The Chemistry, Biochemistry and Pharmacology of Marine Natural Products from Leptolyngbya, a Chemically Endowed Genus of Cyanobacteria

Leptolyngbya, a well-known genus of cyanobacteria, is found in various ecological habitats including marine, fresh water, swamps, and rice fields. Species of this genus are associated with many ecological phenomena such as nitrogen fixation, primary productivity through photosynthesis and algal blooms. As a result, there have been a number of investigations of the ecology, natural product chemistry, and biological characteristics of members of this genus. In general, the secondary metabolites of cyanobacteria are considered to be rich sources for drug discovery and development. In this review, the secondary metabolites reported in marine Leptolyngbya with their associated biological activities or interesting biosynthetic pathways are reviewed, and new insights and perspectives on their metabolic capacities are gained.


Introduction
Cyanobacteria, also known as 'blue-green algae', rank among the oldest prokaryotes found on Earth and evolved to possess remarkable capabilities of oxygenic photosynthesis and nitrogen fixation, allowing them to both adapt to and affect the planet's environmental conditions for over two billion years [1][2][3][4]. In traditional classification schema, filamentous cyanobacteria without akinetes, heterocysts or true-branching were generally assigned to the order Oscillatoriales (Section III) [5]. However, one specific genus of filamentous, nonheterocystous cyanobacteria with granular surface ornamentation was ultimately named Leptolyngbya in 1988, and placed in the order Pseudanabaenales [6][7][8]. The identification of this genus was supported by phylogenetic analysis, including 16S rRNA gene sequences, which at the time showed no clear relationship with any other defined genus of cyanobacteria [9].

Polypeptides
In 2008, McPhail and co-workers reported the discovery of coibamide A (1), a cyclic depsipeptide with a high degree of Nand O-methylation from a Leptolyngbya species collected in Panama ( Figure 1) [32]. The absolute configuration was determined by a series of chiral HPLC analyses of the amino acids resulting from the hydrolyzed peptide as well as some computational modeling. Interest in coibamide A (1) mounted due to its exquisitely potent low-nanomolar in vitro inhibition activity against multiple cancer cell lines, including human NCI-H460, MDA-MB-231, H292, PC-3, SF-295, mouse neuro2a, LOX IMVI, HL-60(TB), and SNB-75 cell lines [32]. The number of cell lines was expanded to include human U87-MG, SF-295 glioblastoma cells and mouse embryonic fibroblasts (MEFs), and it was further revealed that the cyclized structure was crucial for potent biological activity [52]. Several studies subsequently reported the total synthesis of coibamide A (1). Yao and Lim et al. completed the total synthesis of the structure originally reported for coibamide A (2), as well as a synthetic O-desmethyl analogue. However, both the 1 H and 13 C NMR data for the synthetic product 2 differed from those of the natural product, which indicated that the absolute configuration of this compound required revision [33,34]. Additionally, Oishi and Fujii synthesized the d-N-Me-Ala epimer of coibamide A (4) due to an epimerization of this residue during the macrocyclization process ( Figure 1) [35]. In 2015, Fang and Su were able to assign the correct configuration of coibamide A (1) with the revision of the l-HVA and l-N-Me-Ala residues to the d-HVA and d-N-Me-Ala after total synthesis of this alternative along with its diastereomeric analogues ( Figure 1) [30]. The absolute configuration of coibamide A (1) proposed by Fang and Su was further confirmed by McPhail and Cheong using computational methods to calculate NMR data for the conformational space occupied by several possible diastereomers and comparison with experimental values [53]. This considerable effort to resolve the structure of coibamide A (1) was largely motivated by the potent cytotoxicity of this molecule, and as a byproduct, provided a number of new analogues for mechanistic and pharmacological studies. More recently, Su and Fang went on to synthesize 18 new analogues (2, 4, 5-20) including the originally proposed structure of coibamide A (2) as well as the revised correct structure (1) to perform a structure-activity relationship study ( Figure 2) [31]. However, none of the coibamide A analogues were more potent cytotoxins than the natural product, indicating the strong correlation between the observed activity, the core molecular structure and optimization of this structure through natural evolutionary processes. The only analogue that exhibited similar inhibition as natural coibamide A was the [MeAla3-MeAla6]-coibamide (8), which significantly suppressed tumor growth in vivo [31,33,35].

Polypeptides
In 2008, McPhail and co-workers reported the discovery of coibamide A (1), a cyclic depsipeptide with a high degree of N-and O-methylation from a Leptolyngbya species collected in Panama ( Figure  1) [32]. The absolute configuration was determined by a series of chiral HPLC analyses of the amino acids resulting from the hydrolyzed peptide as well as some computational modeling. Interest in coibamide A (1) mounted due to its exquisitely potent low-nanomolar in vitro inhibition activity against multiple cancer cell lines, including human NCI-H460, MDA-MB-231, H292, PC-3, SF-295, mouse neuro2a, LOX IMVI, HL-60(TB), and SNB-75 cell lines [32]. The number of cell lines was expanded to include human U87-MG, SF-295 glioblastoma cells and mouse embryonic fibroblasts (MEFs), and it was further revealed that the cyclized structure was crucial for potent biological activity [52]. Several studies subsequently reported the total synthesis of coibamide A (1). Yao and Lim et al. completed the total synthesis of the structure originally reported for coibamide A (2), as well as a synthetic O-desmethyl analogue. However, both the 1 H and 13 C NMR data for the synthetic product 2 differed from those of the natural product, which indicated that the absolute configuration of this compound required revision [33,34]. Additionally, Oishi and Fujii synthesized the D-N-Me-Ala epimer of coibamide A (4) due to an epimerization of this residue during the macrocyclization process ( Figure 1) [35]. In 2015, Fang and Su were able to assign the correct configuration of coibamide A (1) with the revision of the L-HVA and L-N-Me-Ala residues to the D-HVA and D-N-Me-Ala after total synthesis of this alternative along with its diastereomeric analogues ( Figure 1) [30]. The absolute configuration of coibamide A (1) proposed by Fang and Su was further confirmed by McPhail and Cheong using computational methods to calculate NMR data for the conformational space occupied by several possible diastereomers and comparison with experimental values [53]. This considerable effort to resolve the structure of coibamide A (1) was largely motivated by the potent cytotoxicity of this molecule, and as a byproduct, provided a number of new analogues for mechanistic and pharmacological studies. More recently, Su and Fang went on to synthesize 18 new analogues (2, 4, 5-20) including the originally proposed structure of coibamide A (2) as well as the revised correct structure (1) to perform a structure-activity relationship study ( Figure 2) [31]. However, none of the coibamide A analogues were more potent cytotoxins than the natural product, indicating the strong correlation between the observed activity, the core molecular structure and optimization of this structure through natural evolutionary processes. The only analogue that exhibited similar inhibition as natural coibamide A was the [MeAla3-MeAla6]-coibamide (8), which significantly suppressed tumor growth in vivo [31,33,35].  The dolastatins are an expansive and well-known series of peptidic compounds. These were named after first being discovered from the sea hare Dolabella auricularia, but it was later found that the mollusk accumulates these natural products from cyanobacteria in their diets [37,39,54,55]. Dolastatin 12 (21) is one such compound, and has interesting 4-amino-2,2-dimethyl-3-oxopentanoic acid (Ibu) and (2S,3R)-3-amino-2-methylpentanoic acid (MAP) residues in its cyclic structure ( Figure   Figure 2. Synthetic analogues of coibamide A produced to explore structure-activity relationships (SAR) in this molecular class.
The dolastatins are an expansive and well-known series of peptidic compounds. These were named after first being discovered from the sea hare Dolabella auricularia, but it was later found that the mollusk accumulates these natural products from cyanobacteria in their diets [37,39,54,55]. Dolastatin 12 (21) is one such compound, and has interesting 4-amino-2,2-dimethyl-3-oxopentanoic acid (Ibu) and (2S,3R)-3-amino-2-methylpentanoic acid (MAP) residues in its cyclic structure ( Figure 3). The macrocycle also includes an ester linkage across a 2-hydroxy-3-methylpentanoate (Hmp) residue, and is thus a depsipeptide. While this compound was originally isolated from D. auricularia, it was noted to resemble the Lyngbya majuscula metabolite majusculamide C. Dolastatin 12 (21) was later re-discovered from a mixed cyanobacterial assemblage of L. majuscula and Schizothrix calcicola from Guam [36,39,56,57] as well as from a Leptolyngbya sp. RS03 collected in the Red Sea [36]. Dolastatin 12 (21) may be present as a mixture of diastereomers arising from epimerization of the acid-sensitive Ibu unit during the isolation process [57]. The configuration of this Ibu residue was determined by CD and 1 H NMR analysis after hydrolysis and purification [39]. The correct configuration was determined to be R by comparing the free Ibu unit to Adhpa synthetic standards [57]. In the early studies, dolastatin 12 (21) was shown to inhibit actin polymerization [37,58]. Dolastatin 12 (21) has in vitro MICs <0.05 µg/mL against human nasopharyngeal carcinoma cells and 0.08 µg/mL against human colon adenocarcinoma cells [37]. Further cytotoxicity characterization showed that dolastatin 12 has an in vitro IC 50 > 1 µM against HeLa cells [36,56] and a ED 50 of 7.5 × 10 −2 µg/mL against murine P388 lymphocytic leukemia [38,39].
Molassamide (28) is another cyclic depsipeptide ( Figure 7) that was isolated along with loggerpeptins A-C (25-27) [40]. Compared to loggerpeptins (25)(26)(27), molassamide (28) exhibited much more potent inhibition activity against porcine pancreatic elastase, with an IC 50 value of 50 nM, indicating the Abu residue between Ahp and Thr-1 is important to the antiproteolytic selectivity [40]. Molassamide presents similar binding patterns in molecular docking as loggerpeptin C (27), with the Abu and N-terminal Thr-1, Thr-2 and Ala binding in subsites S1-S4 of HNE and porcine pancreatic elastase [40]. Molassamide also inhibited elastase from cleaving the substrate CD40 in both biochemical and cellular assays, and it also inhibited ICAM-1 cleavage and downregulated elastase-induced ICAM-1 gene expression. Overall, this profile of activity is indicative of molassamide being a promising candidate for potential treatment of breast cancer [40]. Molassamide (28) is another cyclic depsipeptide ( Figure 7) that was isolated along with loggerpeptins A-C (25-27) [40]. Compared to loggerpeptins (25)(26)(27), molassamide (28) exhibited much more potent inhibition activity against porcine pancreatic elastase, with an IC50 value of 50 nM, indicating the Abu residue between Ahp and Thr-1 is important to the antiproteolytic selectivity [40]. Molassamide presents similar binding patterns in molecular docking as loggerpeptin C (27), with the Abu and N-terminal Thr-1, Thr-2 and Ala binding in subsites S1-S4 of HNE and porcine pancreatic elastase [40]. Molassamide also inhibited elastase from cleaving the substrate CD40 in both biochemical and cellular assays, and it also inhibited ICAM-1 cleavage and downregulated elastaseinduced ICAM-1 gene expression. Overall, this profile of activity is indicative of molassamide being a promising candidate for potential treatment of breast cancer [40].   Molassamide (28) is another cyclic depsipeptide ( Figure 7) that was isolated along with loggerpeptins A-C (25-27) [40]. Compared to loggerpeptins (25)(26)(27), molassamide (28) exhibited much more potent inhibition activity against porcine pancreatic elastase, with an IC50 value of 50 nM, indicating the Abu residue between Ahp and Thr-1 is important to the antiproteolytic selectivity [40]. Molassamide presents similar binding patterns in molecular docking as loggerpeptin C (27), with the Abu and N-terminal Thr-1, Thr-2 and Ala binding in subsites S1-S4 of HNE and porcine pancreatic elastase [40]. Molassamide also inhibited elastase from cleaving the substrate CD40 in both biochemical and cellular assays, and it also inhibited ICAM-1 cleavage and downregulated elastaseinduced ICAM-1 gene expression. Overall, this profile of activity is indicative of molassamide being a promising candidate for potential treatment of breast cancer [40].

Simple Esters
Lumyong et al. isolated an antibacterial compound, 2-hydroxyethyl-11-hydroxyhexadec-9-enoate (29) (Figure 8), from Leptolyngbya sp. LT19 [41], as a result of screening for antibacterial activities. They showed compound 25 to be active against Vibrio harveyi and V. parahaemolyticus, with MIC values of 250-1000 and 350-1000 µg/mL. This antibacterial activity could potentially be useful to the shrimp aquaculture industry that is often burdened by the highly damaging Vibrio spp. pathogens [41]. However, the absolute configuration of the single stereocenter in compound 25 has not yet been determined. enoate (29) (Figure 8), from Leptolyngbya sp. LT19 [41], as a result of screening for antibacterial activities. They showed compound 25 to be active against Vibrio harveyi and V. parahaemolyticus, with MIC values of 250-1000 and 350-1000 μg/mL. This antibacterial activity could potentially be useful to the shrimp aquaculture industry that is often burdened by the highly damaging Vibrio spp. pathogens [41]. However, the absolute configuration of the single stereocenter in compound 25 has not yet been determined.  (Figure 9) [42]. Chemical synthesis of honaucin and a number of analogs (30)(31)(32) revealed that the each of the functional groups is critical for both of these biological activities. Further, synthetic honaucin analogues 4-bromo-honaucin A (33) and 4′-iodohonaucin A (35) were discovered to have slightly more potent activity in cellular TRAP activity than honaucin A itself (30), with IC50 values of 0.54 and 0.61 μg/mL compared to 0.63 μg/mL for 30 (Figure 9), [43]. Mechanistic pharmacological investigations of honaucin A (30) indicated that the molecular target(s) involves the Nrf2-ARE (Antioxidant Response Element) pathway, and specifically involves interaction with Cys residues on Keap1 when it is complexed with Nrf2. This allows Nrf2 to be transported to the nucleus, where it activates cytoprotective genes and generates an anti-inflammatory response [64]. Additionally, further investigation of bromo-honaucin A (33) revealed that it provides a protective effect against bone loss in RANKL-treated murine monocyte/macrophage RAW264.7 cells [43].

Macrolides
Leptolyngbyolides A-D (36-39), a series of 22-membered macrolides, were isolated from Leptolyngbya sp. collected in Okinawa, Japan. The absolute configuration of each was assigned following an asymmetric total synthesis of leptolyngbyolide C (38) (Figure 10) [44]. Leptolyngbyolides A-D (36-39) were screened for cytotoxic activity against HeLa S3 cells in vitro and were found to be moderately active, with IC50 values of 0.99, 0.16, 0.64 and 0.15 μM, respectively [44]. Furthermore, these metabolites possessed actin-depolymerizing activity with an EC50 of 12.6, 11.6,  (Figure 9) [42]. Chemical synthesis of honaucin and a number of analogs (30)(31)(32) revealed that the each of the functional groups is critical for both of these biological activities. Further, synthetic honaucin analogues 4-bromo-honaucin A (33) and 4 -iodohonaucin A (35) were discovered to have slightly more potent activity in cellular TRAP activity than honaucin A itself (30), with IC 50 values of 0.54 and 0.61 µg/mL compared to 0.63 µg/mL for 30 (Figure 9), [43]. Mechanistic pharmacological investigations of honaucin A (30) indicated that the molecular target(s) involves the Nrf2-ARE (Antioxidant Response Element) pathway, and specifically involves interaction with Cys residues on Keap1 when it is complexed with Nrf2. This allows Nrf2 to be transported to the nucleus, where it activates cytoprotective genes and generates an anti-inflammatory response [64]. Additionally, further investigation of bromo-honaucin A (33) revealed that it provides a protective effect against bone loss in RANKL-treated murine monocyte/macrophage RAW264.7 cells [43].
Lumyong et al. isolated an antibacterial compound, 2-hydroxyethyl-11-hydroxyhexadec-9enoate (29) (Figure 8), from Leptolyngbya sp. LT19 [41], as a result of screening for antibacterial activities. They showed compound 25 to be active against Vibrio harveyi and V. parahaemolyticus, with MIC values of 250-1000 and 350-1000 μg/mL. This antibacterial activity could potentially be useful to the shrimp aquaculture industry that is often burdened by the highly damaging Vibrio spp. pathogens [41]. However, the absolute configuration of the single stereocenter in compound 25 has not yet been determined.  (Figure 9) [42]. Chemical synthesis of honaucin and a number of analogs (30)(31)(32) revealed that the each of the functional groups is critical for both of these biological activities. Further, synthetic honaucin analogues 4-bromo-honaucin A (33) and 4′-iodohonaucin A (35) were discovered to have slightly more potent activity in cellular TRAP activity than honaucin A itself (30), with IC50 values of 0.54 and 0.61 μg/mL compared to 0.63 μg/mL for 30 (Figure 9), [43]. Mechanistic pharmacological investigations of honaucin A (30) indicated that the molecular target(s) involves the Nrf2-ARE (Antioxidant Response Element) pathway, and specifically involves interaction with Cys residues on Keap1 when it is complexed with Nrf2. This allows Nrf2 to be transported to the nucleus, where it activates cytoprotective genes and generates an anti-inflammatory response [64]. Additionally, further investigation of bromo-honaucin A (33) revealed that it provides a protective effect against bone loss in RANKL-treated murine monocyte/macrophage RAW264.7 cells [43].

Macrolides
Leptolyngbyolides A-D (36-39), a series of 22-membered macrolides, were isolated from Leptolyngbya sp. collected in Okinawa, Japan. The absolute configuration of each was assigned following an asymmetric total synthesis of leptolyngbyolide C (38) (Figure 10) [44]. Leptolyngbyolides A-D (36-39) were screened for cytotoxic activity against HeLa S3 cells in vitro and were found to be moderately active, with IC 50 values of 0.99, 0.16, 0.64 and 0.15 µM, respectively [44]. Furthermore, these metabolites possessed actin-depolymerizing activity with an EC 50 of 12.6, 11.6, 26.9 and 21.5 µM, and this may represent the mechanism for the observed cellular apoptosis caused by these compounds [44].
Mar. Drugs 2020, 18, x FOR PEER REVIEW 15 of 27 26.9 and 21.5 μM, and this may represent the mechanism for the observed cellular apoptosis caused by these compounds [44]. A new macrolide, palmyrolide A (40), was isolated from an environmental assemblage of a Leptolyngbya cf. and Oscillatoria spp. collected from Palmyra Atoll in the Central Pacific Ocean ( Figure  11). Palmyrolide A (40) comprises a rare N-methyl enamide functionality, as well as an intriguing tbutyl branch that likely results biosynthetically from the incorporation of malonate and three methyl groups from S-adenosyl-L-methionine (SAM), as shown for apratoxin [45,65]. The absolute configuration of this molecule was assigned by total synthesis of both the natural (−)-palmyrolide A (40) and its enantiomer, (+)-ent-palmyrolide A. This was necessitated because the t-butyl substituent apparently provides the lactone ester bond with resistance to hydrolysis, precluding chemical degradation studies [66][67][68]. It was speculated that cyanobacterial secondary metabolites possessing this motif, a t-butyl adjacent to an ester, might be stable under a wide variety of environmental conditions, similar to what was observed in the laboratory environment. Palmyrolide A (40) exhibited potent inhibition of calcium oscillations in murine cerebrocortical neurons and sodium channelblocking activity in neuroblastoma (neuro2a) cells [45]. Phormidolide (41) is a 16-membered macrocyclic lactone polyketide-derived metabolite that was discovered from an Indonesian Leptolyngbya sp. (ISB3NOV94-8A) ( Figure 12). This molecule has several unique structural features including a large number of hydroxy and methyl groups on the carbon backbone, several points of both cis and trans unsaturation, and a vinyl bromide at one terminus. Phormidolide (41) showed in vitro brine shrimp toxicity, with an LC50 of 1.5 μM. Multiple NMR experiments, including GHMBC, 2D INADEQUATE, and ACCORD-ADEQUATE, G-BIRDR, X-HSQMBC NMR experiments, enabled a J-based configuration analysis and deduction of relative configuration of stereocenters; a variable temperature Mosher ester analysis was used to assign the absolute configuration [46]. An investigation of the biosynthetic pathway for phormidolide (41) used a genome sequencing approach, and identified the phormidolide biosynthetic gene cluster (phm). The phormidolide (41) gene cluster was found to be of the trans-AT PKS type, which has been relatively A new macrolide, palmyrolide A (40), was isolated from an environmental assemblage of a Leptolyngbya cf. and Oscillatoria spp. collected from Palmyra Atoll in the Central Pacific Ocean ( Figure 11). Palmyrolide A (40) comprises a rare N-methyl enamide functionality, as well as an intriguing t-butyl branch that likely results biosynthetically from the incorporation of malonate and three methyl groups from S-adenosyl-l-methionine (SAM), as shown for apratoxin [45,65]. The absolute configuration of this molecule was assigned by total synthesis of both the natural (−)-palmyrolide A (40) and its enantiomer, (+)-ent-palmyrolide A. This was necessitated because the t-butyl substituent apparently provides the lactone ester bond with resistance to hydrolysis, precluding chemical degradation studies [66][67][68]. It was speculated that cyanobacterial secondary metabolites possessing this motif, a t-butyl adjacent to an ester, might be stable under a wide variety of environmental conditions, similar to what was observed in the laboratory environment. Palmyrolide A (40) exhibited potent inhibition of calcium oscillations in murine cerebrocortical neurons and sodium channel-blocking activity in neuroblastoma (neuro2a) cells [45].
Mar. Drugs 2020, 18, x FOR PEER REVIEW 15 of 27 26.9 and 21.5 μM, and this may represent the mechanism for the observed cellular apoptosis caused by these compounds [44]. A new macrolide, palmyrolide A (40), was isolated from an environmental assemblage of a Leptolyngbya cf. and Oscillatoria spp. collected from Palmyra Atoll in the Central Pacific Ocean ( Figure  11). Palmyrolide A (40) comprises a rare N-methyl enamide functionality, as well as an intriguing tbutyl branch that likely results biosynthetically from the incorporation of malonate and three methyl groups from S-adenosyl-L-methionine (SAM), as shown for apratoxin [45,65]. The absolute configuration of this molecule was assigned by total synthesis of both the natural (−)-palmyrolide A (40) and its enantiomer, (+)-ent-palmyrolide A. This was necessitated because the t-butyl substituent apparently provides the lactone ester bond with resistance to hydrolysis, precluding chemical degradation studies [66][67][68]. It was speculated that cyanobacterial secondary metabolites possessing this motif, a t-butyl adjacent to an ester, might be stable under a wide variety of environmental conditions, similar to what was observed in the laboratory environment. Palmyrolide A (40) exhibited potent inhibition of calcium oscillations in murine cerebrocortical neurons and sodium channelblocking activity in neuroblastoma (neuro2a) cells [45]. Phormidolide (41) is a 16-membered macrocyclic lactone polyketide-derived metabolite that was discovered from an Indonesian Leptolyngbya sp. (ISB3NOV94-8A) ( Figure 12). This molecule has several unique structural features including a large number of hydroxy and methyl groups on the carbon backbone, several points of both cis and trans unsaturation, and a vinyl bromide at one terminus. Phormidolide (41) showed in vitro brine shrimp toxicity, with an LC50 of 1.5 μM. Multiple NMR experiments, including GHMBC, 2D INADEQUATE, and ACCORD-ADEQUATE, G-BIRDR, X-HSQMBC NMR experiments, enabled a J-based configuration analysis and deduction of relative configuration of stereocenters; a variable temperature Mosher ester analysis was used to assign the absolute configuration [46]. An investigation of the biosynthetic pathway for phormidolide (41) used a genome sequencing approach, and identified the phormidolide biosynthetic gene cluster (phm). The phormidolide (41) gene cluster was found to be of the trans-AT PKS type, which has been relatively Phormidolide (41) is a 16-membered macrocyclic lactone polyketide-derived metabolite that was discovered from an Indonesian Leptolyngbya sp. (ISB3NOV94-8A) ( Figure 12). This molecule has several unique structural features including a large number of hydroxy and methyl groups on the carbon backbone, several points of both cis and trans unsaturation, and a vinyl bromide at one terminus. Phormidolide (41) showed in vitro brine shrimp toxicity, with an LC 50 of 1.5 µM. Multiple NMR experiments, including GHMBC, 2D INADEQUATE, and ACCORD-ADEQUATE, G-BIRDR, X-HSQMBC NMR experiments, enabled a J-based configuration analysis and deduction of relative configuration of stereocenters; a variable temperature Mosher ester analysis was used to assign the absolute configuration [46]. An investigation of the biosynthetic pathway for phormidolide (41) used a genome sequencing approach, and identified the phormidolide biosynthetic gene cluster (phm). The phormidolide (41) gene cluster was found to be of the trans-AT PKS type, which has been relatively rarely reported in cyanobacteria. This was based on finding two discrete trans-AT open reading frames along with KS-AT adaptor regions (ATd) within the PKS megasynthase. The megasynthase possesses ketosynthases, ketoreductases, KS-AT adaptor regions, dehydratases, methyltransferases, O-methyltransferase, enoyl-CoA hydratases, an FkbH-like domain, a pyran synthase, an NRPS-like condensation domain and an acyl carrier protein that were consistent with structure of phormidolide (41). The biosynthetic pathway also provided further supporting evidence for the absolute configuration of phormidolide (41), due to the stereospecificity of the ketoreductases observed in phm compared with known homologues [26]. However, subsequent chemical synthesis of key fragments of phormidolide revealed the need for a revision in configuration in at least one stereocenter [69]. Simultaneously, a reanalysis of the biosynthetic gene cluster suggested additional revisions in configuration were possibly required [70]. These findings stimulated a concerted computational and NMR-based re-investigation of phormidolide's complex 3-dimensional structure, leading to a revision in several stereocenters (42) [71].  (41). The biosynthetic pathway also provided further supporting evidence for the absolute configuration of phormidolide (41), due to the stereospecificity of the ketoreductases observed in phm compared with known homologues [26]. However, subsequent chemical synthesis of key fragments of phormidolide revealed the need for a revision in configuration in at least one stereocenter [69]. Simultaneously, a reanalysis of the biosynthetic gene cluster suggested additional revisions in configuration were possibly required [70]. These findings stimulated a concerted computational and NMR-based re-investigation of phormidolide's complex 3-dimensional structure, leading to a revision in several stereocenters (42) [71].

Pyrones
Kalkipyrone A (43) was first reported in a mixed assemblage of Lyngbya majuscula and Tolypothrix sp. (Figure 13), and was found to have potent brine shrimp toxicity (LD 50 = 1 µg/mL) and ichthyotoxicity against Carassius auratus goldfish (LD 50 = 2 µg/mL) [48]. Kalkipyrone A (43) and its analogue kalkipyrone B (44) were later found as metabolites of a Leptolyngbya sp. (ASG15JUL14-6) collected from America Samoa (Figure 12) [47]. Kalkipyrone A (43) and B (44) were moderately toxic to a Saccharomyces cerevisiae strain lacking 16 ATP-binding cassette transporter pump genes (ABC16-monster strain; IC 50 = 14.6 and 13.4 µM, respectively), while the in vitro cytotoxicity of these two natural products against H460 human lung cancer cells was somewhat more potent (EC 50 = 0.9 and 9.0 µM, respectively) [47]. Three related γ-pyrone-containing polyketides described as yoshinones A, B1, and B2 (45)(46)(47), were isolated from Leptolyngbya sp. collected from Ishigaki island Okinawa, Japan ( Figure 14). The absolute configuration of each compound was attempted by a modified Mosher's method; however, these failed possibly as a result of the low amounts of samples available [49]. Thus, to assign the absolute configuration and provide sufficient amounts for further research, the absolute stereochemistry was achieved through total synthesis and comparison of NMR and chiroptical properties between the natural product and synthetic standards [72]. Yoshinone A (45) was found to inhibit adipogenic differentiation against 3T3-L1 cells, with an EC50 of 420 nM and with little cytotoxicity (IC50 = 63.8 μM to S. cerevisiae ABC16-monster cell) [47,49]. The adipogenic differentiation against 3T3-L1 cells of yoshinone B1 (46) and B2 (47) was considerably less potent, with less than 50% activity observed at tested concentrations up to 5 μM [48]. Further examination of structure-activity relationships in this drug class indicated that the position of the pyrone ring and side chain olefin are important for the inhibition of adipogenic differentiation. Further in vitro and in vivo experiments showed that yoshinone A (45) stimulates lactate accumulation deriving from the glycolytic system, and increases fat utilization to compensate for an insufficient energy supply [73]. These properties could possibly support the utility of yoshinone A in anti-obesity indications.

Polyaromatics
A series of brominated polyphenolics, crossbyanols A-D (48-51), were isolated from an extensive benthic Leptolyngbya crossbyana bloom in Hawaii (Figure 15) [50]. In addition to the high level of bromination, crossbyanol B-D (49-51) also have sulfated phenolic functionalities. These metabolites were suggested to play a role in the observed coral toxicity caused by the overgrowing cyanobacteria. Crossbyanol A (48) was found to activate sodium influx in mouse neuroblastoma (neuro2a) cells, with an EC50 20 μg/mL, whereas crossbyanol B (49) possessed antibiotic activity, with an MIC of 2.0-3.9 μg/mL against methicillin-resistant Staphylococcus aureus (MRSA). The latter metabolite also showed moderately potent brine shrimp toxicity (IC50 2.8 μg/mL). Crossbyanols C (50) and D (51) were not observed to have biological activity in these tests, and all four compounds were inactive as cytotoxins to H460 human lung cancer cells [50]. Three related γ-pyrone-containing polyketides described as yoshinones A, B1, and B2 (45)(46)(47), were isolated from Leptolyngbya sp. collected from Ishigaki island Okinawa, Japan ( Figure 14). The absolute configuration of each compound was attempted by a modified Mosher's method; however, these failed possibly as a result of the low amounts of samples available [49]. Thus, to assign the absolute configuration and provide sufficient amounts for further research, the absolute stereochemistry was achieved through total synthesis and comparison of NMR and chiroptical properties between the natural product and synthetic standards [72]. Yoshinone A (45) was found to inhibit adipogenic differentiation against 3T3-L1 cells, with an EC 50 of 420 nM and with little cytotoxicity (IC 50 = 63.8 µM to S. cerevisiae ABC16-monster cell) [47,49]. The adipogenic differentiation against 3T3-L1 cells of yoshinone B1 (46) and B2 (47) was considerably less potent, with less than 50% activity observed at tested concentrations up to 5 µM [48]. Further examination of structure-activity relationships in this drug class indicated that the position of the pyrone ring and side chain olefin are important for the inhibition of adipogenic differentiation. Further in vitro and in vivo experiments showed that yoshinone A (45) stimulates lactate accumulation deriving from the glycolytic system, and increases fat utilization to compensate for an insufficient energy supply [73]. These properties could possibly support the utility of yoshinone A in anti-obesity indications. Three related γ-pyrone-containing polyketides described as yoshinones A, B1, and B2 (45-47), were isolated from Leptolyngbya sp. collected from Ishigaki island Okinawa, Japan ( Figure 14). The absolute configuration of each compound was attempted by a modified Mosher's method; however, these failed possibly as a result of the low amounts of samples available [49]. Thus, to assign the absolute configuration and provide sufficient amounts for further research, the absolute stereochemistry was achieved through total synthesis and comparison of NMR and chiroptical properties between the natural product and synthetic standards [72]. Yoshinone A (45) was found to inhibit adipogenic differentiation against 3T3-L1 cells, with an EC50 of 420 nM and with little cytotoxicity (IC50 = 63.8 μM to S. cerevisiae ABC16-monster cell) [47,49]. The adipogenic differentiation against 3T3-L1 cells of yoshinone B1 (46) and B2 (47) was considerably less potent, with less than 50% activity observed at tested concentrations up to 5 μM [48]. Further examination of structure-activity relationships in this drug class indicated that the position of the pyrone ring and side chain olefin are important for the inhibition of adipogenic differentiation. Further in vitro and in vivo experiments showed that yoshinone A (45) stimulates lactate accumulation deriving from the glycolytic system, and increases fat utilization to compensate for an insufficient energy supply [73]. These properties could possibly support the utility of yoshinone A in anti-obesity indications.

Polyaromatics
A series of brominated polyphenolics, crossbyanols A-D (48-51), were isolated from an extensive benthic Leptolyngbya crossbyana bloom in Hawaii (Figure 15) [50]. In addition to the high level of bromination, crossbyanol B-D (49-51) also have sulfated phenolic functionalities. These metabolites were suggested to play a role in the observed coral toxicity caused by the overgrowing cyanobacteria. Crossbyanol A (48) was found to activate sodium influx in mouse neuroblastoma (neuro2a) cells, with an EC50 20 μg/mL, whereas crossbyanol B (49) possessed antibiotic activity, with an MIC of 2.0-3.9 μg/mL against methicillin-resistant Staphylococcus aureus (MRSA). The latter metabolite also showed moderately potent brine shrimp toxicity (IC50 2.8 μg/mL). Crossbyanols C (50) and D (51) were not observed to have biological activity in these tests, and all four compounds were inactive as cytotoxins to H460 human lung cancer cells [50].

Polyaromatics
A series of brominated polyphenolics, crossbyanols A-D (48-51), were isolated from an extensive benthic Leptolyngbya crossbyana bloom in Hawaii (Figure 15) [50]. In addition to the high level of bromination, crossbyanol B-D (49-51) also have sulfated phenolic functionalities. These metabolites were suggested to play a role in the observed coral toxicity caused by the overgrowing cyanobacteria. Crossbyanol A (48) was found to activate sodium influx in mouse neuroblastoma (neuro2a) cells, with an EC 50 20 µg/mL, whereas crossbyanol B (49) possessed antibiotic activity, with an MIC of 2.0-3.9 µg/mL against methicillin-resistant Staphylococcus aureus (MRSA). The latter metabolite also showed moderately potent brine shrimp toxicity (IC 50 2.8 µg/mL). Crossbyanols C (50) and D (51) were not observed to have biological activity in these tests, and all four compounds were inactive as cytotoxins to H460 human lung cancer cells [50].

Oxazolines
A series of polar oxazolines, named leptazolines A-D (52-55), were isolated from the culture media of a Leptolyngbya sp. (Figure 16) [51]. Their planar structures were characterized by MS and NMR along with formation of acetate derivatives. Relative configuration was determined by comparison of carbon shifts with those calculated by density functional theory (DFT). Interestingly, the calculations were found to vary as a function of the computer operating system (Ubuntu 16, Windows 10, MAC Mavericks, MAC Mojave). Biological assay showed that leptazoline B (53) modestly inhibited the growth of PANC-1 cells, with a GI50 of 10 μM [51], whereas leptazoline A (52) with its aromatic chlorine atom did not show any significant activity to this cell line.

Toxins
Cyanobacteria are known to produce a variety of toxins, including those that are hepatotoxic, neurotoxic, or cardiotoxic, and which generally increase economic burdens and impact public health [23]. Cyanobacterial populations are known to sporadically grow excessively to form blooms, and in some cases these are harmful. A total of 34 species from 15 genera and five families were screened for known toxins including the neurotoxic saxitoxin (56) and the hepatotoxic microcystins (e.g., microcystin-LR, 57) ( Figure 17) [17,23,74]. Leptolyngbya collections from the Red Sea had on average 58.9 μg/g dry wt. and 438-489 μg/g dry wt. of these two toxin classes, respectively. Toxin production

Oxazolines
A series of polar oxazolines, named leptazolines A-D (52)(53)(54)(55), were isolated from the culture media of a Leptolyngbya sp. (Figure 16) [51]. Their planar structures were characterized by MS and NMR along with formation of acetate derivatives. Relative configuration was determined by comparison of carbon shifts with those calculated by density functional theory (DFT). Interestingly, the calculations were found to vary as a function of the computer operating system (Ubuntu 16, Windows 10, MAC Mavericks, MAC Mojave). Biological assay showed that leptazoline B (53) modestly inhibited the growth of PANC-1 cells, with a GI 50 of 10 µM [51], whereas leptazoline A (52) with its aromatic chlorine atom did not show any significant activity to this cell line.

Oxazolines
A series of polar oxazolines, named leptazolines A-D (52-55), were isolated from the culture media of a Leptolyngbya sp. (Figure 16) [51]. Their planar structures were characterized by MS and NMR along with formation of acetate derivatives. Relative configuration was determined by comparison of carbon shifts with those calculated by density functional theory (DFT). Interestingly, the calculations were found to vary as a function of the computer operating system (Ubuntu 16, Windows 10, MAC Mavericks, MAC Mojave). Biological assay showed that leptazoline B (53) modestly inhibited the growth of PANC-1 cells, with a GI50 of 10 μM [51], whereas leptazoline A (52) with its aromatic chlorine atom did not show any significant activity to this cell line.

Toxins
Cyanobacteria are known to produce a variety of toxins, including those that are hepatotoxic, neurotoxic, or cardiotoxic, and which generally increase economic burdens and impact public health [23]. Cyanobacterial populations are known to sporadically grow excessively to form blooms, and in some cases these are harmful. A total of 34 species from 15 genera and five families were screened for known toxins including the neurotoxic saxitoxin (56) and the hepatotoxic microcystins (e.g., microcystin-LR, 57) ( Figure 17) [17,23,74]. Leptolyngbya collections from the Red Sea had on average 58.9 μg/g dry wt. and 438-489 μg/g dry wt. of these two toxin classes, respectively. Toxin production

Toxins
Cyanobacteria are known to produce a variety of toxins, including those that are hepatotoxic, neurotoxic, or cardiotoxic, and which generally increase economic burdens and impact public health [23]. Cyanobacterial populations are known to sporadically grow excessively to form blooms, and in some cases these are harmful. A total of 34 species from 15 genera and five families were screened for known toxins including the neurotoxic saxitoxin (56) and the hepatotoxic microcystins (e.g., microcystin-LR, 57) ( Figure 17) [17,23,74]. Leptolyngbya collections from the Red Sea had on average 58.9 µg/g dry wt. and 438-489 µg/g dry wt. of these two toxin classes, respectively. Toxin production by marine Leptolyngbya poses toxicological risks to marine organisms that may feed on them, or that may be exposed to the cyanotoxins present in seawater [74].
Mar. Drugs 2020, 18, x FOR PEER REVIEW 19 of 27 by marine Leptolyngbya poses toxicological risks to marine organisms that may feed on them, or that may be exposed to the cyanotoxins present in seawater [74].

Non-Toxic Metabolites
Non-toxic secondary metabolites from cyanobacteria include various chemical classes such as phytohormones, siderophores, and UV-absorbing compounds such as mycosporine amino acids (MAAs) and scytonemin (58); all of these have been reported in Leptolyngbya sp. (Figure 18) [23,75,76]. These latter two series of compounds have been shown to protect photosynthetic cyanobacteria from solar UV damage [23]. An investigation of UV-B photoprotective compounds in marine Leptolyngbya discovered shinorine (59) (Figure 18), which is now realized to be one of the most dominant MAAs present in several species of cyanobacteria [75]. Scytonemin has a broader absorption profile than the MAAs, protecting against the solar irradiance damage across the UV (UV-A, -B and -C; 250-425 nm). Interestingly, scytonemin also shows interesting anti-inflammatory activity through inhibition of polo-like kinase stimulated cell proliferation pathways [77]. The biosynthesis of scytonemin, deriving from the assembly of tyrosine and tryptophan derived components, has been studied at the genomic and mechanistic level in several studies; however, all of these studies have been conducted in other genera of cyanobacteria such as Nostoc punctiforme ATCC 29133 and Lyngbya aestuarii [78][79][80][81].

Non-Toxic Metabolites
Non-toxic secondary metabolites from cyanobacteria include various chemical classes such as phytohormones, siderophores, and UV-absorbing compounds such as mycosporine amino acids (MAAs) and scytonemin (58); all of these have been reported in Leptolyngbya sp. (Figure 18) [23,75,76]. These latter two series of compounds have been shown to protect photosynthetic cyanobacteria from solar UV damage [23]. An investigation of UV-B photoprotective compounds in marine Leptolyngbya discovered shinorine (59) (Figure 18), which is now realized to be one of the most dominant MAAs present in several species of cyanobacteria [75]. Scytonemin has a broader absorption profile than the MAAs, protecting against the solar irradiance damage across the UV (UV-A, -B and -C; 250-425 nm). Interestingly, scytonemin also shows interesting anti-inflammatory activity through inhibition of polo-like kinase stimulated cell proliferation pathways [77]. The biosynthesis of scytonemin, deriving from the assembly of tyrosine and tryptophan derived components, has been studied at the genomic and mechanistic level in several studies; however, all of these studies have been conducted in other genera of cyanobacteria such as Nostoc punctiforme ATCC 29133 and Lyngbya aestuarii [78][79][80][81].
Mar. Drugs 2020, 18, x FOR PEER REVIEW 19 of 27 by marine Leptolyngbya poses toxicological risks to marine organisms that may feed on them, or that may be exposed to the cyanotoxins present in seawater [74].

Non-Toxic Metabolites
Non-toxic secondary metabolites from cyanobacteria include various chemical classes such as phytohormones, siderophores, and UV-absorbing compounds such as mycosporine amino acids (MAAs) and scytonemin (58); all of these have been reported in Leptolyngbya sp. (Figure 18) [23,75,76]. These latter two series of compounds have been shown to protect photosynthetic cyanobacteria from solar UV damage [23]. An investigation of UV-B photoprotective compounds in marine Leptolyngbya discovered shinorine (59) (Figure 18), which is now realized to be one of the most dominant MAAs present in several species of cyanobacteria [75]. Scytonemin has a broader absorption profile than the MAAs, protecting against the solar irradiance damage across the UV (UV-A, -B and -C; 250-425 nm). Interestingly, scytonemin also shows interesting anti-inflammatory activity through inhibition of polo-like kinase stimulated cell proliferation pathways [77]. The biosynthesis of scytonemin, deriving from the assembly of tyrosine and tryptophan derived components, has been studied at the genomic and mechanistic level in several studies; however, all of these studies have been conducted in other genera of cyanobacteria such as Nostoc punctiforme ATCC 29133 and Lyngbya aestuarii [78][79][80][81].

Odorous Metabolites
Geosmin (86) and 2-methylisoborneol (87) are two earthy-musty odorous terpenoid secondary metabolites that were first isolated from actinomycetes ( Figure 23). Later, these same molecules were found to have a significant presence in many cyanobacteria species. These simple terpenoids have each been reported numerous times as being among the main causes for off-flavors in water and other products [85]. Wang et al. isolated 86 and 87 from Leptolyngbya bijugata strains, and quantified each at 13.6-22.4 and 12.3-57.5 μg/L, respectively, demonstrating their production by Leptolyngbya [86].

Odorous Metabolites
Geosmin (86) and 2-methylisoborneol (87) are two earthy-musty odorous terpenoid secondary metabolites that were first isolated from actinomycetes ( Figure 23). Later, these same molecules were found to have a significant presence in many cyanobacteria species. These simple terpenoids have each been reported numerous times as being among the main causes for off-flavors in water and other products [85]. Wang et al. isolated 86 and 87 from Leptolyngbya bijugata strains, and quantified each at 13.6-22.4 and 12.3-57.5 μg/L, respectively, demonstrating their production by Leptolyngbya [86].

Odorous Metabolites
Geosmin (86) and 2-methylisoborneol (87) are two earthy-musty odorous terpenoid secondary metabolites that were first isolated from actinomycetes ( Figure 23). Later, these same molecules were found to have a significant presence in many cyanobacteria species. These simple terpenoids have each been reported numerous times as being among the main causes for off-flavors in water and other products [85]. Wang et al. isolated 86 and 87 from Leptolyngbya bijugata strains, and quantified each at 13.6-22.4 and 12.3-57.5 µg/L, respectively, demonstrating their production by Leptolyngbya [86]. Geosmin (86) and 2-methylisoborneol (87) are two earthy-musty odorous terpenoid secondary metabolites that were first isolated from actinomycetes ( Figure 23). Later, these same molecules were found to have a significant presence in many cyanobacteria species. These simple terpenoids have each been reported numerous times as being among the main causes for off-flavors in water and other products [85]. Wang et al. isolated 86 and 87 from Leptolyngbya bijugata strains, and quantified each at 13.6-22.4 and 12.3-57.5 μg/L, respectively, demonstrating their production by Leptolyngbya [86].

Pigments
Phycocyanin is a pigment-protein complex found in cyanobacteria and eukaryotic algae that functions as a light-harvesting pigment. It is widely used in biotechnological, food and pharmaceutical industries [87]. Schipper et al. analyzed the phycocyanin content in Leptolyngbya sp. QUCCCM 56 from a desert environment to reveal that it possesses higher and purer phycocyanin compared to the current commercial source, Arthrospira platensis [87]. Other than absorbing light, phycocyanin also shows potential anti-aging and proteostasis-suppressive activities. With phycocyanin treatment, the life span of wild-type (N2) C. elegans was extended from 14.8 to 19.1 days [88].

Pigments
Phycocyanin is a pigment-protein complex found in cyanobacteria and eukaryotic algae that functions as a light-harvesting pigment. It is widely used in biotechnological, food and pharmaceutical industries [87]. Schipper et al. analyzed the phycocyanin content in Leptolyngbya sp. QUCCCM 56 from a desert environment to reveal that it possesses higher and purer phycocyanin compared to the current commercial source, Arthrospira platensis [87]. Other than absorbing light, phycocyanin also shows potential anti-aging and proteostasis-suppressive activities. With phycocyanin treatment, the life span of wild-type (N2) C. elegans was extended from 14.8 to 19.1 days [88].

Laboratory Cultivation
Despite many attempts, only a few environmental collections of Leptolyngbya have been propagated under laboratory culture conditions and, from these, there has been a relatively low rate of natural product isolation. Rather, most of the compounds reported in this genus have been discovered from environmental samples (Table 1). Martins et al. screened five Leptolyngbya strains among 28 cyanobacteria samples collected from the Portuguese Coast and cultured in Z8 medium, and showed cytotoxic activities of the extracts against multiple human tumor cell lines [15,16,89]. The most commonly used media for cyanobacteria culture is BG11 [53], originally formulated in 1988 to possess synthetic sea salt, microelements mixture, deionized water and vitamin mixtures [28]. In most cases, even with suitable media and abiotic factors, filamentous cyanobacteria are slow-growing life forms, normally with growth rates much slower than algae and other bacteria, thereby presenting a challenge for secondary metabolite discovery due to the small quantities of biomass produced in cultures [28,90]. The 2-hydroxyethyl-11-hydroxyhexadec-9-enoate, by contrast, was able to be isolated from the laboratory culture of Leptolyngbya with the relatively small biomass of 383.6 g because of its reduced structural complexity and higher production yield [41].

Conclusions
Leptolyngbya is a widely distributed genus of cyanobacteria that has emerged to be a very rich source of structurally novel and biologically active natural products. However, to date, this genus appears to be underexplored for its chemical, biological and biosynthetic potential when compared to some other genera of filamentous cyanobacteria, such as Moorena and Symploca [90]. Working with Leptolyngbya has been challenging due to the difficulty in bringing it into culture in the laboratory environment. This has impeded not only new compound discovery, but also exploration of its genomic characteristics, including those that are responsible for natural product biosynthesis. It is possible that developments with the heterologous expression of cyanobacterial natural product pathways will enable a more extensive exploration of the rich secondary metabolome of this genus in the future [91,92].