or similar values and solved structures was 39. Additional articles were integrated from an additional search of terms referring to similar action mechanisms of the compounds, closely related molecules with similar activities or structures, and from a search of the protein data bank of protein complexes (Fig. 1). A search for similar sequences was performed on the protein BLAST of the NCBI server using standard parameters (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Images of chemical structures were obtained from ChemSpider and the Natural Products Atlas (see Acknowledgments). Some images of chemical structures were created again to clarify the visualization using KingDraw professional chemical structure editor software (http://www.kingdraw.cn). The moiety figures of supplemental figures and other modifications (e.g., linker chains of nostocarboline dimers) were prepared using KingDraw. The names of any moieties (e.g., azole moieties were uniformized) were defined with ACDlab Freeware (version 2021). Figures were rendered according to the parameter of Tachyon in VMD program v1.9.4a53 (Humphrey et al., 1996).

Fig. 1.

Workflow diagram of the literature search in Pubmed.

A search of the scientific literature was performed in the PubMed server (https://pubmed.ncbi.nlm.nih.gov/) using Boolean operators, in three steps. At each step the titles and abstracts were revised; in the last step all articles collected were read to select the final set.

4. Compounds produced by cyanobacteria

For 50 years, the ability of marine, freshwater, and terrestrial cyanobacterial extracts to produce biological effects has been well known (Mynderse et al., 1977). Simple extracts (e.g., Nostoc commune and Rivularia biasolettiana) have been tested against parasites, including P. falciparum (Lee et al., 1994; Becher et al., 2005; Broniatowska et al., 2011). The exploration of cyanobacterial natural compounds began by obtaining extracts (using mainly methanol, ethanol, and chloroform mixtures) from cultured or sampled cyanobacteria from accessible habitats. These extracts were subjected to chromatographic fractionation and purification and tested in vitro against P. falciparum strains (Veerabadhran et al., 2014). Other works have evaluated the value of the antimalarial activity of food supplements such as Spirulina (e.g., the cyanobacterium Spirulina platensis) (Wulandari et al., 2018). Many compounds with high chemical variability are isolated and characterized (mainly modified peptides), followed by the testing of antiplasmodial activity (most commonly against blood stages), other anti-parasite tests, and searches for anti-cancer properties. However, it is desirable to explore additional habitats for extract and access the natural variability of compounds, such as cyanobacteria living in artificial lagoons/ponds and garbage dumps (containing human waste), in rainforests (tree bark and soil), in uncommon places such as hydrothermal vents close to surface water or deep marine places, as well as symbiotic gut microbiome cyanobacteria (e.g., the non-photosynthetic group of Melainabacteria Class) (Puente-Sánchez et al., 2018; Hu and Rzymski, 2022).

4.1. Modified peptides from cyanobacteria

4.1.1. Depsipeptides

Peptides and modified peptides are common products of cyanobacterial secondary metabolism. Cyclic depsipeptides (peptides with the amide(s) group(s) replaced by a corresponding ester group or amino acids and hydroxy acids linked by ester and amide bonds) containing compounds such as Dhoya moiety have been isolated from cyanobacteria, which likely synthesized by non-ribosomal peptide synthetases and polyketide synthetases. Cyclic depsipeptides isolated from cyanobacteria with antiplasmodial properties and anti-cancer activity have been found, such as dudawalamides A to D (cyclic depsipeptides containing an R/S-Dhoya moiety with five/six residues) from the marine cyanobacteria Moorea producens, which exhibit an IC₅₀ of 10,000–3500 nM against the P. falciparum chloroquine-resistant W2 strain and minimal mammalian cell cytotoxicity (Table 1 and see Table S1, which contains all structures of the cyanobacterial compounds with antiplasmodial activity described here). Dudawalamides contain non-modified L-amino acids (e.g., Gly, Val, Pro, and Ala) alternating with modified residues (e.g., N-M-Ile, N-M-Phe, Hmba, D-allo-Hmpa, N-M-Phe, NO-diM-Tyr, and N-M-Val) in the sequence determined by the planar structure of the peptides. These cyclic depsipeptides are covalently linked between C- and N-terminals from L-amino acids through the R/S-Dhoya moiety as a connector. Dudawalamides C and D contain two unusual moieties, Hmba and D-allo-Hmpa, respectively and, dudawalamide A contains an unusual moiety, Lac bound to R/S-Dhoya. The mechanisms to produce unusual moieties are unclear, although they probably are produced by deamination of usual amino acids or the insertion of organic acids such as pyruvate (Almaliti et al., 2017). Dudawalamides belong to a family of lipo-depsipeptides of marine origin (e.g., viequeamides, wewakpeptines, and pitipeptolides) with moderate cytotoxicity against mammalian cell lines with the R/S-Dhoya moiety as a connector of the cyclic depsipeptides.

Table 1.

Peptides and compounds from cyanobacteria with antiplasmodial activity with chemical diversity.

Compounds		Reference
Dudawalamide A IC50: 3600 nM; Pf: W2; Si: 8.3; Ri: NA Seq: Ala: N-M-Ile:Pro: N-M-Phe: Gly: Dhoya:Lac: Stage affected: erythrocytic stages H-640 lung cancer cells: No cytotoxicity at 30 μM.		Almaliti et al. (2017)
Viridamide A IC50: 5800 nM; Pf: unknow; Si: NA; Ri: NA Seq: Mdyna: N-M-Ile:Val: Val: N-M-Val: Hmpa: O-M-Pro Stage affected: unknow		Simmons et al. (2008)
Symplocamide A IC50: 950 nM; Pf: W2; Si: 0.04–0.03; Ri: NA IC50: 40 nM; H-460 lung cancer cells IC50: 29 nM; Neuro-2a neuroblastoma cells Seq: L-Cit: Ahp: L-Ile: NO-diM-Br-Tyr: L-Val: Aba-L-Gln-But Stage affected: erythrocytic stages Tg: hypothetical inhibition of serine proteases		Linington et al. (2008)
Companeramide B IC50: 220 nM; Pf: D6; Si: 4.5; Ri: NA IC50: 230 nM; Pf: Dd2; Si: 4.3; Ri: 1.04 IC50: 700 nM; Pf: 7G8; Si: 1.4; Ri: 3.18 Seq: N-M-Ala: Val: N-M-Val: Ile: Pro: N-M-Ala: Val: N-M-Val: Amoya: Hmba Stage affected: erythrocytic stages NCIH460 non-small-cell lung carcinoma, MDA-MB-231 breast adenocarcinoma, SF-295 glioblastoma, and SK-OV3 ovarian carcinoma cells: No significant cytotoxicity at 1 μM		Vining et al. (2015)
Lagunamide A IC50: 190 nM; Pf: NF54; Si: 0.033; Ri: NA IC50: 6.4 nM; P388 leukemia cell lines Seq: Hmpa: Ala: N-M-Phe: N-M-Gly: Ile: N-M-Ala: Dtea1 Stage affected: erythocytic stages Tg: Hypothetical caspase-mediated mitochondrial apoptosis		Tripathi et al. (2010)
Kakeromamide B EC50: 890 nM; Pf: Dd2; Si: 2.5; Ri: NA EC50: >2300 nM; HEK293T, HepG2 cell lines Seq: Val: NO-M-Tyr: NO-M-Tyr: Ala-Tzl-Ca: Amoa Stage affected in RBC: Blood stages		Sweeney-Jones et al. (2020)
Venturamide B IC50: 5200 nM; Pf: W2; Si: 10.7–10.3; Ri: NA IC50: 56 000 nM; vero cells IC50: >54 000 nM; MCF-7 cancer cells Seq: Thr-Tzl-Ca: Ala-MOzl-Ca: Val-Tzl-Ca Stage affected: erythrocytic stages		Linington et al. (2007)
Aerucyclamide B IC50: 700 nM; Pf: K1; Si: 171; Ri: NA IC50: >120 000 nM Rat Myoblast L6 Cells Seq: Gly-Tzl-Ca: Ile-MOzn-Ca: Ile-Tzl-Ca Stage affected: erythrocytic stages		Portmann et al. (2008)
Balgacyclamide B IC50: 8200 nM; Pf: K1; Si: 18; Ri: NA IC50: >150 000 nM Rat Myoblast L6 Cells Seq: Ile-Tzl-Ca: Val-MOzn-Ca: Thr: Ala Stage affected: erythrocytic stages		Portmann et al. (2014)
Macrocycle 4 IC50: 180 nM a synthetic molecule; Pf: K1 Si: NA; Ri: NA Seq: mOzl: Ile: Tzl: Ile: Tzl: Gly Stage affected: erythrocytic stages		Peña et al., 2012)
Dolastatin 15 IC50: 200 nM; Pf: FCH5.C2; Si: 0.011; Ri: NA IC50: 2.2 nM; HCT116 cell line Seq:Dolavaline:Val:Val:Pro:Pro:Hmba:Benp Stage affected in RBC: Trophozoite/Schizonts arrested		(Fennell, 2003
Malyngamide X ED50: 5440 nM; Pf: K1; Si: 1.5/0.75/1.3; Ri: NA ED50: 8200 nM kB cell line ED50: 4120 nM NCI-H187 cell line ED50: 7030 nM BC cell line Seq: Mtea: N-M-Ala: Mpna: Lpmp Stage affected: erythrocytic stages		Suntornchashwej et al. (2007)
Ikoamide IC50: 140 nM; Pf: 3D7; Si: NA; Ri: NA HeLa; HL60 cells: No significant growth-inhibitory activity at 10 μM Seq: Dmo: N-M-D-allo-Ile: D-Ser: D-Leu: N-M-Leu: L-Thr: N-M-Ile: N-M-Gln: NO-diM-L-Tyr-O-M Stage affected: erythrocytic stages		Iwasaki et al. (2020)
Hoshinoamide A IC50: 520 nM; Pf: 3D7; Si: 19.23; Ri: NA HeLa cells: not exhibit any cytotoxicity against at 10 μM Seq: Hba: Aha: N-M-Leu: Ile: N-M-D-Val: Gln: Val: N-M-D-Phe: Pro-O-M Stage affected: erythrocytic stages		Iwasaki et al. (2018)
Santacruzamate A Extract fraction H, 99% inhibition of growth 10 μg/ml; Pf: Indochina W2 Subfraction purificaction contained Santacruzamate A. IC50: 0.119/>1000/434 nM; HDAC: HDAC2, HDAC4, HDAC6; Si: NA; Ri: NA GI50: 29 400/1400 nM; HCT-116, HuT-78 cancer cell lines Stage affected: erythrocytic stages		Pavlik et al. (2013)
Malyngolide dimer IC50: 19 000 nM; Pf: W2; Si: NA; Ri: NA H-640 lung cancer cells: cytotoxicity at 9 μM (110% survival) and 55 μM (10% survival). Stage affected: erythrocytic stages		Gutiérrez et al. (2010)
Hierridin A/B mixture Hypothetical mixture 50/50% IC50: 9772 nM; Pf: D6; Si: NA; Ri: NA IC50: 3.7 μg/ml; Pf: D6 Hierridin A = R1 C17H35 Hierridin B R1 = C15H31		Papendorf et al. (1998)
Ambigol A IC50: 1744.3 and 3346 nM; Pf: K1/NF54 MIC: 206,185 nM; Rat myoblast L6 cells. Si: 118/61; Ri: 0.52 Stage affected: erythrocytic stages		Wright et al. (2005)
Lyngbyaloside EC50: >790 nM; Pf: Dd2; Si: 1.6/1.6; Ri: NA EC50: >1300/1300 nM; HEK293 T/HepG2 Stage affected: erythrocytic stages		Sweeney-Jones et al. (2020)
Biselyngbyaside IC50: 3400 nM; Pf: K1; Si: 0.7/0.1; Ri: 0.7 IC50: 4400 nM; Pf: FCR3; Si: 0.56/0.09; Ri: NA IC50: 2500/400 nM; HeLA/MRC-5 cell lines Stage affected: erythrocytic stages		Sato et al. (2018)
Biselyngbyolide B IC50: 24 000 nM; Pf: K1 Si: 0.0001/0.009; Ri: 1.02 IC50: 23 500 nM; Pf: FCR3 Si: 0.0001/0.009; Ri: NA IC50: 2.8/230 nM; HeLA/MRC-5 cell lines Stage affected: erythrocytic stages

Pf = P. falciparum strain; Tg = target/activity; Seq = peptide sequence; Si = Selectivity index; Ri = Resistance index; NA = not acredited, not evaluated, experiment not carried out.

Other depsipeptides include viridamides A and B, which are non-cyclic lipo-depsipeptides with Mdyna moiety (methoxylated fatty acid moiety) bound in the N-terminal of the peptides, which contains two L-amino acids and four modified residues. Viridamides were isolated from the Oscillatoria nigro-viridis strain OSC3L, a dark green and very small cyanobacterium. Notably, viridamide A showed activity against P. falciparum with an IC₅₀ of 5800 nM and activity against cancer cell lines evaluated by disk diffusion zone of inhibition (250 and 200 mm, respectively). Mdyna moiety is bound to the N-M-Ile residue at the N-terminal followed by a non-modified Val, a N-M-Val, Hmpa (viridamide A)/Hmba (viridamide B), and O-M-Pro as final residue (Simmons et al., 2008). No mechanisms of action of dudawalamides and viridamides activities have been associated with these compounds.

Symplocamide A is a brominated cyclic depsipeptide containing an unusual L-Cit residue, followed by an Ahp, L-Ile, NO-diM-Br-Tyr, L-Val and covalently linked by Aba-L-Gln-But moiety as a connector between N-terminal of L-Cit and C-terminal of L-Val. Symplocamide A exhibits activity against the P. falciparum W2 strain, showing an IC₅₀ of 950 nM and cytotoxicity to cancer cells in vitro, with a specific serine protease inhibitor activity in trypsin and chymotrypsin. Symplocamide A was isolated from a marine cyanobacterium Symploca sp. collected at a depth of 25 m on Sunday Island in Papua New Guinea (Linington et al., 2008). Symplocamide belongs to the cyclic Ahp-containing depsipeptides family of cyanobacterial origin that exhibits serine protease inhibitor activities. Examples of cyclic Ahp-containing depsipeptides family are inhibitor A90720A from freshwater Microchaete loktakensis (Lee et al., 1994); lyngbyastatin 5, 6, and 7, as well as kempopeptin A and B, from marine Lyngbya sp. (Matthew et al., 2007; Nogle et al., 2001); lyngbyastatin 4 from marine Lyngbya confervoides (Matthew et al., 2007); somamide A and B from the marine assemblage of Lyngbya majuscula and Schizothrix sp., (Nogle et al., 2001); symplostatin 2 from marine Symploca hydnoides (Harrigan et al., 1999); molassamide from marine Dichothrix utahensis (Salvador et al., 2013); and cyanopeptolin-type peptides (Mazur-Marzec et al., 2018). Unfortunately, many of these cyanobacterial cyclic Ahp-containing depsipeptides are untested against Plasmodium species. However, a hypothetical mechanism of action can be proposed, mediated by the inhibition of P. falciparum essential serine proteases. Additionally, we found the availability of protein structure complexes on the RCSB Protein Data Bank (PDB) database such as inhibitor A90720A in complex with Bos taurus trypsin (PDB ID: 1TPS, UniProt ID: P00760·TRY1_BOVIN, Serine protease 1), as well as Tutuilamide A and Lyngbyastatin 7 with Sus scrofa pancreatic elastase (PDB ID: 6TH7 and 4GVU, UniProt ID: P00772·CELA1_PIG, Elastase-1) (Keller et al., 2020; Lee et al., 1994; Salvador et al., 2013). A rapid inspection of the structures shows that the main chain conformation of the hhh-Ahp-xxx-Aba-Val-Tyr residue section (hhh represents hydrophobic residues, xxx is any residue, and Tyr can be modified) is over the top of the catalytic residues (Ser and His) on the active site pocket of the serine protease (Fig. 2, panel A). These residues are a kind of signature of the activity against serine proteases. An interesting observation is that Val and Ahp maintain two main chain hydrogen bridges, allowing the folding of the cyclic Ahp-containing depsipeptides, and the other residues contribute to binding and specificity (Lee et al., 1994). The Ahp units on the depsipeptides are considered essential for inhibiting of serine proteases due to extensive complementarity preventing the enzymatic catalytic process (Vining et al., 2015). The Aba-Val residues are conserved in the Ahp-containing depsipeptides mentioned here (all from cyanobacteria); however, other Ahp-containing depsipeptides (streptopeptolins) from bacteria such as Streptomyces olivochromogenes contains Aba-Ala (Tiwari et al., 2021). The Ahp-containing depsipeptides are interesting lead compounds to be used as scaffolds with a core of Ahp-xxx-Aba-Val/Ala and the remaining side chains as variable positions for redesign to manage specificity against Plasmodium enzymes.

Fig. 2.

Protein structures complex with similar molecules reported with antiplasmodial activity.

Panel A) Crystal structure of the mammalian serine protease trypsin (UniProt ID: P00760 · TRY1_BOVIN, Serine protease 1, Anionic trypsin I, TRY1) from Bos taurus in complex with inhibitor A90720A (from freshwater blue-green alga Microchaete loktakensis) bound on the active site cavity (PDB ID: 1TPS). A90720A is a compound of the cyclic Ahp depsipeptides family depicted in a licorice in black color (drawing method of VMD program; Humphrey et al., 1996). A90720A is deeply buried in trypsin active site cavity and maintains 13 hydrogen bonds (only represented 5 to avoid saturation with residues in licorice in grayscale colors). Non-bonded contacts detected (less than 4 Å around) are 112 according to the activity (IC₅₀ of 10 nM for bovin trypsin). The catalytic triad (residues His57, Asp102 and Ser214) is close to the motif Ahp: L: NMY(N-methyl-Tyr):V (underlined residues, all standard residues in 1-letter code), similar to symplocamide A motif Ahp:Ile:NO-diM-Br-Tyr:Val. A90720A maintain the elliptic form observed in cyclic peptides with Ahp. Panel B) Electron microscopy structure (PDB ID: 7TR3) of the Tubulin alpha-1B chain (UniProt ID: Q2XVP4·TBA1B_PIG) and Tubulin beta-2B chain subunits (UniProt ID: P02554·TBB_PIG, Tubulin beta chain) from Sus scrofa in complex with a Kinesin-like protein from Candida albicans (UniProt ID: A0A1D8PKA4·A0A1D8PKA4_CANAL) and with dolastatin 10 bound (cyanobacterial compound), phosphoaminophosphonic acid-adenylate ester (ATPase inhibitor), guanosine-5′-diphosphate (GDP) and guanosine-5′-triphosphate (GDP). Tubulin beta-2B (newcartoon in light grayscale colors) show bound to dolastatin 10 inhibitor (licorice in black color abbreviated D-10) in a superficial groove over the binding cavity of GDP bound (molecule licorice in grayscale colors). Dolastatin 10 inhibitor probably interfering with the replacement of GTP-GDP activity producing inhibits tubulin polymerization. Dolastatin 10 shown only non-bonded contacts with tubulin (detected 44) with residues depicted (licorice in light grayscale colors) probably mediated mainly by hydrophobic interactions. Panel C) Crystal structures of carmaphycin A peptide (PDB ID: 4HRC; licorice in black color, a cyanobacterial compound) bound to Saccharomyces cerevisiae 20S proteasome protein complex. Yeast proteasome complex is formed by 14 different proteinases but carmaphycin A bound only in proteasome subunits β-type 1, 2 and 5 (UniProt ID: P30656·PSB5_YEAST, P38624·PSB1_YEAST and P25043•PSB2_YEAST) located in the central rings of the complex with an irreversible covalently bound to residue Thr 1(licorice in grayscale colors). Carmaphycin A and carfilzomib (PDB ID: 4QW4; not represented) maintain 6 hydrogen bonds with the same residues (Thr21, Gly47, Ala49 and Asp124 of H subunit β-type 2). The figure only shows H subunit β-type 2 in complex with carmaphycin A and 6 residues. Non-bonded contacts detected are similar between the two compounds but have few differences due to molecular size affecting probably the mechanism of binding but not the action mechanism. Panel D) Crystal structure of the mammalian sarcoplasmic/endoplasmic reticulum Calcium ATPase 1 (Calcium pump SERCA protein, PDB ID: 4YCN, UniProt ID: P04191·AT2A1_RABIT) from Oryctolagus cuniculus with bound marine cyanobacterial macrolide BLLB. BLLB is depicted as licorice in black color and SERCA protein is depicted as newcartoon in light grayscale colors. BLS and BLLB macrolides are bound on the same inhibitory site, a big cavity insight the SERCA Protein. BLS and BLLB maintain two hydrogen bonds with SERCA protein (Gln56 and Asn101) with large number of non-bonded contacts (51 detected) with 17 residues. Only are depicted 7 to represent the main helices forming the cavity (one to left, two to right, one on the bottom).

The cyclic depsipeptide companeramides A and B contain an Amoya β-amino acid, an Hmba, and eight hydrophobic α-amino acid residues with similar sequences N-M-Ala, Ile/Val, N-M-Val, Ile, Pro, N-M-Leu/N-M-Ala, Ala/Val and N-M-Val, respectively. Companeramides were isolated from the marine cyanobacteria collection of Leptolyngbya sp. Additionally, companeramides were tested against the P. falciparum chloroquine-sensitive D6, chloroquine-resistant Dd2, and 7G8 strains, resulting in significant activity for companeramide B (IC₅₀ of 220–700 nM); however, companeramide A showed a specific activity in strain D6 (IC₅₀ of 500 nM). The compounds show three residues with different side chains, which is the main factor explaining the difference in activity. As for human cancer cell lines (non-small cell lung carcinoma NCIH460, breast adenocarcinoma MDA-MB-231, glioblastoma SF-295, and ovarian carcinoma cells SK-OV3), the study showed no significant cytotoxicity at 1 μM (Vining et al., 2015). The action mechanisms of companeramides are likely associated with the lipidic Amoya-Hmba moieties and their hydrophobical properties.

Isolated from the marine cyanobacteria Lyngbya majuscule, cyclic depsipeptide lagunamides A, B, and C are cytotoxic compounds with an unusual 27-membered ring system. These molecules showed activity against the P. falciparum drug-sensitive NF54 strain (IC₅₀ values of 190, 910, and 290 nM, respectively). Lagunamides A, B, and C contain five residues (N-M-Ala, Ile, N-M-Gly, N-M-Phe, and Ala) followed by a Hmpa moiety and different polyketide moiety for each lagunamide. The polyketide moieties in lagunamides A and B are highly similar molecules (Dtea1 and Dtea2); however, their difference causes lagunamide A to produce 4.78 more antiplasmodial activity than lagunamide B. In lagunamide C, the polyketide moiety is Dtuea—a compound slightly larger than Dtea1/2. These compounds exhibited cytotoxic activity with an IC₅₀ range of 2.1–24.4 nM against cancer cell lines (such as P388 murine leukemia, A549 lung carcinoma, PC3 prostate cancer, HCT8 ileocecal colorectal adenocarcinoma, and SK-OV ovarian cancer), inducing apoptosis by an unknown mechanism (Tripathi et al., 2012, 2011, 2010). Lagunamides belong to a family of cyclic depsipeptides isolated from marine cyanobacteria and mollusks (aurilides, lagunamides, palau'amide, kulokekahilide, and odoamide) with growth-inhibitory activities in cancer cell lines but with few studies on protozoan cells strains. Studies of the mechanisms in cancer cell lines showed activity in the caspase-mediated mitochondrial apoptosis pathway (lagunamide A) and the dissipation of mitochondrial membrane potential (Δφm) with the overproduction of reactive oxygen species (ROS). Aurilide inhibits the interaction of the prohibitin 1 protein (PHB1, UniProt ID: P35232·PHB1_HUMAN) with spastic paraplegia 7 metalloprotease (SPG7, UniProt ID: Q9UQ90·SPG7_HUMAN), resulting in the activation of SPG7. Active SPG7 cleaves the optic atrophy 1 GTPase (the dynamin-like 120-kDa protein OPA1, UniProt ID: O60313·OPA1_HUMAN), triggering the mitochondria-induced apoptosis pathway (Huang et al., 2016; Luo et al., 2019; Sato et al., 2011; Tripathi et al., 2012). We performed three searches on the BLASTp Suite program (Altschul, 1997) finding PHB1, SPG7, and OPA1 human homologs in diverse Plasmodium species: i) PHB1 putative Plasmodium homologs with 46–49% identity and a query cover factor of 91–97%; ii) SPG7 Plasmodium homologs (putative ATP-dependent zinc metalloprotease FTSH) with 42–44% identity and a query cover factor of 73–78%; and iii) OPA1, a Plasmodium homolog named dynamin-like protein (25–27% identity and low cover factor of ∼30%). These interesting data reveal that a similar mechanism likely mediating a conserved mitochondrial ancestral pathway, giving rise to antiplasmodial activity. Lagunamides are candidates to be redesigned in chemical libraries to obtain new derivatized chemical compounds to find drugs with antiplasmodial activity and reduced toxicity in host cells. On the other hand, compounds with homologous protein-associated mechanisms between Plasmodium and cancer cells could be applied to shared chemical libraries with antiplasmodial/anticancer drug screening to identify various selectivity properties against similar protein targets.

Lyngbyabellin A, ulongamide A, and kakeromamide B isolated from the marine cyanobacterium Moorea producens exhibited activity against the P. falciparum Dd2 strain. Notably, the activity was initially defined as EC₅₀ (half-maximal effective concentration) and later reported as IC₅₀ (half-maximal inhibitory concentration) (Sweeney-Jones et al., 2020). This natural collection's second most potent compound (Table 2) was lyngbyabellin A, a cyclic depsipeptide whose activity exhibited an EC₅₀ of 0.15 nM and IC₅₀ of 1.5 nM. This compound contains interesting moieties as a dichlorinated dClha and the thiazole Ile-like unit (Ile-Tzl-Ca) connected through a Gly to a second thiazole hydroxyl-Val-like unit (hVal-Tzl-Ca). About azole units, see the next Cyclamides section). The lyngbyabellin A sequence order is interpreted using the unique normal residue Gly, followed by Ile-Tzl-Ca, dClha, and hVal-Tzl-Ca. Lyngbyabellin A was initially isolated from Lyngbya majuscule and tested against smooth muscle A-10 cells line to observe filament-disrupting activity using rhodamine-phalloidin; however, whether the mechanism of action is similar in Plasmodium species remains unknown (Fathoni et al., 2020; Luesch et al., 2000; Sweeney-Jones et al., 2020). Ulongamide A depsipeptide is a small cyclic depsipeptide with an antiplasmodial activity of EC₅₀ of 990 nM and was initially isolated from Lyngbya sp. Notably, ulongamide A is structured by an N-M-Phe and N-M-Val connected to a thiazole Ala-like unit (Ala-Tzl-Ca) followed by a β-amino acid moiety Amha connected to Lac moiety closing the ring with the amino of N-M-Phe (Luesch et al., 2002, 2000). Isolated compounds are usually found in different species of cyanobacteria, thus highlighting the relevance of finding the most divergent cyanobacterial lineages with the hope of discovering the chemical diversity generated by evolution.

Table 2.

The most promising natural peptides and most relevant synthetic derivatives with better IC₅₀ values (<40 nM) from Cyanobacteria with anti-Plasmodium activity.

Compounds		Reference
Lyngbyabellin A: EC50: 0.15 nM; IC50: 1.5 nM; Pf: Dd2 Si: 12 000–2200/28.6–481.3; Ri: NA EC50: 1900-330 nM; HEK293T, HepG2 cells IC50: 43–722 nM; KB, LoVo cancer cell lines Tg: filament-disrupting activity, probably the actin network Seq: Gly: Ile-Tzl-Ca: dClha: hVal-Tzl-Ca Stage affected in RBC: Blood stages		(Luesch et al., 2000; Sweeney-Jones et al., 2020)
Dolastatin 10: IC50: 0.1 nM; Pf: FCH5.C2; Si: 0.59–5; Ri: NA IC50: 0.059–0.5 nM; P388, L1210 line cells Tg: Microtubule network, Mammalian cells is the Tubulin Seq: Dolavaline: Val: Dolaisoleuine: O-M-Dolaproine: DolaPhenine Stage affected in RBC: Trophozoite/Schizonts arrested		Fennell (2003)
Carmaphycin B: IC50: 4.10 nM nM asexual blood stage IC50: 160 nM sexual stage IC50: 61.6 nM liver stage; Pf: Dd2; Si: 3.07/0.07/0.2; Ri: NA IC50: 12.6 nM; HepG2 cell line Tg: 20S proteasome β5-subunit inhibitor Seq: NHex: L-Val: L-Met-On: L-Leu: EK Stage affected in RBC: Asexual and Sexual stages		(LaMonte et al., 2017; Li et al., 2012)
Carfilzomib: IC50: 25 nM on extracted subunits EC50: 28.8 nM Trophozoites EC50: 492 nM; Human foreskin fibroblast Pf: 3D7/D10 Tg: proteosomal beta5 subunit Seq: Morpholino:Phe: Leu: Phe: Leu:EK Stage affected in RBC: Asexual stage, Trophozoite
Carmaphycin B derivative 1: IC50: 4.11 nM; Pf: Dd2; Si: 33.3; Ri: NA IC50: 134 nM; HepG2 cell line Tg: 20S proteasome β5-subunit inhibitor Seq: NHex: L-Val: L-Nle: L-Leu: EK Stage affected in RBC: Asexual stage
Carmaphycin B derivative 5: IC50: 5.55 nM; Pf: Dd2; Si:14.5; Ri: NA IC50: 80.2 nM; HepG2 cell line Seq: NHex: 4BNH-L-Phe: L-Nle: L-Leu: EK Stage affected in RBC: Asexual stage
Carmaphycin B derivative 17: IC50: 9.38 nM; Pf: Dd2; Si:36.9; Ri: NA IC50: 346 nM; HepG2 cell line Seq: NHex: L-Trp: L-Trp: L-Phe: EK Stage affected in RBC: Asexual stage
Carmaphycin B derivative 18: IC50: 3.27 nM asexual blood stage IC50: 130 nM sexual blood stage Pf: Dd2; Si: 379.2; Ri: NA IC50: 1240 nM; HepG2 cell line Tg: 20S proteasome β5-subunit inhibitor Seq: NHex: D-Val: L-Nle: L-Leu: EK Stage affected in RBC: Asexual and sexual stages
Carmaphycin B derivative 19: IC50: 2.92 nM; Pf: Dd2; Si: 313.4; Ri: NA IC50: 915 nM; HepG2 cell line Tg: 20S proteasome β5-subunit inhibitor Seq: NHex: D-Trp: L-Nle: L-Leu: EK Note: Carmaphycin derivatives are not available in the databases. Stage affected in RBC: Asexual stage

Pf = P. falciparum strain; Tg = targe/activityt; Seq = peptide sequence.

Si = Selectivity index was calculated IC₅₀ from cell mamalian line/IC₅₀ from parasites. Any articles report maximum values in nM for mammalian cells lines toxicity, although is not an IC₅₀ values were considered for appreciate an aproximated index value. Ri = Resistance index was calculated IC₅₀ from parasites resistant line/IC₅₀ values of parasites sensitive line. NA = not acredited, not evaluated, experiment not carried out.

Kakeromamide B (isolated from Moorea producens) is a cyclic pentapeptide with an EC₅₀ of 890 nM against the P. falciparum Dd2 strain. This compound stimulated mammalian actin polymerization in a dose-dependent manner, suggesting an antiplasmodial homolog mechanism of lyngbyabellin A (Sweeney-Jones et al., 2020). Kakeromamide A is a related compound previously discovered in Moorea bouilloni, with a biological activity that induces the differentiation of neural stem cells into astrocytes (at 10 μM; in vitro model of mouse ES cells) but remains untested as an antiplasmodial compound. Kakeromamides share a similar structure sequence with a standard section of a Val, followed by two units of NO-M-Tyr connected to a thiazole unit Val-Tzl-Ca moiety and then followed by Amha for kakeromamide A. Differences in kakeromamide B are on the last residues showing an Ala-Tzl-Ca, followed by an Amoa moiety (Nakamura et al., 2018; Sweeney-Jones et al., 2020). These characteristics can explain the shared activity and probable shared mechanisms of action on the cytoskeleton (Nakamura et al., 2018). Although ulongamide A shares a structure with kakeromamides, no studies have supported the mechanism of action responsible for the activity.

4.1.2. Cyclamides: cyclic peptides with azole units (venturamides, aerucyclamides, microcyclamides, and balgacyclamides)

The cyclamides are cyclic peptides with thiazole (Tzl), thiazoline (Tzn), oxazole (Ozl), oxazoline (Ozn), methyl-oxazole (MOzl), and methyl-oxazoline (MOzn) moieties formed by aromatic heterocyclization reaction of one residue (Cys, Ser or Thr). The formation of the azole occurs when the side chain is included as an aromatic heterocyclic ring in the backbone with the substitution of the usual carbonyl moiety by an ether inside the ring; as a result of the reaction, the peptide bound with the adjacent residue is lost forming azole units with two residues; note that second residue contains their side chain intact. The peptide bounds connect the azole units as standard peptides (see Figs. S1, S2, and S3). Venturamides A and B are cyclic peptides isolated from the marine cyanobacteria Oscillatoria sp. and showed activity against the P. falciparum chloroquine-resistant W2 strain with IC₅₀ values of 8200 and 5200 nM, respectively (Tables 1 and S1). These peptides have exhibited IC₅₀ values of 86,000/56 000 and 13,100/>54,000 nM, respectively, against Vero cells and MCF-7 cancer cells (Linington et al., 2007). Venturamides are cyclic hexapeptides containing azole units with Tzl, Ozl, and MOzl moieties connected by peptide bonds; note that the chemical variations are in the side chains of the second residue of each azole unit. Venturamide A is formed by Ala-Tzl-Ca, Ala-MOzl-Ca, and Val-Tzl-Ca; venturamide B is formed by Thr-Tzl-Ca, Ala-MOzl-Ca, and Val-Tzl-Ca. Their IC₅₀ values suggest that venturamides require chemical optimization to enhance their activity/specificity against hitherto unknown plasmodial targets and explore the properties of azole moieties. (Fennell, 2003). Venturamide cyclic hexapeptides are related to the cyclic octapeptides patellamides produced by Prochloron didemni, a symbiotic cyanobacterium from the sea squirt Lissoclinum patella (Ascidiacea Class); unfortunately, the patellamides are untested against Plasmodium species. The biosynthetic pathway of venturamides and patellamides has been identified partially in the marine cyanobacterium Trichodesmium erythraeum and symbiotic cyanobacterium Prochloron didemni. Notably, these peptides are recognized for being chelating agents of Cu²⁺ and Zn²⁺ and for their cytotoxicity (Schmidt et al., 2005).

Aerucyclamides with antiplasmodial activity isolated from the freshwater cyanobacterium Microcystis aeroginasa PCC 7806 are cyclic hexapeptides containing azole units with Tzl, Ozl, Tzn, and MOzn moieties. Aerucyclamide A contains the novel azole combinations Gly-Tzl-Ca, Ile-MOzn-Ca, and Ile-Tzn-Ca and showed activity against the P. falciparum K1 strain (IC₅₀ of 5000 nM). Notably, aerucyclamide B has a better effect against P. falciparum K1 strain (IC₅₀ of 700 nM) structured by Gly-Tzl-Ca, Ile-MOzn-Ca, and Ile-Tzl-Ca. A unique difference concerning Aerucyclamide A is the oxidation state of the last moiety (Ile-Tzn-Ca change by Ile-Tzl-Ca). Aerucyclamides C and D showed activity against the P. falciparum chloroquine-resistant K1 strain (IC₅₀ values of 2300 and 6300 nM, respectively) but contained additional azole moieties. Val-Ozl-Ca, Ala-MOzn-Ca, and Ile-Tzl-Ca structure Aerucyclamide C, while aerucyclamide D structure by Gly-Tzl-Ca, Phe-MOzn-Ca, and Met-Tzn-Ca. Microcyclamides 7806A and 7806B are synthetically derived compounds of aerucyclamide C (prepared by acidic hydrolysis) that lack the Ala-MOzn-Ca moiety, replaced by two residues (Thr and Ala) with different statuses of atom connectivity; however, these two compounds were not evaluated as antiplasmodial compounds (Portmann et al., 2008).

Balgacyclamides A and B are cyclic hexapeptides containing azole units isolated from the freshwater cyanobacterium M. aeroginasa EAWAG 251 with activity against the P. falciparum K1 strain (EC₅₀ values of 900 and 820 nM, respectively), with no toxicity against L6 cells. Ile-Tzl-Ca, Val-MOzn-Ca, and Ala-MOzn-Ca structure Balgacyclamide A compared with Balgacyclamide B that maintains Ile-Tzl-Ca and Val-MOzn-Ca, and the last moiety is replaced by Thr and Ala residues, suggesting that activity could be in the azole units.

Balgacyclamide C is a similar structure solved but not evaluated as an antiplasmodial compound structured with Ile-Tzl-Ca, Phe-MOzn-Ca, and the last moiety replaced by Thr and Gly residues (Portmann et al., 2014). Macrocycle 4, a predicted analog of aerucyclamide B, is structured by Gly-Tzl-Ca, Ile-MOzl-Ca, and Ile-Tzl-Ca (a moiety contained in lyngbyabellin A), was synthesized and exhibited enhanced activity against the P. falciparum K1 strain (IC₅₀ of 180 nM) (Peña et al., 2012). The biological activities, mechanism, and protein targets of cyclamides against Plasmodium species and cancer cell lines are unknown. However, studies of azole-based compounds in P. falciparum exhibited the inhibition of lactate dehydrogenase—a key enzyme in glycolysis—coupled with homolactic fermentation (Cameron et al., 2004). The targets of azole cyclic hexapeptides could be the catalytic sites of diverse enzymes using dinucleotides, with the side chains of the moieties being the main factors for specificity/discrimination. Although the same argument would apply to other azole-cyclized peptides, the compounds require optimization and selectivity-mediated modifications with effective azole molecules (e.g., pyrazole) and the exploration of another moiety on the side chains.

4.1.3. Other modified peptides

Gallinamide A is a non-cyclic pentapeptide isolated from Schizothrix sp. with activity against the P. falciparum W2 strain (IC₅₀ of 8400 nM) that has four modified residues. The sequence with N-diM-Ile, Ica, L-Leu, Apa, and Mmp moieties structure this compound similar to dolastatins, compounds with antiplasmodial and antiproliferative effects in mammalian cells (Linington et al., 2009). Dolastatins 10 and 15 (microtubule inhibitors) were initially found on Dolabella auricularia (a mollusk) and tested against the P. falciparum FCH5.C2 strain (IC₅₀ values of 0.1 and 200 nM, respectively) (Fennell, 2003). In the years that followed, diverse dolastatins and analogs (e.g., symplostatin 1 and somamida A) were discovered in marine cyanobacteria from the genera Lyngbya and Symploca, suggesting that the compounds come from the mollusks’ diets. Notably, the dolastatin 10 compound showed effectiveness against P. falciparum with an IC₅₀ of 0.1 nM (Table 2; the compound with the highest effect in this review), presumably with the same target observed on diverse mammalian cell lines (microtubule network disruption) and structured with dolavaline connected with an ordinary Val followed by three unusual residues: dolaisoleuine, O-M-dolaproine, and dolaphenine (Gao et al., 2021). Dolastatin 15 is similar but with less potency (IC₅₀ 200 nM), with the sequence dolavaline, Val, Val, Pro, and Pro connected to Hmba followed by a moiety Benp (similar to dolaphenine). The effects and potential mechanism of action of dolastatin 10 can be revised at a structural level on the 7TR3 structure of the PDB database (Fig. 2, panel B) in a complex with a mammalian tubulin ring (Hunter et al., 2022; Luesch et al., 2002; Ratnayake et al., 2020). The tubulin ring 7TR3 is an electron microscopy structure formed by two subunits, the Tubulin alpha-1B chain and Tubulin beta-2B chain subunits (UniProt ID: Q2XVP4·TBA1B_PIG and P02554·TBB_PIG) from Sus scrofa in complex with a Kinesin-like protein from Candida albicans (UniProt ID: A0A1D8PKA4·A0A1D8PKA4_CANAL) and with dolastatin 10 bound. Additional structures with dolastatin 10 analogs deposited on the PDB and obtained by x-ray diffraction are available (PDB ID: 4X1K and 4X20).

Carmaphycins A and B are modified tripeptides isolated from Symploca sp. cyanobacteria. Carmaphycin B showed activity against asexual, liver, and sexual stages of the P. falciparum Dd2 strain (Table 2; IC₅₀ values of 4.1, 61.6 and 160 nM respectively) and consists of L-Val, Met-On, and L-Leu, while the peptide capped on the N- and C-terminals with an NHex group and α, β-epoxyketone (EK) group, respectively. Carmaphycin B differs from the Met-On concerning carmaphycin A, which contains a Met-Ox—a different oxidation state of the Met. Carmaphycins are 20S proteasome β5-subunit inhibitors such as carfilzomib and PR3 compounds (synthetic tetrapeptides). Carfilzomib is used to combat multiple myeloma and is active against P. falciparum at the asexual blood stage (IC₅₀ of 28.8 nM). The structural similarity in the L-Leu:EK and hydrophobic patterns between carfilzomib, PR3, and carmaphycin drugs suggest similar mechanisms of action on the 20S proteasome (Table 2. Fig. 2, panel C). New antimalarial compounds based on the carmaphycins were synthesized and reported with enhanced selectivity and activity against the P. falciparum 20S proteasome. Twenty new carmaphycin derivatives were generated and tested against the P. falciparum Dd2 strain, with similar or slightly increased activities being observed (for derivatives 1, 2, 3, 6, 8, 16, 18, and 19, with IC₅₀ values ranging between 2.92 and 4.11 nM). The carmaphycin B derivatives 18 and 19 were the most promising compounds with the lower IC₅₀ values (Table 2; IC₅₀ values of 3.27 and 2.92 nM, respectively) and with the best selectivity index (Si values of 379.2 and 313.4), with the substitutions of the central L-Val and Met-On residues by D-Trp and L-Nle while maintaining the capped N- and C-terminals with NHex and L-Leu:Ek. Six carmaphycin B derivatives with the best IC₅₀ values contain NHex, L-Nle, and L-Leu:Ek moieties, while two contain NHex, 4-Pyr-LAla, L-Nle, and Ek moieties (LaMonte et al., 2017; Li et al., 2012). The mechanism of action of carmaphycins and carfilzomib can be reviewed at a structural level on the 4HRC and 4QW4 structures of the PDB database. 4HRC and 4QW4 crystal structures of 20S proteasome from Saccharomyces cerevisiae were obtained by x-ray diffraction in complex with carmaphycin analog 3 and carfilzomib. The 20S proteasome is a protein complex constituted by 14 different proteinases (α and β-type subunits forming α and β-rings) with a total of 28 subunits (2 of each type). Each ring consists of seven subunits and is arranged as follows α-β-β-α in a cylindrical-like structure. The carmaphycin analog 3 and carfilzomib compounds bound to β-type 1, 2 and 5 subunits from the proteasome located in the β-rings (UniProt ID: P30656·PSB5_YEAST, P38624·PSB1_YEAST and P25043·PSB2_YEAST). Additional structures can be downloaded from the PDB database from S. cerevisiae 20S proteasome in complex with EK derivatives, such as 1G65 and 4NO9.

The linear alkynoic lipotetrapeptides (with N-methyl amide residues) isolated from the marine cyanobacterium L. majuscula Gomont (Panamanian strain) Dragomabin, Dragonamide A (known as Dragonamide 3), and Carmabin A were tested against the P. falciparum chloroquine-resistant W2 strain (IC₅₀ values of 6000, 7700, and 4300 nM, respectively). Dragonamide B was inactive, showing a side chain of Val replacing the unique Phe of Dragonamide A. Carmabin A shows the moiety Mdya on the N-terminal, and the other three show the moiety Moya. Moreover, the C-terminals of the four peptides are amidated. Carmabin A was more cytotoxic to the Vero cell line (IC₅₀ 9800 nM) than dragomabin (IC₅₀ 182 300 nM) or dragonamide A (IC₅₀ 67 800 nM) (McPhail et al., 2007). These alkynoic lipotetrapeptides contain four types of hydrophobic residues (only N-M-Phe, N-M-Ala, N-M-Val, and NO-diM-Tyr), with Mdya and Moya producing highly hydrophobic modified peptides; however, the potential mechanisms of action and targets remain unknown.

The lipopeptide malyngamide X (a fatty acid amide) showed activity against the P. falciparum K1 strain (expressed as ED₅₀ of 5440 nM), and the sequence structured by the moieties Mtea, N-M-Ala, Mpna, and Lpmp isolated from the mollusk Bursatella leachii. Malyngamide X is likely of cyanobacterial origin stemming from the mollusk's diet. Malyngamide X showed similar cytotoxicity against human epidermoid carcinoma of the nasopharynx (KB), human small-cell lung cancer (NCI-H187), and breast cancer cell lines with ED₅₀ values of 8200, 4120, and 7030 nM, respectively (Suntornchashwej et al., 2007). Although malingamides are compounds synthesized by cyanobacteria, malyngamide X is a unique compound of the family tested against Plasmodium and found in a mollusk (Ozaki et al., 2019). Mollusks are known to collect, select, and accumulate specific metabolites as a defense by cleptochemistry, so their capture or breeding could be advantageous; for example, sea hares fed with specific cyanobacteria could accumulate new pre-selected compounds with biological activities. Malyngamides have been described with some biological activity against tumoral, carcinogenic, and leukemic cells, as well as anti-HIV, anti-inflammatory, and anti-nociceptive activity (Li et al., 2021; Villa et al., 2010). Recently, bacterial riboflavin synthase—an enzyme present only in bacteria, plants, and fungi—was proposed as the target protein of malyngamide V using in silico predictions but was not tested (Alturki et al., 2022). However, the protein targets in Plasmodium species remain to be discovered or predicted as the theoretical function of the compound family.

Mabuniamide is a lipopeptide with entirely hydrophobic residues isolated from Okeania sp. marine cyanobacterium. It has a moiety of Bta connected to the N-terminal of the sequences N-M-Ala, Pro, N-M-Phe, Leu, N-M-D-Val, Val, Gly, N-M-Phe, and N-M-Pro, with an IC₅₀ of 1400 nM against the P. falciparum 3D7 strain. Mabuniamide stimulates glucose uptake in skeletal muscle cell line (L6 myoblasts) cultures and showed no cytotoxicity at 10–40 μM (Ozaki et al., 2019). Although no further functional studies have been conducted, the effect on the L6 myoblasts cell line suggests the intervention of glucose uptake homolog proteins on P. falciparum. Ikoamide is another octalipopeptide isolated from the Okeania sp. with an IC₅₀ of 140 nM against the P. falciparum 3D7 strain and no cytotoxicity at 10 μM on human cancer cell lines. The Ikoamide has a moiety Dmo bound in the N-terminal with D-amino acids (N-M-D-allo-Ile, D-Ser, and D-Leu), followed by N-M-Leu, L-Thr, N-M-Ile, N-M-Gln and NO-diM-L-Tyr-O-M with the C-terminal methylated (Iwasaki et al., 2020). The cyclic dodecapeptides tychonamide A and B were isolated from the marine cyanobacterium Tychonema sp. with activity against the P. falciparum K1 strain (IC₅₀ values of 988 and 1235 nM, respectively) and L6 myoblasts cell line (IC₅₀ values of 2764.5 and 3315.7 nM, respectively). These compounds were evaluated against several cancer cell lines exhibiting cytotoxic activity (mean IC₅₀ values of 201.79–874.42 nM and 686.48–10228.6 nM, respectively). The tychonamides structures contain D/L residues, an unusual residue, and the moiety β-amino acid Atpoa. Additionally, Atpoa is bound to an external residue N-A-M-Met-Leu. Atpoa connects residues in the following order: D-Gln, Gly, L-Pro, L-Pro, D-allo-Ile, L-Ser, Dhb, L-Thr, O-M-D-Hty/D-Hphe (tychonamide A and tychonamide B, respectively), and L-Pro for closing the ring. Moreover, Atpoa has the interesting characteristic of allowing three residues to be connected simultaneously (an external and two on the ring) and could thus be recruited as a scaffold compound (Mehner et al., 2008). The studies of these compounds are limited, and their mechanisms of action remain unknown.

The non-cyclic hoshinoamides A, B, and C are lipopeptides with unusual TYR-like moieties isolated from the marine cyanobacterium Caldora penicillata, and they inhibited the growth of the P. falciparum 3D7 strain with IC₅₀ values of 520, 1000, and 960 nM, respectively. Hoshinoamides at 10,000 nM do not cause toxicity effects in HeLa cells. Hoshinoamides are similar peptides with differences observed in the moieties bound to the N-terminal, combinations of Hba moiety, and compounds derived from fatty acids (Aha, Ana, and Amoc moieties) bound to N-terminal Leu residues of each hoshinoamide. Hoshinoamides A and B contain seven central residues with similar sequence (L/D residues) and methylation patterns, while hoshinoamide C contains six residues and differs in sequence and methylation pattern from hoshinoamides A and B. Hoshinoamide compounds have been little studied, and is desirable to uncover the mechanism of action and improve the effectiveness and selectivity-mediated synthetic modifications (Iwasaki et al., 2021, 2018).

4.2. Alkaloids produced by cyanobacteria

Marine and freshwater cyanobacteria produce other secondary metabolites as alkaloids derived from amino acids (Table 3 and Table S2). The chlorinated alkaloid nostocarboline isolated from the freshwater cyanobacterium Nostoc 78-12A inhibits the acetyl/butyryl-cholinesterase and trypsin proteins. Nostocarboline showed toxicity against the P. falciparum K1 strain (IC₅₀ of 194 nM). Nostocarboline dimers (bis-β-carbolinium homo-dimers) were synthesized using 10 symmetrical dihalogen linkers and tested against the P. falciparum K1 strain. The new compounds increase the antiplasmodial activity in eight of the derivatives (IC₅₀ of 14–121 nM) and increase the selectivity index in compounds 7 and 8 (2625/1810, respectively) (Barbaras et al., 2008; Becher et al., 2005). Subsequently, the same research group expanded the nostocarboline derivatives and additional tests on the P. berghei mouse model for in vivo evaluation and in vitro verification against P. falciparum K1 erythrocytic stages. The P. berghei parasitemia model explores the activity of the compound administered intraperitoneally for four consecutive days; parasitemia was determined on day four post-infection (24 h after the last treatment). The nostocarboline dimers (previously described) show antiplasmodial activity in vitro but weak or no activity in the P. berghei parasitemia mouse model, resulting in the natural nostocarboline compound being the best for in vivo activity but less effectiveness for in vitro activity. This effect is interesting to elucidate since the natural product is the best option (perhaps this molecule can pass the membranes but not the dimers). Additional derivatives tested, such as the nostocarboline monomer (with substitutions of the 2-methyl by N-Alkyl derivatives) and closely related compounds, including eudistomin N and eudistomin 8 (8-brome-β-carboline, although it is named eudistomin O in the article of Bonazzi et al., the last compound in PubChem is 7-brome-β-carboline) and its derivatives. The nostocarboline monomer derivatives showed weak effects on P. falciparum K1, and two derivatives of eudistomin N showed elevated activity and notable selectivity against the same strain (Table 3): the compound eudistomin N derivative 23 and eudistomin N derivative 24, which have an IC₅₀ range of 18–32 nM and 4783/2443 selectivity index, respectively (Bonazzi et al., 2010). These experiments suggest that the halogenated position of nostocarboline and the compounds eudistomin N derivatives 23 and 24 are key, as well as the 2-methyl position, which enhances the activity of these compounds. Eudistomins are natural marine compounds isolated from ascidian animals but have structural similarities with nostocarboline. Nostocarboline was isolated from cyanobacteria and showed strong algicidal effects; however, eudistomins compounds could be synthesized by the symbiotic cyanobacterium of the ascidian. Nostocarboline and the brominated equivalent were produced by feeding Nostoc 78–12A with precursors (Portmann et al., 2009).

Table 3.

Alkaliods and other compounds produced by Cyanobacteria with better IC₅₀ values (<100 nM) anti-Plasmodium activity.

Compounds		Reference
Nostocarboline: IC50: 194 nM; Pf: K1; Si: 622; Ri: NA IC50: 120 900 nM; Rat myoblast L6 cells Tg: probably inhibition of hydrolases and IkB kinase/Casein kinase II type enzymes. Stage affected: not described		Barbaras et al. (2008)
Nostocarboline dimers design All were tested against Pf: K1 The linkers chains: Nostocarboline dimer 5: IC50: 56 nM; Si: 408; Ri: NA IC50: 23 000 nM; Rat myoblast L6 cells Nostocarboline dimer 6: IC50: 18 nM; Si: 2625; Ri: NA IC50: 479 000 nM; Rat myoblast L6 cells Nostocarboline dimer 7: IC50: 20 nM; Si: 1810; Ri: NA IC50: 36 200 nM; Rat myoblast L6 cells Nostocarboline dimer 8: IC50: 18 nM; Si: 423; Ri: NA IC50: 7500 nM; Rat myoblast L6 cells Nostocarboline dimer 9: IC50: 14 nM; Si: 575; Ri: NA IC50: 8200 nM; Rat myoblast L6 cells Nostocarboline dimer 10: IC50: 23 nM; Si: 186; Ri: NA IC50: 4300 nM; Rat myoblast L6 cells Rn = Nostocarboline molecule Note: not available in the databases Stage affected: not described
Eudistomin N derivative 23 R = CH3:		Bonazzi et al. (2010)
IC50: 18 nM; Pf: K1; Si: 4783; Ri: NA
IC50: 86 100 nM; Rat myoblast L6 cells
Eudistomin N derivative 24 R = C2H5:
IC50: 32 nM; Pf: K1; Si: 2443; Ri: NA
IC50: 78 800 nM; Rat myoblast L6 cells
Tg: probably inhibition of hydrolases and IkB kinase/Casein kinase II type enzymes.
Stage affected: erythrocytic stages
Calothrixin A (1a):		Rickards et al. (1999)
IC50: 58/185 nM; Pf: FAF6/FCR-3
Si: 0.68; Ri: 0.31
IC50: 40 nM; cervical cancer HeLa cells
Tg: probably DNA topoisomerase I
Stage affected: erythrocytic stages
Bastimolide A:		(Shao et al., 2015, 2018)
IC50: 80 nM; Pf:TM90-C2A; Si: 26/38; Ri: 0.03
IC50: 90 nM;Pf: TM90-C2B; Si: 23/34; Ri: 0.03
IC50: 140 nM; Pf: W2; Si: 15/22; Ri: 0.05
IC50: 270 nM; Pf: TM91-C235; Si: 7/11; Ri: 0.1
IC50: 2600 nM; Pf:HB3; Si: 0.8/1.19; Ri: NA
IC50: 2100/3100 nM; Vero/MCF-7 cell lines
Tg: The action mechanism remained undetermined
Stage affected: erythrocytic stages

* Pf = P. falciparum strain; Tg = target/activity; Si = Selectivity index; Ri = Resistance index; NA = not acredited, not evaluated, experiment not carried out.

Santacruzamate A is a carbamate derivative (Table 1) compound similar to histone deacetylase (HDAC) inhibitor SAHA used to treat refractory cutaneous T-cell lymphoma. Santacruzamate A was isolated from a non-well-characterized marine cyanobacteria (a dark-brown tuft-forming species) closely related to the Symploca genus. Santacruzamate A has activity against HCT-116 colon cancer and HuT-78 cutaneous T-cell lymphoma (reported growth inhibition (GI₅₀) values of 29,400 and 1400 nM, respectively) and demonstrated slightly better activity against cutaneous T-cell lymphoma than SAHA (GI₅₀ of 3000 nM). The initial extract fraction H exhibited activity against the P. falciparum chloroquine resistant Indochina W2 strain growth (at 10 μg/ml). The posterior fraction eluting with a solvent gradient of 1:1 MeOH–H₂O was subjected to Reverse Phase-High Performance Liquid Chromatography (RP-HPLC) purification, resulting in santacruzamate A isolation (Pavlik et al., 2013). Although the santacruzamate A protein target in Plasmodium was not determined, the hypothetical target is the homolog protein HDAC from the parasite since SAHA binds to the active site of the mammalian HDAC protein. However, additional research is required to verify this prediction and will be necessary for a chemical library to generate specificity.

The indolo[3,2-j]phenanthridine alkaloids calothrixins A and B—isolated from the cyanobacterium Calothrix sp. — are polycyclic compounds with activity against the P. falciparum FAF6 strain, with IC₅₀ values from 58 to 180 nM respectively (Rickards et al., 1999). N-alkyl derivatives from calothrixins A and B were synthesized and tested for activity against the P. falciparum FCR-3 strain (Table 3 and S2). This strain is less sensitive to calothrixins A and B, with IC₅₀ values of 185 and 120 nM, respectively. Among the derivatives, the best candidate was calothrixin B 2c, with an IC₅₀ value of 220 nM (Matsumoto et al., 2012). O-methyl-calothrixin A and N-methyl-calothrixin B derivatives, as well as calothrixins A and B, have demonstrated the capacity to inhibit human DNA topoisomerase I—apparently the main effect against human leukemia CEM cell line tested (IC₅₀ values ranging from 200 to 5130 nM). Although none of these derivatized O-methyl-calothrixin A and N-methyl-calothrixin B derivatized compounds have been tested on antiplasmodial activity, these data suggest the possible mechanism of these compounds on P. falciparum (Khan et al., 2009).

Many compounds isolated from cyanobacteria have shown activities against parasites, such as tumonoic acids isolated from the Lyngbya majuscula/Schizothrix calcicola assemblage and Blennothrix cantharidosma (unaccepted by synonym: Blennothrix cantharidosmum Komárek, formerly Hydrocoleum Gomont). Tumonoic acids are acyl proline/N-acyl-amino acid derivatives with a fatty acid on the N-terminal, with variations such as 2-methyl decanoic and octanoic acid (Octa), all of which are small depsipeptides of one residue of Pro with moieties (for this reason, we place them in this section). The proline on the C-terminal contains combinations of hydroxy acid derivatives of Ala, Ile, or Val of known moieties (e.g., Lac, Hmpa, and Hmba) to generate ten tumonoic acids and two methylated variations. Only tumonoic acid I shows activity against P. falciparum (an unknown strain), with an IC₅₀ of 2000 nM (sequence: Octa:Pro:Hmpa:Hmpa). Tumonoic acids D to H have reported IC₅₀ values for concentrations of up to 10 μg/ml (35,285–20678 nM), while tumonoic acids A to C are not reported to have antiplasmodial activity (Clark et al., 2008; Khan et al., 2009). Only one compound shows antiplasmodial activity, but studies of the potential mechanisms still need to be completed.

4.3. Other compounds produced by the secondary metabolism of cyanobacteria

The malyngolide dimer (Table 1) is a natural δ-lactone with two aliphatic chains attached to the δ-position of the lactone ring isolated from the marine cyanobacterium L. majuscula and derived from the malyngolide unit isolated from the same cyanobacterium. Malyngolide dimer has an IC₅₀ of 19,000 nM against the P. falciparum chloroquine resistant W2 strain and an IC₅₀ of 9000 nM against H-460 human lung tumor cells, resulting in more toxicity for these cells than for P. falciparum (Gutiérrez et al., 2010). The action mechanism and targets for malyngolide and its derivatives remain unknown.

Hierridin A and B were isolated like an inseparable mixture (by HPLC with an unknown proportion of each one) from the methylene dichloride extract of the marine cyanobacterium Leptolyngbya ectocarpi (unaccepted by synonym: Phormidium ectocarpi). The initial methylene dichloride extract has shown antiplasmodial activity (IC₅₀ of 2.1 μg/ml). The inseparable mixture was isolated and tested against the P. falciparum chloroquine resistant W2 strain and chloroquine-sensitive D6 strain, with IC₅₀ values of 5.2 and 3.7 μg/ml (a hypothetical 50/50 hierridin mixture with an average MW: IC₅₀ values of 13,734 and 9772 nM, respectively), the differences between the extract and purified mixture are indicative of the presence of additional compounds working synergically (Papendorf et al., 1998). Hierridins are methylated hydroquinones with a long alkane chain and target the mitochondrial function in cancer cell lines (Freitas et al., 2016). Hydroxynaphthoquinones such as lapachone and hydrolapachone show antiplasmodial activity by oxidative stress via ROS production, and hierridin likely produces a related action mechanism; however, the mechanism should be experimentally validated (Carneiro et al., 2016).

Different chlorinated aromatic molecules can be produced by the terrestrial cyanobacterium Fischerella ambigua when cultured in diverse mediums—for example, BG-11 medium produced ambigol A and C compounds and Z-medium ambigol B. The main effects against the P. falciparum K1 and NF54 strains are from the compounds ambigol A (IC₅₀ values of 1744.3 and 3346 nM respective to each strain) and ambigol C (IC₅₀ values of 3169.1 and 5150 nM respective to each strain). Ambigol B was found to be relatively inactive against the P. falciparum chloroquine sensitive D6 and chloroquine resistant W2 strains (IC₅₀ value > 20618.6 nM for both) (Falch et al., 1995; Wright et al., 2005). Notably, some discrepancies in the literature—especially with ambigol A—make it necessary to re-evaluate the data and action mechanism.

Many new molecules were extracted from the marine cyanobacteria genus Okeania as modified peptides (Section 4.1.3.) and macrolides, such as bastimolides A and B (isolated from O. hirsute), which exhibit antiplasmodial activity. Bastimolides are described as 40-membered-ring polyhydroxy macrolides with a tertbutyl terminus, with the differences between A and B resulting from the ring's cycling position. Bastimolide A showed activity against the P. falciparum-resistant strains TM90-C2A, TM90-C2B, TM91-C235, and W2 in the nM range (IC₅₀ of 80–270 nM) and against the activity of the P. falciparum chloroquine-sensitive HB3 strain in the μM range (IC₅₀ of 2600 nM) (Table 3). The derivatives bastimolide A-1i, bastimolide A-1j, and bastimolide A-1k show similar activities to bastimolide B against the P. falciparum chloroquine-sensitive strain HB3 in the μM range (IC₅₀ of 20,000–9700 and 5700 nM, respectively) (Table S2). Moreover, the 2-(E)-bastimolide A isomer (C2–C3 double-bond isomerization) showed better activity against the P. falciparum chloroquine-sensitive HB3 strain, albeit also in the μM range (IC₅₀ of 1400 nM) (Shao et al., 2015, 2018). The bastimolides structure is an interesting lead compound (like antibiotic 14- to 16-membered-ring macrolides) with the potential to generate large chemical libraries. The macrolides are an extensive group, and diverse cyanobacterial versions, such as palstimolide A isolated from the marine Leptolyngbya sp., with a structure related to bastimolide A (and others not tested against Plasmodium, such as nuiapolide and amantelides). Palstimolide A demonstrates activity against the P. falciparum Dd2 strain (IC₅₀ of 172.5 nM) and toxicity to liver HepG2 cells (IC₅₀ of 5040 nM). The difference between palstimolide A and bastimolide A is in the arrangement and number of hydroxy groups, while the action mechanism of these compounds remained undetermined (Keller et al., 2020).

18E-lyngbyaloside C and lyngbyaloside—isolated from the marine cyanobacteria Moorea producens—are small hydroxyl 14-/16-membered macrolides that are brominated, embedded with 2,6-cis tetrahydropyran ring (oxane), and linked to a 2,3,4-tri-O-methyl-6-deoxy-R-mannopyranoside, with activity against the P. falciparum Dd2 strain expressed as a half-maximal effective concentration: EC₅₀ of 1900–790 nM. Lyngbyalosides contain the Br linked on a side chain described as bromobuta-1,3-diene moiety. In contrast, 18E-lyngbyaloside C contains the Br linked to a side chain described as a bromohexa-1,3-diene moiety. Both compounds were previously isolated from Moorea bouilloni (unaccepted: Lyngbya bouilloni). 18E-lyngbyaloside C and lyngbyaloside were tested against HEK293T and HepG2 cell lines, resulting in similar toxicity with EC₅₀ values of 3100–1300 nM (Chang et al., 2015; Klein et al., 1997; Sweeney-Jones et al., 2020). These kinds of brominated macrolides have received limited research attention but represent interesting structures without the pyranose as a scaffold.

Biselyngbyaside (BLS) and biselyngbyolide B (BLLB) are hydroxyl 18-membered macrolides with a hydrophobic side chain ((4E)-2-methylhexa-1,4-diene) isolated from the marine cyanobacterium Lyngbya. The presence of a pyranose ring on BLS differentiates both compounds. Notably, BLS showed activity against the P. falciparum chloroquine-resistant K1 (IC₅₀ of 3400 nM) and chloroquine-sensitive FCR3 (IC₅₀ of 4400 nM) strains. In contrast, BLLB showed reduced antiplasmodial activities against the same strains (IC₅₀ values of 24.0 and 23.5 μM, respectively). In contrast, BLS and BLLB compounds showed effective cytotoxicity against the MRC-5 regular cell line (IC₅₀ values of 400 and 230 nM, respectively). BLS, BLLB, and cyclopiazonic acid (CPA) are inhibitors of mammalian SERCA protein (Ca2+ ATPases responsible for pumping the cytosolic Ca2+ into the sarcoplasmic reticulum), the crystal structures of the SERCA protein from Oryctolagus cuniculus (PDB ID: 4YCN, 4YCM, and 3FPB; UniProt ID: P04191·AT2A1_RABIT) in complex with BLS, BLLB and CPA have been solved (Sato et al., 2018). Their structures show BLS, BLLB (Fig. 2, panel D), and CPA molecules bound on the same cavity of mammalian SERCA protein. CPA is a small molecule concerning BLS and BLLB macrolides and is observed bound in the same cavity but occupies only a minimal fraction. In contrast, the macrolides occupy a large section of the cavity. The site of CPA is occupied by BLS and BLLB, suggesting to be a sensible region for the activity of SERCA proteins. CPA inhibits the homologous of SERCA in P. falciparum (PfATP4 and PfATP6) and may also inhibited by BLS and BLLB (Morita et al., 2015). Based on the aforementioned data, a homology model of PfATP6 and the comparative analysis of SERCA complexes were used to design three new compounds (BLLB 3, BLLB 4, and BLLB 5), which are derivatives based on substitution of the side chain outside the BLLB ring with 2-methylprop-2-en-1-ol, N-(2-methylprop-2-en-1-yl) acetamide and N-ethyl-2-methylprop-2-enamide, respectively. The new compounds were synthesized and tested against P. falciparum with a slight improvement in the IC₅₀ values of BLLB 4 and 5 (IC₅₀ values of 14,000 and 19,000 nM, respectively) (Sato et al., 2018). This kind of work is complex and requires sophisticated approximations to design compounds. At the same time, additional in silico tests (molecular modeling techniques) are needed to obtain better predictions, even though it is an interesting way to build predictive theoretical chemical libraries with demanding analysis of the spatial orientations to predict molecular binding modes.

5. Discussion

Natural products of cyanobacterial origin (marine and freshwater) have become an important source of biologically active agents, providing novel compounds with potential uses in drug research. The biological activities found on cyanobacterial compounds cover a broad range of activities and notable structural diversity with novel moieties fragments and scaffolds (see Figs. S1, S2, and S3 for all peptide moieties collected here). Of all the compounds contained herein (total in tables (S1 and S2): 106 tested and 5 not tested; in addition to numerous related compounds not tested), only two showed antiplasmodial activity, with IC₅₀ values of <2 nM exhibited by the depsipeptide type molecules dolastatin 10 and lyngbyabellin A (representing the 1.9% of the total tested purified compounds compiled herein). However, dolastatin 10 and lyngbyabellin A share the capacity to produce interventions on the cytoskeletal network with different targets and mechanisms of action. Dolastatin 10 binds to the tubulin curved ring (PDB: 7TR3), and lyngbyabellin A disrupts the actin network; however, the binding site remains unknown. In anticancer research was uncovered the compound monomethyl auristatin E (a synthetic analog of dolastatin 10) that is used clinically as a component of approved monoclonal antibody-drug conjugates (e.g. brentuximab antibody) (Doronina et al., 2003). Unfortunately, no similar development based on dolastatin 10 against P. falciparum is approved. An inspection of the molecular formulas of the two compounds uncovered a few similarities (e.g., the presence of the thiazole ring and Ile side chain on dolastatin 10 and lyngbyabellin A), highlighting the different chemical connectivity. While lyngbyabellin A is a cyclic depsipeptide with many modifications and chlorinated (dClha), dolastatin 10 is a linear peptide maintaining all the peptide bonds with discrete modifications (Hunter et al., 2022; Luesch et al., 2000). On the other hand, lyngbyabellin A showed a better selectivity index ratio (Si: 28–481) than dolastatin 10 (Si: 0.59) but is not considered a drug of use.

In an effort to classify the IC₅₀ nM activities of collected compounds, we selected the IC₅₀ value of <40 nM like a cutting separation because we considered that it could be appropriate for selecting antimalarial candidates from natural sources (in the range of “potential utility” according to IC₅₀ values). In general, IC₅₀ values of ≤1 μM are considered highly active compounds (approximately 50% of the compounds compiled here), but not all compounds are suitable candidates because they require improvements. For this criterion, we based on the clinical work described by Wakoli and Cols (Wakoli et al., 2022), in which clinical P. falciparum isolates (2008–2021) were tested for in vitro susceptibility to antimalarial drugs (piperaquine, arthemeter, dihydroartemisinin, chloroquine, and lumefantrine) and reported the IC₅₀ values (expressed as median IC₅₀s interquartile range). In addition to clinical isolates, P. falciparum strains, such as the chloroquine-resistant W2 strain, the chloroquine-sensitive D6 strain, and the drug-sensitive 3D7 strain, were tested. Temporal changes (2008–2021) of IC₅₀s values of these isolates remained unchanged by dihydroartemisinin, arthemeter, and chloroquine, with values being <20 nM. The P. falciparum W2, D6, and 3D7 strains controls observed similar behavior in dihydroartemisinin and arthemeter (IC₅₀s values of <20 nM). In clinical isolates of P. falciparum concerning lumefantrine IC₅₀s values, a trend of increasing resistance is observed, with values of <20 nM rising to <40 nM in 2018–2021. Clinical isolates of P. falciparum concerning piperaquine IC₅₀s values remained unchanged in <40 nM since 2008. Regarding the IC₅₀s values of the P. falciparum chloroquine-resistant W2 strain, resistance of <80 nM for chloroquine was observed, suggesting a cutting limit as regards the effectiveness of new compounds; however, the resistance of each compound may differ. The clinical P. falciparum isolates monitoring individual IC₅₀ values (outside the interquartile) on the range of chloroquine-resistant W2 strain for all compounds until 100 nM, suggesting that IC₅₀ values 20–100 nM range represent the trend of resistance and loss of effectiveness (Wakoli et al., 2022). Based on the above, we propose a simple activity classification against P. falciparum strains for the compounds collected in this work. Compounds with IC₅₀ < 40 nM are in the range of “potential utility” (most promising compounds), and compounds between 40 and 100 nM are in the range of the trend of resistance and loss of effectiveness. However, other parameters need to be considered, such as selectivity index ratio (Si), which is used to define differences in toxicity among cells.

In the range of “potential utility” according to the IC₅₀ values (most promising compounds), 25 compounds were found (IC₅₀ values of 2–40 nM, 23% of total), corresponding to nostocarboline dimer derivatives 6 to 10, eudistomin N derivative 23 to 24, carmaphycin B, and carmaphycin B derivatives 1 to 8, 10, 11 and 14 to 20, all of which are synthetic compounds except for carmaphycin B. Carfilzomib, a synthetic tetrapeptide, is an irreversible proteasome inhibitor used on chemotherapy that shares the same EK moiety with natural compounds carmaphycins and its synthetic derivatives, as well as the same type of biological activity. The EK moiety causes the inhibition of the 20S proteasome β5-subunit with an irreversible binding mode, and the residues contribute to the species selectivity. The tested carmaphycin B derivative compounds resulted in the discovery of eight chemical variations with IC₅₀ values slightly better than or similar to those of original natural carmaphycin B. A similar research work was developed from carmaphycin B, obtaining the carmaphycin-17 with specificity against Trichomonas vaginalis, thus demonstrating the efficacy of applying the chemical libraries (O'Donoghue et al., 2019). The libraries produced changes in the functional group of the side chains of the tripeptide attached to the EK moiety without exploring changes in the N-terminal moiety NHex. Carmaphycin peptide is an attrative scaffold model (Nhex-tripeptide-EK) with the possibility of 20³ standard residue combinations that will likely facilitate new functional derivatives. However, the selectivity index ratio could be applied to investigate toxicity against mammalian cells. The compounds nostocarboline dimer derivatives 6 and 7, and eudistomin N derivatives 23 and 24 showed better Si values (Si: 2625, 1810 and 4783, 2443, respectively). Regarding the Si values of carmaphycins and their derivatives, these inhibitors exhibit lower Si ratios (derivatives 18 and 19 show the best values Si: 387 and 313, respectively). The better Si values of collected compounds are on the β-carbolines and carmaphycins of the most promising compounds. The use of strategies, as a payload system based on antibodies has been applied in carmaphycins against cancer (Almaliti et al., 2019). Applying combined strategies could increase the efficiency of some compounds against malaria complex diseases, considering that antibodies against plasmodial membrane protein targets can be additional attractive developments (e.g., against erythrocyte invasion PfCRCR complex). The theory proposes a simple system in which the drug is specifically transported and released (ideally), and the antibodies (monoclonal or selected polyclonal antibodies of high affinity) can block a vital function of the parasites. Regarding the nostocarboline and eudistomin N derivatives, it is necessary to confirm/discover the targets in Plasmodium species, and redesigning the compounds to gain selectivity and activity is necessary. Nostocarboline studies of inhibition are limited to hydrolases (butyrylcholinesterase and trypsin). Notably, eudistomin N alone has not yet been tested as an antiplasmodial compound but is known for its inhibitory activity against IkB kinase (IKK). The compound 6-chloro-norharmane (the synthetic precursor of nostocarboline dimers) shared the inhibition activity of IKK with eudistomin N and other β-carbolines. In general, β-carbolines are inhibitors of mammalian kinases such as casein kinase II (CKII), kinase A (PKA), kinase C (PKC), and IKK and other enzymes as endonuclease 1 and monoamine oxydases A and B. Nostocarboline, eudistomin N and 6-chloro-norharmane are halogenated β-carboline compounds and likely shared similar inhibitory activities aforementioned, as inhibition of some kinases (Castro et al., 2003). At present in human beings, we know that IkB kinase proteins are three: IκBα, IκBβ, and IκBε; other as PKC has several isoforms, and CKII has at least two homologous genes encoding catalytic subunits. Using β-carboline compounds against kinase Plasmodium species represents a relatively unexplored area; however, the investigation of Plasmodium kinases as a potential drug target is an area that started to be explored 20 years ago. Carmaphycin B and nostocarboline are compounds with activity in the nM range. They are considered here as important examples of how the synthetic modification of natural compounds can enhance and redirect biological activities.

The remaining compounds collected in this review, such as various cyclic peptides, cyclamides, various non-cyclic peptides, macrolides, and alkaloids, contain a rich chemical diversity with a wide variety of moieties (e.g., azoles, fatty acids, modified amino acids, hydroquinones, halogenated molecules). These compounds will likely be enhanced if chemical libraries are built, as observed with nostocarboline. We have selected a few compounds to propose some valuable observations that could help to adjust strategies to improve the activity of the compounds described above and obtain a better antiplasmodial activity. Since macrolides (bastimolides, palstimolide, lyngbyalosides, BLS, BLLB, and similar compounds as malyngolide dimer) are molecules with considerable molecular sizes and weights, it is desirable to redesign the scaffold and build new, lighter hybrid versions with appropriate druglikness rules. Lyngbyalosides (18E-lyngbyaloside C and lyngbyaloside) and BLS are two types of macrolides evaluated against Plasmodium with similar activity measures (EC₅₀ and IC₅₀). Notably, both groups share the presence of the pyranose ring and similar hydrophobic side chains, while the differences are in the size and type of the macrolide ring (lyngbyalosides are smaller with an oxane ring) and the presence of bromine atom on the side chain in lyngbyalosides. Lyngbyalosides are structures reminiscent of the erythromycin and azithromycin antibiotics with one pyranose ring and brominated on the hydrophobic side chain, and these compounds could be considered scaffolds. BLS and BLLB (a version without a pyranose ring) are promising large structures (622.37 g/mol) that require more research to redesign the best species-specific SERCA protein pump inhibitors with structural bioinformatics approaches.

We collect 15 compounds with at least one azole moiety (Tzl, Tzn, Ozl, Ozn, MOzl, and MOzn) resulting from an aromatic heterocyclization reaction. Cyclamides (10 compounds) are interesting compounds that stand out due to their high azole moiety content. Among these compounds, an interesting example is the macrocyclo 4 (a synthetic derivative of aerucyclamide B) enhanced from 700 to 180 nM with few modifications, producing noteworthy geometrical changes responsible for its activity. Macrocycle 4 has the potential to be a lead compound for generating chemical libraries based on side-chain substitutions of the azole units. An interesting proposal is that other azole groups (e.g., isotiazole, imidazole, or pyrazole) could improve antiplasmodial activity by substituting the MOzl and Tzl units. Natural compounds are molecules whose probability of being used as antiplasmodial compounds is very low and far from being able to be redesigned to improve their anti-parasitic activity; however, it is most likely that redesigns will focus on screening for anti-cancer compounds. The strategy for alternative uses for these molecules as antiplasmodial compounds is mediated by considering the homologous targets detected on cancer cells and searching for them on Plasmodium species—the next step of strategy followed by chemical adaptation through the design of specific chemical libraries. From the azole-containing peptides collected, it is noteworthy that Tzl units are present in all compounds (including dolastatin 10 and Lyngbyabellin A) and MOzn/MOzl units in the 66%. Ozl and Tzl molecules are important scaffolds used as intermediaries in the synthesis of new compounds mediated by the substitution patterns that serve in the reaction outcome and focus mainly on the delimitation of biological activities (including anti-parasitic activities) and valuable leads for pharmaceuticals (Kakkar and Narasimhan, 2019; Singh et al., 2020). The similar compounds pyrazole and imidazole are also used and usually provide better potency in biological activities. Although the production of ergothioneine and ovothiol (imidazole moieties) derived from His on cyanobacterial is known, we did not find imidazole compounds tested against Plasmodium species (Liao and Seebeck, 2017). The azole peptides can be classified into three types: i) cyclic peptides with one or two azole units such as lyngbyabellin A, ulongamide, and kakeromamides; ii) the linear peptide, such as dolastatin 10 with one Tzl unit; iii) cyclic peptides with three azole units, such as aerucyclamides, venturamides, balgacyclamides, and the synthetic macrocycle 4. The azole pyrrole was not considered because it is found regularly in proline resides, although interesting moieties are uncovered. The azole moieties formed by the heterocyclization of one residue merge with the next residue to form moieties with a side chain residue-like azole unit (e.g., Ile-Tzl-Ca). The heterocyclization produces the loss of the peptide bond by liberation of the oxygen, connecting the side chain (sulfur or oxygen) with the carbonyl atom. The azole units (e.g., aa-Tzl-Ca) observed in cyanobacteria could be chemical building blocks for redesigning new molecules with the additional characteristic of changing the side chain (see Fig. S3 for the moieties of azole units). Finally, the activities of compounds, such as lagunamides, are driven by the Hmpa-Dtea1 and Hmpa-Dtuea moieties, and others, such as Dmo from Ikoamide could be chemical building blocks. Thus, thoroughly analyzing the different chemical building blocks could result in designing new compounds with better activity.

The cyanobacterial compounds grouped here focus on antiplasmodial activity, mainly directed toward etiological agents of severe malaria P. falciparum-resistant strains. However, these compounds showed antiplasmodial effects, and several compounds showed anti-cancer activity with a toxic effect on mammalian cells, revealing that these compounds may be general toxic agents against eukaryotes. Although we do not presently understand the reason(s) why cyanobacteria create these compounds (e.g., defense, symbiotic interactions with host and environmental microbiomes, environmental responses, or others), understanding this biological issue may help us to determine how we can obtain compounds with specific characteristics or help plan the methodological forms to searching for cyanobacteria and identify the places to look for them as a mean to exploit the genetic diversity contained in this ancestral clade. Additional strategies, such as cultivation with different media and variations in the environmental conditions, could be helpful to exploit the cyanobacterial “abilities” to increase the production of compounds with lower concentrations and specific biological activities. On the other hand, monitoring activity using cancer cell lines is an essential test because it can help detect compounds with antiplasmodial activity mediated by comparing the homologous targets shared between eukaryotic organisms. The differences in target activity can be re-oriented by developing a chemical library (as is observed in the carmaphycin family of compounds) to produce species-specific compounds. Another approach that can be used to search for and locate new cyanobacterial-like compounds is the strategy of developing chemical libraries and total synthesis adaptations using naturally observed moieties (e.g., the lists of moieties in Figs. S1, S2, and S3) and connections with azole-type units.

6. Conclusions

This review showed that P. falciparum is one of the best-evaluated protozoan parasites for marine-derived compounds with anti-parasitic activity focused on cyanobacterial sources. Marine cyanobacteria are a significant resource for finding and developing new drug leads because they can build peptides with modifications that improve chemical properties. Although sponges are recognized as the most important source of marine-derived products with antiplasmodial activity (additional data not shown), uncultured associated cyanobacteria are important fractions of the sponge microbiome in several species (Konstantinou et al., 2021; Mutalipassi et al., 2021). In particular, cyanobacteria represent a notable source of new and underexplored compounds. Only in recent years has an intense search for new compounds of cyanobacterial origin begun. However, only a few marine compounds have been evaluated for in vivo activity against P. falciparum. Therefore, the chemical synthesis of the most promising compounds identified is a priority for studying their effects on stages of Plasmodium parasites using in vivo animal models. Improving our understanding of the biological context in which the cyanobacterial compounds are produced is an exciting research branch that requires attention for planning the rational exploitation of marine biological resources. The fight against parasites is an evolutionary process because drug resistance will occur. Therefore, we must remain one step ahead regarding drug development and uncover compounds with potential activity and selectivity.

Funding

This work was supported by the Autonomous University of Mexico City (UACM) awarded to MEAS and the grant of the Colegio de Ciencia y Tecnología for the project UACM CCYT-2023-IMP 03, and to the National Institute of Psychiatry Ramón de la Fuente Muñiz (INPRFM). We appreciate the excellent technical assistance of MsC Laura Vazquez Carrillo and the technical support of Alfredo Padilla of the UACM.

Declaration of competing interest

The authors declare no conflict of interest.

Abbreviations

Moiety compound abbreviations and IUPAC names for any compounds (this is a list not a sequence order, the sequences order are on the tables, non-modified residues (L/D) were omitted).

Depsipeptides → Dudawalamides

N-M-Ile

(N-methyl-Isoleucine)

N-M-Phe

(N-methyl-Phenylalanine)

Dhoya

(2,2-dimethy-3-hydroxy-7-octynoic acid)

D-allo-Hmpa

(isoleucic acid or 2-hydroxy-3-methyl pentanoic acid)

Hmba

(2-hydroxyl isovaleric acid or 2-hydroxy-3-methyl butanoic acid)

NO-diM-Tyr

(N,O-di-methyl-Tyrosine)

N-M-Val

(N-methyl-Valine)

Lac

(lactate or 2-hydroxy propanoic acid)

Viridamides

N-M-Ile

(N-methyl-Isoleucine)

N-M-Val

(N-methyl-Valine)

Mdyna

(methoxylated fatty acid moiety, 5-methoxydec-9-ynoic acid)

Hmpa

(isoleucic acid or 2-hydroxy-3-methyl pentanoic acid)

Hmba

(2-hydroxyl isovaleric acid or 2-hydroxy-3-methyl butanoic acid)

O-M-Pro

(O-methyl-Proline)

Symplocamide A

L-Cit

(citrulline residue)

NO-diM-Br-Tyr

(N, O-dimethyl-3-bromotyrosine moiety)

Ahp

(3-amino-6-hydroxy-2-piperidone)

Aba

(alpha-aminobutyric acid)

L-Gln-But

(Glutamine-butanoyl)

Companeramides

Amoya

(β-amino acid 3-amino-2-methyl-7-octynoic acid)

Hmba

(2-hydroxyl isovaleric acid or 2-hydroxy-3-methyl butanoic acid)

N-M-Ala

(N-methyl-Alanine)

N-M-Val

(N-methyl-Valine)

N-M-Leu

(N-methyl-Leu)

Lagunamides

Hmpa

(isoleucic acid or 2-hydroxy-3-methyl pentanoic acid)

N-M-Phe

(N-methyl phenylalanine)

N-M-Gly

(N-methyl glycine)

N-M-Ala

(N-methylalanine)

Dtea1

((2E, 5S, 6S, 7S, 8S) −5,7-dihydroxy-2,6,8-trimethyl dec-2-enoic acid)

Dtea2

((2E, 5S, 6S, 7R, 8E) −5,7-dihydroxy-2,6,8-trimethyldeca-2,8-dienoic acid)

Dtuea

((2E, 5R, 6S, 8S, 9R)-5,8-dihydroxy-2,6,9-trimethylene-2-enoic acid)

Lyngbyabellin A

Ile-Tzl-Ca

(Ile-like unit: 2-(1-amino-2-methyl butyl) −1,3-thiazole-4-carboxylic acid)

dClha

(dichlorinated beta-hydroxy acid moiety or 7,7-dichloro-3-acyloxy-2,2-dimethyl octanoate)

hVal-Tzl-Ca

(hydroxyl-Val-like unit: 2-(1,2-dihydroxy-2-methyl propyl) −1,3-thiazole-4-carboxylic acid)

Ulongamide A

N-M-Phe

(N-methyl phenylalanine)

N-M-Val

(N-methyl valine)

Ala-Tzl-Ca

(Ala-like unit: 2-(1-aminoethyl)-1,3-thiazole-4-carboxylic acid)

Amha

(β-amino acid moiety 3-amino-2-methylhexanoic acid)

Lac

(lactate or 2-hydroxy propanoic acid)

Kakeromamides

NO-M-Tyr

(O, N-dimethyl tyrosine)

Val-Tzl-Ca

(Val-like unit: 2-(1-amino-2-methyl propyl)-1,3-thiazole-4-carboxylic acid)

Ala-Tzl-Ca

(Ala-like unit: 2-(1-aminoethyl)-1,3-thiazole-4-carboxylic acid)

Amha

(β-amino acid moiety 3-amino-2-methylhexanoic acid)

Amoa

(3-amino-2-methyloctanoic acid)

Cyclamides → Venturamides

Ala-Tzl-Ca

(Ala-like unit: 2-(1-aminoethyl)-1,3-thiazole-4-carboxylic acid)

Ala-MOzl-Ca

(Ala-like unit: 2-(1-aminoethyl)-5-methyl-1,3-oxazole-4-carboxylic acid)

Val-Tzl-Ca

(Val-like unit: 2-(1-amino-2-methyl propyl)-1,3-thiazole-4-carboxylic acid)

Thr-Tzl-Ca

(Thr-like unit: 2-(1-amino-2-hydroxypropyl)- 1,3-thiazole-4-carboxylic acid)

Aerucyclamides A & B

Gly-Tzl-Ca

(2-(aminomethyl)-1,3-thiazole-4-carboxylic acid)

Ile-MOzn-Ca

(2-(1-amino-2-methyl butyl)-5-methyl-4,5-dihydro-1,3-oxazole-4-carboxylic acid)

Ile-Tzn-Ca

(2-(1-amino-2-methyl butyl)-4,5-dihydro-1,3-thiazole-4-carboxylic acid)

Ile-Tzl-Ca

(2-(1-amino-2-methyl butyl) −1,3-thiazole-4-carboxylic acid)

Aerucyclamides C & D

Val-Ozl-Ca

(2-(1-amino-2-methylpropyl)-1,3-oxazole-4-carboxylic acid)

Ala-MOzn-Ca

(2-(1-aminoethyl)-5-methyl-4,5-dihydro-1,3-oxazole-4-carboxylic acid)

Ile-Tzl-Ca

(2-(1-amino-2-methyl butyl) −1,3-thiazole-4-carboxylic acid)

Gly-Tzl-Ca

(2-(aminomethyl)-1,3-thiazole-4-carboxylic acid)

Phe-MOzn-Ca

(2-(1-amino-2-phenylethyl)-5-methyl-4,5-dihydro-1,3-oxazole-4-carboxylic acid)

Met-Tzn-Ca

(2-[1-amino-3-(methylsulfanyl)propyl]-4,5-dihydro-1,3-thiazole-4-carboxylic acid)

Balgacyclamides

Ile-Tzl-Ca

(2-(1-amino-2-methyl butyl) −1,3-thiazole-4-carboxylic acid)

Val-MOzn-Ca

(2-(1-amino-2-methylpropyl)-5-methyl-4,5-dihydro-1,3-oxazole-4-carboxylic acid)

Ala-MOzn-Ca

(2-(1-aminoethyl)-5-methyl-4,5-dihydro-1,3-oxazole-4-carboxylic acid)

Phe-MOzn-Ca

(2-(1-amino-2-phenylethyl)-5-methyl-4,5-dihydro-1,3-oxazole-4-carboxylic acid)

Macrocycle 4

Gly-Tzl-Ca

(2-(aminomethyl)-1,3-thiazole-4-carboxylic acid)

Ile-MOzl-Ca

(2-(1-amino-2-methylbutyl)-5-methyl-1,3-oxazole-4-carboxylic acid)

Ile-Tzl-Ca

(2-(1-amino-2-methyl butyl) −1,3-thiazole-4-carboxylic acid)

Other modified peptides. → Gallinamide A

N-diM-Ile

(N, N-diMethyl-isoleucine)

Ica

(isocaproic acid)

Apa

(4-(S)-amino-2-(E)-pentenoic acid)

Mmp

(methyl-methoxy pyrrolinone)

Dolastatin 10

Dolavaline

(N-diMethyl-valine)

Dolaisoleuine

((3R,4S,5S)-3- methoxy-5-methyl-4-(methylamino) heptanoic acid)

O-M-Dolaproine

((2R,3R)-3-methoxy -2-methyl-3-[(2S)-pyrrolidin-2-yl] propanoic acid)

Dolaphenine

((1S)-2-phenyl-1-(1,3-thiazol-2-yl)ethanamine)

Dolastatin 15

Dolavaline

(N-diMethyl-valine)

Hmba

(2-hydroxyl isovaleric acid or 2-hydroxy-3-methyl butanoic acid)

Benp

(5-benzyl-4-methoxy-1,5-dihydro-2H-pyrrole-2-one)

Carmaphycin A & B

NHex

(N-hexanoyl)

Met-Ox

(L-methionine-sulfoxide)

Met-On

(L-methionine-sulfone)

EK

(β-epoxyketone)

Carfilzomib

Mor

(morpholino)

EK

(β-epoxyketone)

Carmaphycin B derivative 1, 2, 6, 16, 18, 19

NHex

(N-hexanoyl)

L-Nle

(2-amino hexanoic acid)

EK

(β-epoxyketone)

Carmaphycin B derivative 3, 8

NHex

(N-hexanoyl)

4-Pyr-LAla

(3-(4-pyridyl)-L-alanine)

L-Nle

(2-amino hexanoic acid)

EK

(α, β-epoxyketone)

Dragomabin and Dragonamides

Moya

(2-methyloct-7-ynoic acid)

N-M-Phe

(N-methyl-L-phenilalanine)

N-M-Ala

(N-methyl-L-alanine)

NO-diM-Tyr

(N,O-dimethyl-L-tyrosine)

N-M-Val

(N-methyl-L-valine)

Carmabin A

Mdya

(2,4-dimethyldec-9-ynoic acid)

N-M-Phe

(N-methyl-L-phenilalanine)

N-M-Ala

(N-methyl-L-alanine)

NO-diM-Tyr

(N,O-dimethyl-L-tyrosine)

Malyngamide X

Mtea

(7-methoxytetradec-4-enoic acid)

N-M-Ala

(N-methyl-L-alanine)

Mpna

(4-amino-3- hydroxy-2-methylpentanoic acid)

Lpmp

(isopropyl-4-methoxy-D3-pyrrolin-2-one)

Mabuniamide

Bta

(butanoic acid)

N-M-Ala

(N-methyl-L-alanine)

N-M-Phe

(N-methyl-L-phenilalanine)

N-M-D-Val

(N-methyl-D-valine)

N-M-Pro

(N-methyl-L-proline)

Ikoamide

Dmo

(3,5-dimethoxy octanoic acid)

N-M-D-allo-Ile

(N-methyl-D-allo-isoleucine)

N-M-Leu

(N-methyl-L-leucine)

N-M-Ile

(N-methyl-L-isoleucine)

N-M-Gln

(N-methyl-L-glutamine)

NO-diM-L-Tyr-O-M

(N,O-dimethyl-L-tyrosine-O-methyl)

Tychonamides

Atpoa

(β-amino acid 3-amino-2,5,7-trihydroxy-8-phenyloctanoic acid)

Dhb

(dehydrobutyrine)

O-M-D-Hty

(O-methyl-D-homotyrosine)

D-Hphe

(D-homophenylalanine)

N-A-N-Met-Leu

(N-acetyl-N-methyl-leucine)

Hoshinoamides

Hba

(4-(4-hydroxyphenyl)-butanoic acid moiety)

Aha

(6-aminohexanoic acid moiety)

Ana

(9-aminononanoic acid)

Amoc

(8-amino-4-methyloctanoic acid)

N-M-D-Phe

(N-methyl-D-phenilalanine)

O-M-Pro

(O-methyl-Proline)

Alkaloids

Nostocarboline

(N-methyl-6-chloro-carbolinium)

Nostocarboline dimer 7

(2,2'-(pentane-1,5-diyl) di (N-methyl-6-chloro-carbolinium)

Nostocarboline dimer 8

(2,2'-(octane-1,8-diyl) di (N-methyl-6-chloro-carbolinium)

Eudistomin N

(6-Brome-beta-carboline)

Eudistomin 8

(8-Brome-beta-carboline)

Eudistomin N derivative 23

(6-Bromo-2-methyl-9H-β-carbolin-2-ium iodide)

Eudistomin N derivative 24

(6-Bromo-2-ethyl-9H-β-carbolin-2-ium iodide)

SAHA

(suberoylanilide hydroxamic acid)

Santacruzamate A

(ethyl N-[4-oxo-4-(2-phenylethylamino) butyl]carbamate)

Tumonoic acid I

Octa

(octanoic acid)

Hmba

(2-hydroxyl isovaleric acid or 2-hydroxy-3-methyl butanoic acid)

Hmpa

(isoleucic acid or 2-hydroxy-3-methyl pentanoic acid)

Calothrixins A

(20-oxido-10-aza-20-azoniapentacyclo [11.8.0.03,11.04,9.014,19] henicosa-1(13),3(11), 4,6,8,14,16,18,20-nonaene-2,12-dione)

Calothrixins B

(10,20-diazapentacyclo [11.8.0.03,11.04,9.01419] henicosa-1(13),3(11), 4,6,8,14,16,18,20-nonaene-2,12-dione)

Others compound

Malyngolide dimer

((3R,6S,9R,12S)-6,12-bis(hydroxymethyl)-3,9-dimethyl-6,12-di(nonyl)-1,7-dioxacyclododecane-2,8-dione)

Hierridin A

(2,4-dimethoxy-6-heptadecylphenol)

Hierridin B

(2,4-dimethoxy-6-pentadactyl-phenol)

Ambigol A

(3,5-dichloro-2-(3,5-dichloro-2-hydroxyphenyl)-6-(2,4-dichlorophenoxy)phenol)

Ambigol B

(3,5-dichloro-2,6-bis(2,4-dichlorophenoxy)phenol)

Ambigol C

(2,6-dichloro-3,5-bis(2,4-dichlorophenoxy)phenol)

Bastimolide A

((10S,12R,20R,24R,28S,36R,40S)-40-tert-butyl-10,12,16,20,22,24,28,32,36-nonahydroxy-4-methyl-1-oxacyclotetracont-3-en-2-one)

Bastimolide B

((3Z,10S,12R,16S,20R,22S,24R)-10,12,16,20,22-pentahydroxy-4-methyl-24-[(4S,8S,12R,16S)-4,8,12,16-tetrahydroxy-17,17-dimethyloctadecyl]-1-oxacyclotetracos-3-en-2-one)

18E-Lyngbyaloside C

((1S,5R,7R,8R,11S,13R)-5-[(3E,5E)-6-bromohexa-3,5-dienyl]-1,7-dihydroxy-5,8-dimethyl-13-[(2R,3R,4R,5S,6S)-3,4,5-trimethoxy-6-methyloxan-2-yl]oxy-4,15-dioxa bicyclo[9.3.1]pentadecan-3-one)

Lyngbyaloside

((1R,8Z,13S,15S)-5-[(1E,3E)-4-bromobuta-1,3-dienyl]-1,11-dihydroxy-2,7,12-trimethyl-15-[(2S,3S,4S,5R,6R)-3,4,5-trimethoxy-6-methyloxan-2-yl]oxy-4,17-dioxabicyclo[11.3.1]heptadec-8-en-3-one)

BLS

(Biselyngbyaside)

BLLB

(biselyngbyolide B)

Appendix A. Supplementary data

The following is the Supplementary data to this article.