Environmental and cultured cyanobacteria as sources of Aedes aegypti larvicides

In tropical countries, the control of the mosquito Aedes aegypti is a public health priority due to its role as a vector of important viral diseases. Marine cyanobacteria are recognized as abundant sources of bioactive compounds, and they constitute a potential source of insecticides useful for controlling mosquito populations and preventing epidemic outbreaks. We collected 30 benthic cyanobacterial mats in Providencia and Rosario islands (in the Colombian Caribbean) belonging to the genera Phormidium, Symploca, Oscillatoria, Lyngbya, Pseudoanabaena, Leptolyngbya, Moorea, and Dapis. Fractions of organic extracts from the most abundant environmental samples were evaluated in three bioassays, assessing (i) larvicidal activity against A. aegypti, (ii) toxicity against the brine shrimp (Artemia salina) nauplii, and (iii) acetylcholinesterase inhibition. Non-polar fractions exhibited larvicidal activity. The polar fraction from one Dapis pleuosa extract showed larvicidal activity without being toxic against A. salina nauplii. Extracts from Moorea producens exhibited the greatest toxicity against A. aegypti larvae and A. salina nauplii. From 23 cultured cyanobacterial samples, only five grew under laboratory conditions and produced enough biomass to yield organic extracts. Of these, three extracts showed strong larvicidal activity, but only the extract from Phormidium tenue showed reduced toxicity against A. salina nauplii. We detected variation among the chemical profiles and larvicidal activity of cyanobacterial consortia depending on sites and dates of collection. Our findings suggest that despite variation in chemical profiles, extracts of marine benthic cyanobacteria can be further developed as effective control agents against insect vectors, in their larval stages. The culture of marine benthic cyanobacteria needs to be further explored to provide enough biomass leading to the identification of bioactive compounds with public health applications.


Introduction
Mosquitoes (Diptera: Culicidae) are important vectors of several viral diseases and are a topic of public health concern in tropical countries. Diseases such as dengue, chikungunya, zika, and yellow fever are mainly transmitted via Edited by Juan Carlos Salcedo-Reyes (salcedo.juan@javeriana.edu.co) the mosquito Aedes aegypti [1]. These four diseases have a large impact on public health systems due to their high incidence, increasing infection rates, and increasing treatment costs. In Latin America, the recent outbreaks of zika and chikungunya exemplify the gravity of these diseases [2,3]. Efforts to control these diseases have focused on targeting either their underlying viruses or their vectors [4]. Vector control via insecticide applications constitutes a widespread alternative to avoid the growth of large mosquito populations. However, insecticide-associated environmental and health impacts, together with the development of insecticide resistance in many mosquitoes, including A. aegypti, [5], call for new sources of compounds to control mosquito populations.
Cyanobacteria are microorganisms that thrive in a great diversity of aquatic and terrestrial environments. Their survival is linked to their broad array of metabolic products associated with defensive and competitive (algaecides) functions. These metabolites include, but are not limited to, cyclic and linear peptides, guanidines, purines, lipids, macrolides [6,7]. Recently, some studies have evaluated the potential of cyanobacteria and their metabolites towards the development and production of application in the areas of biofuels, anticancer agents, and pesticides [6]. Marine cyanobacteria-derived compounds, such as unsaturated fatty acids and sulfated glycolipids, have revealed promising activities against larvae of the mosquito species A. aegypti and Aedes albopictus [8,9]. Hence, cyanobacterial metabolites represent a potential source of compounds that may be used as an alternative for controlling Aedes larvae.
One of the main challenges when utilizing marine natural products in the pharmaceutical and agrochemical industries is their supply [10]. This challenge escalates when promising compounds are derived from marine organisms with low population densities. With cyanobacteria, however, this may not be the case. The increase, recurrence, and persistence of large cyanobacterial blooms, as a consequence of excessive nutrient input into coastal waters and global warming [11], propitiates the obtention of sufficient amounts of cyanobacterial compounds with biological activity. Also, the culture of marine microorganisms is a feasible approach to overcome supply limitations [12][13][14][15]. Most chemical studies of cyanobacteria have been thus far performed on environmental samples [16,17]. Therefore, the culture of cyanobacteria, although challenging because these organisms do not easily respond to in vitro conditions [18][19][20], could help understanding bloom dynamics and provide enough biomass to obtain promising bioactive compounds [21,22]. As a continuation of our studies on the chemistry and potential use of cyanobacteria from the Colombian Caribbean [23], in this work we explored the insecticidal activity of polar and non-polar extracts of marine cyanobacteria against A. aegypti larvae. We assessed the potential toxicity of these extracts with assays involving Artemia salina nauplii. Lastly, we investigated a possible mechanism of action of these extracts as acetylcholinesterase (AChE) inhibitors. All our evaluated extracts were obtained from environmental samples and from cultured cyanobacterial mats ( Fig. 1) that yielded enough biomass to perform the assays.

Cyanobacterial mat collection
Samples of benthic cyanobacterial mats were collected in waters off the coasts of Old Providence Island and the Archipelago of Islas del Rosario, Colombia. Old Providence Island is part of the Archipelago of San Andrés, Old Providence, and Santa Catalina, a group of oceanic islands, submerged banks, and atolls located approximately 300 km off the coast of Nicaragua [25,26]. The shallow reef complex in Old Providence has a total area of 285 km 2 and is characterized by the development of different coral formations [27][28][29]. The archipelago of Islas del Rosario (Departamento de Bolívar) consists of a group of islands and low coral reefs of recent origin. In Old Providence, samples were collected at Felipe's Place (13 • 18'-13 • 24' N; 81 • 19-81 • 25' W), at depths between 10-20 m, and in Islas del Rosario sampling sites were Ministerio (10 • 11'06" N, 75 • 43'55" W), at depths between 10-20 m and Pavitos (10 • 10'30" N, 75 • 46'17" W), at a depth between 20-25 m. All sites are well-developed fringing reefs, with a predominance of patches with low coral diversity that determine their ecology [30].
Immersions were performed at each collection site from June to August 2009August , 2010August , 2015, and 2016 to detach and collect cyanobacterial mats from sand and live substrates. During the sampling months, the northeast winds intensify, temporarily inhibiting the rainy season [24]. Upon collection, cyanobacterial mats were deposited in mesh bags, keeping track of site and depth of collection. The collected material was cleaned from debris and kept frozen (-4 • C) until chemical studies were performed. Small mat samples (50 cc) were stored in 10 % formalin in seawater to perform morphometric analyses. Vouchers for each cyanobacterial mat were deposited at the IBUN (Instituto de Biotecnología-Universidad Nacional de Colombia) collection.

Identification of cyanobacterial samples
Cyanobacterial identification was performed on morphological characters. To identify the different cyanobacterial morphotypes, at least three portions from each sample were observed under a Nikon optical microscope connected to a digital camera. The captured images were analyzed using NIS-Elements Br 2.30 Nikon Imaging Software, which allowed performing measurements of each morphological character under several magnifications up to 100X. Distinctive characters included: trichome length, width, and number of ramifications; the presence/absence of specialized cells such as heterocysts or akinetes; presence of a facultative (fine to stratified) sheath; cell length to width ratio; the presence of calyptrae; the shape of apical cells; intercellular constrictions; and presence of mucilaginous cases [31][32][33].
Chemical separation analyses were performed with the following analytical quality solvents: MeOH, DCM, and BuOH (Merck, Darmstadt, Germany). Thin layer chromatography was performed on aluminum plates precoated with silica gel 60F254 (Merck, Darmstadt, Germany). HPLC-ELSD analyses were performed on a Thermo Dionex ultimate 3000 system, coupled to an ELSD Sedex 85 detector (Sedere, France) with a gain of 10 for the ELSD detector and a temperature of 80 • C.

Extraction and chemical partitions of environmental cyanobacteria samples
The extraction step on each sample was performed using a DCM/MeOH (1:1) mixture. Solvents were removed under reduced pressure to yield a crude extract. Each crude extract was weighed and then resuspended in water and mixed with dichloromethane to yield the DCM (FD) fraction. The water layer was then extracted with butanol to yield each butanol (FB) fraction and the residual water fraction (WW). Finally, FD and FB fractions were assayed as is described below. A further separation scheme was performed with sample IBUN-02224. The crude extract from this cyanobacterial mat was fractionated over a DIOL cartridge (5 g) using the following mobile phase composition, Hex/EtOAc 8:2, Hex/EtOAc 1:1 EtOAc 100 %, EtOAc/MeOH 1:1, and MeOH and yielding five fractions of increasing polarity (FI-FV).

Cyanobacterial mat culture conditions
Small portions of cyanobacterial mats (1 g) were cleaned from debris and stored in sterile vials with seawater from the sampling site. Samples were kept under sunlight and brought to the laboratories of Universidad Nacional and Universidad Jorge Tadeo Lozano in Bogota. Once in the laboratory, samples were rinsed with sterile artificial seawater (salinity 35) and then suspended in sterile SWBG-11 culture medium (30 mL) [34]. Cyanobacterial samples were grown at 25 • C, with a 12 h light/dark regime and without aeration, seeking to not affect biofilm development. Cultured cyanobacteria were kept under these conditions for two months changing the SWBG-11 culture medium every other week.
Culture escalations were performed every two months. Surviving cyanobacterial samples were transferred to 100 mL in a 250-ml Erlenmeyer flask, maintaining the same culture conditions as described above. Then, a further escalation was done to 200 mL of SWBG-11 medium in 500-mL Erlenmeyer flasks, and two months afterwards they were transferred to 300 mL of SWBG-11 medium in 750-mL Erlenmeyer flasks.

Extraction and chemical partitions of cultured cyanobacteria
Cultured cyanobacterial samples were extracted using a DCM/MeOH (1:1) mixture as described previously, in the environmental sample extraction section. Solvents were removed under reduced pressure yielding a crude extract. Each crude extract was weighed, resuspended in water, and extracted with EtOAc.

Larvicidal bioassay against Aedes aegypti
Larvicidal assays were performed following Berry et al. [9]. A. aegypti eggs were allowed to hatch in hypoxic deionized water at 28 • C during 30-40 min. Assays were performed in 24-well plates. The tested fractions were either the FD or FB fractions from the environmental samples or the EtOAc fraction from the cultured cyanobacteria. All fractions were evaluated in triplicate. For the assay, fractions were resuspended in CHCl 3 and MeOH (1 mg/mL) to solubilize them; then 50 µg/mL of each test solubilized fraction were added in each well and mixed with deionized water to a final volume of 1 mL. Solvents were removed by evaporation before water was added to each well. Negative controls consisted of 10 µL of CHCl 3 and MeOH, the positive control consisted of 10 µg/mL 4,4'-DDT. Four freshly hatched (instar I) mosquito larvae were added to each well. Plates were incubated at 28 • C, with 12/12 h light/darkness periods for 6 days. Larval development was monitored during all four instars (six days). Each larva was fed with 20 µL of a 1 % liver powder solution. The number of dead/live larvae in each well was counted every 24 h. Test fractions that exhibited mortality greater than 50 % after 6 days were considered active. assay, 10 µL or 50 µL from a 5 mg/mL solution of the FD and FB fractions or the EtOAc fraction of cultured cyanobacteria were added in each well and mixed with artificial sea water to reach a final volume of 2 mL. Acetone was used to solubilize the FD and FB fractions from environmental samples and the EtOAc fractions from cultured cyanobacteria. Negative controls consisted of 10 µL of acetone, the positive control consisted of 10 µg/mL of malathion. 10 Artemia nauplii were added to each well. Plates were incubated at 28 • C. After 24 h dead/live nauplii were counted in each well. Test fractions that exhibited a LC 50 ≤ 12.5 µg/mL after 24 hours were considered active [35].

Acetylcholinesterase inhibition bioassay
Crude extracts and fractions were tested for their AChE inhibitory activity following the Marston method [36]. Briefly, an AChE solution (4 UA/mL) was prepared in phosphate buffer 0.1 M, pH = 7.4, with BSA (1 mg/mL). 1-naphthyl acetate was used as a substrate solution and prepared in 96 % ethanol (2.5 mg/mL). A Fast-Blue Salt (FBS) dye solution in water (2.5 mg/mL) was prepared prior to use. All test fractions were resuspended in dichloromethane or methanol and applied on a TLC plate for the AChE inhibition test. Malathion (10 µg/spot) was used as a positive control. TLC plates were eluted with a mixture of n-hexane/EtOAc in a 7:3 ratio for FD fractions, whereas a mixture of CHCl 3 /MeOH in a 9:1 ratio was used for the FB fractions of environmental samples. TLC plates were dried and sprayed uniformly with an AChE solution and incubated for 30 minutes at 37 • C.
Plates were then sprayed with the 1-naphtyl-acetate solution and incubated for 30 min at 37 • C. Finally, the plates were sprayed with the FBS dye solution until a purple coloration was observed (1-2 minutes). The presence of white spots on the plate was associated with AChE inhibition [37].

Results and discussion
Cyanobacterial mat composition and culture Thirty samples of cyanobacterial benthic mats were collected from different marine substrates such as sediments, algae, and soft corals. We have recently documented the potential allelopathic effects of cyanobacterial extracts from live corals and soft corals [38]. All collected cyanobacteria showed filamentous and turf morphologies ( Table 1). The preliminary taxonomic classification based on morphometric parameters allowed the identification of cyanobacteria belonging to the genera Phormidium, Symploca, Oscillatoria, Lyngbya, Pseudoanabaena, Leptolyngbya, Moorea, and Dapis. A mat of Dapis Islas del Rosario pleousa was identified using morphological and chemical traits, namely the production of malyngolide (unplublished data). Malyngolide is a diagnostic chemical marker for this species and may be used as a chemotaxonomical marker according to Engene et al. [39]. Some of the collected samples were not identified because of insufficient sampled biomass and failure to grow under laboratory conditions.
Among the collected cyanobacteria, Lyngbya-like mats were very conspicuous in the Caribbean Sea. Almost 350 compounds have been isolated from Lyngbya-like species, an unusual number for a single genus [34]. Results of morphological and molecular trait studies in Lyngbya, by Engene et al., revealed that this genus is a polyphyletic group. Consequently, the genus has been reassigned in three different genera, Moorea [16], Okeania [40], and Dapis [39]. However, there are still other Lyngbya species that have not been reclassified yet [34]. All these three genera produce a vast number of metabolites, many of which are genus or species-specific. These metabolites are used nowadays as chemotaxonomic markers [33].  Cyanobacterial cultures and further escalation were initiated with the most common mats (Table 1 and Table 2). The least abundant mats were just cultured in order to obtain enough biomass for chemical studies. Thanks to these cyanobacteria cultures enough biomass was available to perform thorough molecular characterizations of these microorganisms and to isolate, analyze, and elucidate their bioactive compounds. In a previous work, and following Bertin et al. [41], we established that the SWBG-11 culture medium was the most appropriate for culturing marine cyanobacteria (Unpublished data). Our formulation of the SWBG-11 medium (equivalent to natural eutrophication conditions [20]) included high concentrations of phosphorus, nitrogen, and iron (III).  (Table 3). Cyanobacteria that tolerated well the escalation process belonged to the genera Lyngbya, Leptolyngbya, and Phormidium. These are filamentous cyanobacteria with very narrow trichomes that may be effective for nutrient uptake [19,20,42]. After 9 months, only five samples attained enough biomass to provide enough material to obtain extracts and run larvicidal assays (Table 3).
Larvicide and ecotoxic activities of fractions from environmental cyanobacterial samples As explained in the methods section, a crude extract from each sampled cyanobacterial mat with a wide polarity range was obtained with DCM/ MeOH (1:1). Further fractionation yielded a fraction of low-medium    Table 2.
Larvicidal activity tests against A. aegypti were monitored trough all four instars (six days in total). Larvae in the control solvent elicited a mortality below 10 %, which is within the range larval natural mortality [9]. Eight out of 35 fractions tested were active. Most of them corresponded to the FD fractions (with a low-medium polarity). These results are consistent with those reported by Harada [8] and Berry [9]. These studies detected larvicidal activity in the lipid fraction, due to the presence of unsaturated fatty acids (oleic, linoleic, and γ -linolenic acid), and polar fractions due to the presence of sulfated glycolipids [8,9]. In our case, only a polar fraction, obtained from mat IBUN-02213 (Depis pleousa), showed larvicidal activity. Additionally, the FIII fraction (medium polarity) from Phormidium tenue (IBUN-2224) was active against A. aegypti larvae.
The FD fractions of the Phormidium submembranaceum-Symploca hynoides consortium (PNN-18) and M. producens (IBUN-03496) mats were the most active extracts against mosquito larvae (eliciting 100 % mortality at 50 µg/well). Symploca species are well known to produce potent, cytotoxic compounds such as symplostatin 1 and 2 [44,45]. The Moorea genus is know as a rich source of bioactive natural compounds, however, there are no reports of insecticidal compounds isolated from this genus according to the Marinlit database [46]. The non-polar fraction (FD) of the Lyngbya-Oscillatoria consortium (PNM-28) revealed good larvicidal activity. Unsaturated fatty acids with larvicidal activity have been isolated from Oscillatoria aghardii [7]. This kind of compunds may be responsible for the activity observed in this fraction.
Both larvicidal activity and metabolic profiles of the P. We also observed spatial variation in the bioactivity and metabolic profiles of FD fractions from M. producens mats collected at different locations. Samples IBUN-03495, IBUN-03496, IBUN-03497, and IBUN-03498) were collected off Isla Grande (Islas del Rosario) in June 2016 at different but nearby locations. The FD fractions from IBUN-03495 and IBUN-03496 elicited A. aegypti larvae mortalities of 83 % and 100 %, respectively, whereas samples IBUN-03497 and IBUN-03498 resulted in mortalities of 45 % and 23 %, respectively. The HPLC-ELSD analysis from these FD fractions showed different chemical profiles, revealing chemical profile differences between the most active and the least active fractions (Fig. 2). This suggests that differences in larvicidal activity may reflect varying metabolic profiles associated to particular microenvironmental conditions at each collecting site or ecological interactions taking place at the time of collection. A single cyanobacterial species may produce different metabolites depending on biotic and abiotic conditions such as temperature, light intensity, or nutrient availability [47].
Studies taking into consideration both the variation of environmental parameters and ecological interactions (i.e. space competition or protection from grazing), must be conducted before ascribing metabolite production to a particular factor. On the other hand, the observed metabolic variation could be explained by genetic differences. For instance, the Phormidium sp.  Both samples, showed differences in larvicidal activity at 50 µg/mL (eliciting 0 % and 83 % larvae mortality, respectively), but their chromatograms were very similar (Fig. 3). Possibly, the compounds responsible for larvicidal activity were in very low quantities and were not readily detected in the chemical profiles.
Lethality tests against the brine shrimp A. salina are a general approach to evaluate the toxicity of extracts, fractions, or compounds. However, some researchers have correlated those results with insecticidal activity [48] and to the selectivity of a particular compound acting as an insecticide [49]. We expected a promising larvicidal substance to be very active against A. aegypti larvae and exhibiting little toxicity against A. salina. However, we did not discard extracts active against both A. aegypti and A. salina. In our toxicity assays, test fractions that exhibited a LC 50 ≤ 12.5 µg/mL after 24 hours were considered active. All the non-polar (FD) fractions caused some mortality, but only 12 out of 15 were considered toxic, while only 5 of the 15 polar fractions tested (FB) were toxic against nauplii of A. salina. All FD fractions were rich in lipidic compounds. Some fatty acids have been reported active against A. salina and other arthropods, inhibiting Na + and K + ATPases [50]. The presence of fatty acids in FD fractions could explain the lethality results against A. salina. The most toxic extracts were obtained from samples of M. producens (IBUN-03495, IBUN-0349, IBUN-03497, and IBUN-03498) and mixed consortia of P. submembranaceum-S. hynoides showed varying toxicity. Three out of the four consortia extracts were toxic. For instance, sample PNM-08 showed a 100 % mortality at a concentration of 50 µg/mL, whereas extract PNM-18 was not toxic at any of the concentrations tested. Again, a varying degree in biological activity became evident in samples collected in different years. Samples PNM-07, PNM-08, and PNM-13, collected in 2010 at Isla Grande, were toxic to Artemia nauplii. However, a sample of M. producens collected at the same site in 2009 did not exhibit such activity. Discussing the putative causes of such variation is out of the scope of this paper, but they are likely related to the interaction of environmental variables such as increased nutrient input and increased water temperatures affecting the reefs at Islas del Rosario [51]. Our results failed to establish a positive correlation between larvicidal activity and toxicity to A. salina, as opposed to other reports in the literature [48].

AChE inhibitory effect of fractions from environmental cyanobacterial samples
In the AChE inhibition assay, all the FD fractions tested inhibited acetylcholinesterase. The most active samples consisted of Lyngbya-Oscillatoria consortia (sample PNM-28); revealing strong activity against A. aegypti larvae and low toxicity against A. salina. This result could reflect insecticidal activity mediated by AChE inhibition. Interestingly, some of the health hazards linked to freshwater cyanobacterial blooms have been related to acetylcholinesterase inhibition [52]. Furthermore, neurotoxic effects on birds are caused by cyanobacterial toxins able to inhibit cholinesterases [53]. The most representative AChE inhibitor compound from cyanobacterial sources is anatoxin-a, isolated from the planktonic species Anabaena flos-aquae and produced by other freshwater cyanobacteria [54].
Before conducting the AChE inhibition assays, we hypothesized that strong AChE inhibitors would show significant larvicidal activity against A. aegypti, following the mode of action of insecticidal organophosphates and carbamates [55]. However, some of the cyanobacterial samples tested showed strong AChE inhibition without being larvicidal. This could be partly ascribed to structural differences in the acetylcholinesterase of insects compared to that . Alternatively, we propose that compounds in the test extracts may not have crossed insect biological barriers, as an explanation for the inhibition activity on the enzyme and the absence of activity on the mosquito larvae. Nevertheless, this hypothesis remains to the be tested.
Larvicidal extracts that do not inhibit acetylcholinesterase are promising insecticides because of the wide-spread insecticide resistance against this target [56,57]. Additionally, a mechanism of action that does not involve this enzymatic target, allows for better selectivity and less toxicity in non-target organisms. This pattern was observed in the FD fraction of sample PNM-18 (from a P. submembraceum-S. hydnoides consortium) as well as in fraction III of the extract from D. pleousa (sample IBUN-02214).
In summary, the most promising cyanobacterial extracts were obtained from M. producens (samples IBUN-03495 and IBUN-03496), followed by extracts from P. submembranaceum-S. hynoides mixed consortium (PNM-18), Lyngbya spp.-Oscillatoria spp. mixed consortium (PNM-28), and Phormidium sp. (IBUN-03494). Mats of M. producens were locally abundant, thick, and conspicuous but exhibited large chemical variation depending on collection sites and dates. Extracts from mixed cyanobacterial consortia and from Phormidium sp. were less abundant and yielded small amounts of crude extracts. The larvicidal activity of the FIII fraction of the Phormidium tenue extract (IBUN-0224) showed larvicidal activity being moderately toxic against A. salina. Hence, this extract could be considered a selective larvicidal agent. Additionally, this fraction did not inhibit AChE, suggesting a different mode of action, which could be of interest in the light of the recurrent reports of insecticide resistance.

Larvicide and ecotoxic activities of cultured cyanobacteria extracts
Crude extracts of five cyanobacterial cultures were partitioned between EtOAc and water to yield a fraction of low-medium polarity. Three out of the five fractions tested, from cultures of the genera Phormidium and Lyngbya, showed strong larvicidal activity (100 % mortality; Table 4), and two fractions from cultures the genus Leptolyngbya exhibited mild-to-low larvicidal activity (≤ 40 % mortality; Table 4). In contrast to Phormidium fractions, those from Leptolyngbya were less active against mosquito larvae and were characterized by their fast growth. It is thought that fast growing cyanobacteria are more effective in nutrient uptake, whereas slow growing mats invest more energy resources in defensive compound biosynthesis. Studies from Leptolyngbya collected in Florida showed larvicidal activity, but also spatial and temporal metabolic variations with direct consequences on their biological activity [7,9].  Table 4. Bioassays results for crude extract of cultured cyanobacteria. The Aedes aegypti larvae percent mortality was evaluated at 50 μg/ml. evaluation, while the Artemia salina percent mortality was tested at 12.5 μg/ml after six days.
Larvicidal extracts from Lyngbya consortia (samples IBUN-02220 and IBUN-02221) were also active against A. salina nauplii, showing activity comparable to that of organophosphate insecticides [58]. They did not inhibit the acetylcholinesterase, however. The organic extract of cultured Phormidium tenue (sample IBUN-02224) caused a 97 % mortality of A. aegypti larvae after six days of exposure, had a moderate AChE inhibitory activity, and exhibited very low toxicity against A. salina nauplii (Table 4). After the assay, surviving mosquito larvae stopped their development in instars 2 and 3, suggesting an ontogenic development interruption elicited by this extract. There are few reports on this regard because most studies evaluate mosquito mortality in the third and fourth instars of their development. Growth interruption implies that mosquito larvae will not reach their adult stage, and therefore, they will not become vectors of viral diseases.

Conclusions
The mats of marine benthic cyanobacteria subject of this study were complex consortia belonging to the genera Phormidium, Symploca, Oscillatoria, Lyngbya, Pseudoanabaena, Leptolyngbya, Moorea, and Dapis in varying proportions. We were able to culture and perform gradual escalations of In general, the non-polar fractions of the tested cyanobacterial extracts showed the greatest larvicidal activity, were toxic against A. salina nauplii, and showed significant AChE inhibition. Nonetheless, the polar fraction from D. pleuosa showed larvicidal activity without being toxic to A. salina nauplii. Since benthic mats of this cyanobacterium can be locally abundant, they are worth a thorough chemical study. The non-polar fractions from the extract of M. producens (IBUN-03496) showed strong larvicidal activity being toxic to A. salina nauplii. The medium-polarity fraction from Phormidium tenue showed interesting larvicidal activity, without being toxic against A. salina nauplii, and without inhibiting AChE.
We found important differences in A. egypti larvicidal activity and metabolic profiles of cyanobacterial samples collected at the same sampling sites in different years. For example, between mixed consortia of Phormidium submembranaceum-Symploca hynoides collected in Old Providence in 2009 and 2010. From the cultured cyanobacteria mats, the extracts of Phormidium and Lyngbya showed strong larvicidal activity but only the extract of Phormidium tenue showed larvicidal activity with a potential low toxicity, as determined in the lethality assay against A. salina nauplii.
Despite their chemical profile variations, our marine benthic cyanobacteria extracts could be further developed as control mechanisms against larval stages of A. aegypti due to their larval toxicity, potential selective mechanisms, and low ecotoxicity in many cases. The cultured of marine benthic cyanobacteria requires further explored to provide enough biomass and to obtain interesting bioactive compounds.

Farja I. Ayala
Farja Isabel Ayala. Chemist. MSc in Chemical Sciences (Universidad Nacional de Colombia). Catheadratic professor in Universidad de Ciencias Aplicadas y Ambientales. She did her master studies in the research group "Estudio y Aprovechamiento de Productos Naturales Marinos y Frutas de Colombia" on biprospection of cyanobacterial mats. Interested in structural determination of natural products and biomolecules in signaling pathways.

Laura M. Becerra
Laura Becerra, born in Bogotá -Colombia in 1990, received her B.Sc. in Marine Biology from Jorge Tadeo Lozano University. Then, she earned her master's degree in microbiology at Universidad Nacional de Colombia. Her thesis was performed in the research group "Estudio y Aprovechamiento de Productos Naturales Marinos y Frutas de Colombia", in which she focused on cyanobacterial mats biprospection by metabolomics. Currently, she is lecturer in Universidad Manuela Beltrán.

Jairo Quintana
Jairo Quintana, chemist from Universidad Nacional de Colombia, carried out his undergraduate and masters degree studies at "Estudio y Aprovechamiento de Productos Naturales Marinos y Frutas de Colombia" research laboratory working on the search of quorum sensing inhibitors from marine macroorganisms and bioprospection of marine cyanobacteria, respectively. He is currently conducting his PhD studies at Universidade de São Paulo.

Lina M. Bayona
Lina María Bayona born in 1990 and studied her bachelor in chemistry and MSc in Chemical Sciences in the Universidad Nacional de Colombia. Her master's thesis was done in the research group "Estudio y Aprovechamiento de Productos Naturales Marinos y Frutas de Colombia" focused on marine cianobacteria metabolomics. Currently she is a PhD candidate in Universiteit Leiden and currently working on metabolomics of marine sponges . Interested in natural products chemistry, NMR and LC-MS based metabolomics. Vast experience in the chemical ecology of marine invertebrates. In the last 10 years, she has studied the ecological implications of blooms of marine benthic cyanobacteria in the Caribbean. She is also involved in seaweed culture initiatives and the development of seaweed derived products at Providencia Island.

Fredy Duque
Fredy Duque was born in Bogota, Colombia in 1985. He performed his undergraduate's studies in Biology in Universidad Nacional de Colombia. He did his master studies in Universidad Jorge Tadeo Lozano, in Environmental Sciences, on benthic cyanobacterial bloom in the Colombian Caribbean. Currently, he is performing the taxonomy of benthic cyanobacteria in the Seaflower Expedition in Archipelago of San Andres in Colombia.

Leonardo Castellanos
Leonardo Castellanos (Chemist, Ph.D.) is the head leader of "Estudio y Aprovechamiento de Productos Naturales Marinos y Frutas de Colombia" research group at Universidad Nacional de Colombia. He is interested in bioprospecting of marine invertebrates and their biotechnological applications. Currently, he has an associated professor of the Chemistry department of Universidad Nacional de Colombia.