Abstract
Mosquito larvae continuously encounter microbes in their aquatic environment, which serve as food and play a critical role in successful development. In previous work, we isolated a Chromobacterium sp. (C.sp_P) with larvicidal activity from the midgut of dengue vector Aedes mosquitoes in Panama. In this study, we found a positive correlation between initial concentrations of C.sp_P and larval mortality rates, and that C.sp_P is more efficient at inducing larval mortality in a high nutrient environment. Multiple Chromobacterium species induce larval mortality with similar efficacy to C.sp_P except for C. subtsugae. We also found that a non-lethal dose of C.sp_P lengthens development time and increases mortality over multiple developmental stages, suggesting persistent effects of exposure. Additionally, we showed that larvicidal activity persists in the larval breeding water after removal of live bacteria, and that the larvicidal factor in C.sp_P-treated water is smaller than 3 kDa, heat resistant to 90 °C, and lost after vacuum centrifugation. We showed that C.sp_P produces hydrogen cyanide in culture and in larval water at concentrations sufficient to kill An. gambiae larvae, and treatment of the larval water with a cyanide antidote eliminated larvicidal activity. We conclude that a potential mechanism by which C.sp_P can induce larval mortality is via production of hydrogen cyanide.
Similar content being viewed by others
Introduction
Mosquitoes, including those transmitting human diseases, co-exist with diverse communities of bacteria throughout their lives, especially during the larval and pupal stages, when they dwell in bacteria rich bodies of water1,2,3,4. Bacteria serve as a food source for larvae, and colonization of the larval gut by bacteria is critical for proper larval development5,6. Recent work suggests that bacteria serve to reduce oxygen levels in the larval midgut which is a necessary trigger for metamorphosis7. The midgut microbiota of larvae is diverse and thought to be derived primarily from the larval habitat2,4,5. Its composition is commonly dominated by Proteobacteria, while Firmicutes, Bacteroidetes, Actinobacteria, and Cyanobacteria are also frequently detected in larvae1,2,4,8.
Bacteria in larvae can be transstadially transmitted to adult midguts5, though the mechanism of this transmission is unclear. This has implications for vector borne disease prevention, as the midgut microbiota of adult female mosquitoes has been shown to influence vector competence for Plasmodium falciparum (the etiological agent of malaria) and dengue virus8,9. For example, multiple bacteria prevent infection by P. falciparum when present in the midgut of adult female An. gambiae, a primary mosquito vector of this parasite10,11,12. Additionally, introduction of specific bacterial species to the midgut of Ae. aegypti (the primary vector of dengue) has been shown to decrease12,13 or increase14 mosquito susceptibility to dengue virus. Investigating and potentially exploiting interactions between environmental microbes and mosquito larvae can provide a better understanding of vector competence and presents novel opportunities for reducing vector borne disease transmission.
In addition to the potential influence on vector competence and vectorial capacity, study of mosquito-microbe interactions has the potential to yield novel biocontrol agents. The facts that larvae are filter feeders, are relatively stationary (compared to adults), and are confined to their breeding water render them good targets for microbial biocontrol15. The gram-positive sporulating bacteria Bacillus thuringiensis var israelensis was isolated from natural mosquito breeding sites and has been used in mosquito population control efforts successfully for decades, although mosquito resistance has been reported16,17,18. This and other species from this genus are ingested by mosquito larvae and release toxins in the larval digestive tract16. These toxins are primarily pore-forming proteins that cause lysis of cells in the insect gut resulting in eventual death19,20,21. Continued investigation of mosquito-associated bacteria in field populations has the potential to yield additional biocontrol agents.
In previous work, we found that Chromobacterium sp. Panama (C.sp_P), a bacterium isolated from the midguts of Aedes aegypti in Panama13, has larvicidal activity against Anopheles gambiae and Ae. aegypti larvae12. Other species of Chromobacterium have been shown to have insecticidal properties, including C.vaccinii which causes mortality in A. aegypti larvae as well as diamondback moths22. C. subtsugae induces mortality in multiple insect species, though not against larval Culex pipiens mosquitoes, and is the main ingredient in the biopesticide Grandevo® (Marrone BioInnovations)23. The most extensively studied member of the genus, Chromobacterium violaceum, produces multiple metabolites with medicinal and industrial applications, including violacin, which has antimicrobial and antiparasitic activities, and hydrogen cyanide (HCN), which can be used for bioleaching of gold from electronic waste and ore24,25,26,27. HCN is produced by hydrogen cyanide synthase encoded by the hcnABC operon which is present in C. violaceum, C. vaccinii, and multiple Pseudomonad genomes22,25,28,29. HCN production during Pseudomonas infection has been shown to contribute to lethality in Caenorhabditis elegans and Drosophila melanogaster30,31. C. vaccinii also produces HCN in culture, though it remains unclear whether HCN production induces mortality in larvae treated with these bacteria22,29. In the present study, we aimed to further characterize the larvicidal activity of C.sp_P and to better understand the mechanism of larval killing by the bacteria.
Results
Concentration of C.sp_P and larval diet influence larvicidal activity
We investigated the conditions necessary to induce An. gambiae larval mortality by varying both the number of C.sp_P as well as the amount of larval food present in the larval water. To test the role of bacterial number, we added C.sp_P to larval breeding water at concentrations of 107, 105, 103, 101 and 10−1 Colony Forming Units (CFU) per 5 mL larval breeding water. To test the effect of larval food concentration, we conducted this experiment in water containing 3 mg ground larval food per 5 mL (“high nutrient,” Fig. 1b) and 1 mg ground larval food per 5 mL (“low nutrient,” Fig. 1b). We found that for both concentrations of larval food, adding 107 C.sp_P was the only treatment that induced significant mortality (Fig. 1). By day four post exposure, probability of survival in low nutrient and high nutrient conditions was 32.4% and 0%, respectively. In addition, we showed that larval mortality is not induced by treatment with another bacterium isolated from the mosquito midgut, Comamonas sp., when added to larval breeding water at similar concentrations as C.sp_P (Fig. S1).
Multiple species in the Chromobacterium genus induce larval mortality in An. gambiae
We assessed whether other species related to C.sp_P were also capable of inducing larvicidal activity. For these assays, we chose C. violaceum, the most well-studied member of the genus, which produces multiple secondary metabolites with bioactive properties24,25,26. We also chose C. subtsugae which is the active ingredient in the bioinsecticide Grandevo® (Marrone BioInnovations)23,32, and C. vaccinii which has documented insecticidal activity against Ae. aegypti larvae22, though the insecticidal mechanisms of C. subtsugae and C. vaccinii remain unclear. We also used C. aquaticum, which does not produce violacein and is more closely related to C.sp_P than the other three species12. We found significant larval mortality over four days after treatment of larval breeding water with C.sp_P (p = 2 × 10−16), C. aquaticum (p = 5.46 × 10−8), C. violaceum (p = 4.63 × 10−5), and C. vaccinii (p = 3.66 × 10−13), but not C. subtsugae (p = 0.846) relative to the PBS control (Fig. 2). Bacterial concentrations of cultures used to treat larvae were highly similar between C.sp_P, C. aquaticum, and C.vaccinii (4 × 108 CFU/mL on average), but higher for C. violaceum (8 × 108 CFU/mL on average) and C. subtsugae (1 × 109 CFU/mL on average) (Figure S2A). A regression analysis revealed no correlation between the CFUs added for each species and survival by day four (r2 = 0.01091, p = 0.2952, Figure S2B).
Larvicidal activity persists after removal of live C.sp_P bacteria
To determine whether live bacteria are necessary to induce larval mortality, we exposed An. gambiae larvae to C.sp_P-treated breeding water from which all live bacteria had been removed. To do this, we treated An. gambiae larval breeding water with C.sp_P until larval mortality was 100% and then passed the breeding water through a 0.2 µm filter. We then exposed additional An. gambiae larvae to the filtered breeding water and observed mortality after an overnight incubation. Larval mortality after exposure to filtered C.sp_P-treated breeding water was approximately 78%, while that in the PBS treated control was 3% (Fig. 3, p = 3.06 × 10−5).
We also investigated whether larvae incubated with C.sp_P could be contributing to the larvicidal activity in the filtered water (e.g. by releasing toxins after death). We repeated the experiment described above but excluded larvae from the initial breeding water. We observed significant mortality in the C.sp_P-treated water, similar to that seen when larvae were present in the initial incubation (Fig. S3, p = 0.00097). We were interested in determining the source of the larvicidal activity and began by first investigating the nature of the factor(s) causing larval mortality.
C.sp_P larvicidal factor(s) are less than 3 kDa, heat stable, and volatile
We size fractionated filtered (i.e. cell-free) C.sp_P- and PBS-treated larval water using centrifuge filtration to obtain three fractions: (1) molecules greater than 10 kDa, (2) molecules less than 10 kDa but more than 3 kDa, and (3) molecules less than 3 kDa. We then treated An. gambiae larvae with each fraction as well as the whole filtrate. We found that larvae treated with the whole C.sp_P filtrate and the <3 kDa fraction experienced similar levels of larval mortality which significantly differed from the respective PBS controls as well as the C.sp_P fractions containing larger molecules (Fig. 4a; C.sp_Pwhole vs. PBSwhole, p = 0.0011; C.sp_P<3kDa vs. PBS<3kDa, p = 0.0043).
We also investigated the heat sensitivity of the C.sp_P-produced larvicidal factor. We filter sterilized C.sp_P- and PBS-treated larval breeding water, incubated the water for one hour at room temperature (RT), 37 °C, 55 °C, 70 °C, or 90 °C, and then treated An. gambiae larvae with each sample and observed mortality. We found that exposure to C.sp_P-treated water resulted in significant larval mortality relative to the respective PBS control for all treatments (Fig. 4b; C.sp_PRT vs. PBSRT, p = 1.17 × 10−4; C.sp_P 37° vs. PBS37°, p = 0.00354; C.sp_P 55° vs. PBS55°, p = 8.28 × 10−7; C.sp_P 70° vs. PBS70°, p = 4.57 × 10−4; C.sp_P 90° vs. PBS90°, p = 4.39 × 10−9).
Finally, we investigated whether the larvicidal factor produced by C.sp_P is volatile. We centrifuged filtered C.sp_P- and PBS-treated larval breeding water in a vacuum centrifuge for 30 minutes to allow volatile compounds to evaporate. We then treated An. gambiae larvae with centrifuged and uncentrifuged water. Exposure to C.sp_P-treated water resulted in significant larval mortality in uncentrifuged samples (C.sp_P vs. PBS, p < 1.0 × 10−4), but not in centrifuged samples (Fig. 4c; C.sp_P VC vs. PBSVC, p = 0.972).
C.sp_P produces cyanide in breeding water at concentrations sufficient to induce larval mortality
Hydrogen cyanide, known to be produced by multiple Chromobacterium species22,33, is a volatile and heat stable small molecule34,35 and therefore could be the source of C.sp_P’s larvicidal activity. We first determined whether C.sp_P produces HCN in culture and the dynamics of its production during growth. We found that cyanide levels began to increase during the log phase of C.sp_P growth and reached approximately 0.6 mg/L by the beginning of stationary phase (Fig. 5a).
We next assessed the concentration of cyanide in the aquatic environment that would be necessary to cause larval mortality. We submerged An. gambiae larvae for 24 hours in a dilution series of NaCN solution, with concentrations ranging from 0 mg/L to 3.6 mg/L. We found that larval mortality was similar to the negative control (approximately 20% mortality) in the 0.23 mg/L treatment but increased to 73.3% in the 0.45 mg/L treatment and was 100% in the 0.9 mg/L treatment (Fig. 5b).
Additionally, we assessed the level of HCN produced by C.sp_P in larval breeding water and the extent to which its production correlates with larval mortality (Fig. 5c). At 4 h post exposure, average cyanide concentrations in larval breeding water were 0.063 mg/L and larval mortality was 0.7%. At 6.5 h post exposure, larval mortality was 45.7% and average cyanide concentrations 0.29 mg/L. By 7.5 h post exposure, larval mortality was 90.6% and average cyanide concentrations had increased to 0.513 mg/L. By 24 h and 48 h, average cyanide concentrations decreased to 0.055 mg/L and less than 0.01 mg/L, respectively.
Finally, we investigated whether HCN evaporates from C.sp_P-treated larval water when centrifuged under a vacuum. Vacuum centrifugation eliminated larvicidal activity (Fig. 4c); if HCN is causing larval mortality, we predicted it would no longer be present after vacuum centrifugation. We found that after vacuum centrifugation, nearly all HCN was lost, with HCN levels decreasing nearly ten-fold (Fig. 5d, C.sp_P HCN = 0.175 mg/L, C.sp_PVC, HCN = 0.02 mg/L).
Treatment with the cyanide antidote hydroxocobalamin eliminates C.sp_P larvicidal activity
We supplemented filtered (i.e. cell-free) C.sp_P-treated larval breeding water with hydroxocobalamin (OHCbl), which sequesters cyanide and is used as an antidote for cyanide poisoning36. We found significant mortality in larvae exposed to C.sp_P-treated water (Fig. 6, C.sp_P vs. PBS, p = 0.024), but no effect on larval survival when C.sp_P-treated water was supplemented with OHCbl (Fig. 6, C.sp_POHCbl vs PBSOHCbl, p = 0.834).
Surviving initial exposure to C.sp_P slows development and reduces survival over time
We were interested in determining whether there are persistent effects of non-lethal exposure to C.sp_P on mosquito survival and/or fitness. We exposed An. gambiae larvae to a moderate dose of C.sp_P (or PBS as a control) for two days. After two days, we recorded larval survival, which averaged 91.1% in PBS and 70.4% in C.sp_P, rinsed larvae from both treatments in clean distilled water and transferred larvae to clean trays containing additional distilled water. We then monitored pupation events, pupal death, and eclosion events until all pupae had either eclosed or died. We also recorded the total number of larvae that either failed to pupate by 11 days or died as larvae. We found that early exposure to C.sp_P significantly slowed eclosion (Fig. 7a, C.sp_P vs. PBS, p = 0.016). This effect was strongest at five days after transfer to clean water, and by day seven the probability of eclosion was similar between the two treatments. We also found that when larvae were treated with C.sp_P, by the end of the experiment a lower proportion had successfully pupated (Fig. 7b; overall percent pupation for C.sp_P = 53.1%, PBS = 74.6%), and of those that pupated, C.sp_P-treated individuals were less likely to eclose (Fig. 7b; overall percent of pupae that eclosed: C.sp_P 53.4% PBS, 79.2%).
Discussion
In the current work, we aimed to characterize the larvicidal activity of C.sp_P and determine potential mechanism(s) by which C.sp_P causes mosquito larval mortality when present in the larval breeding water. We found that larvicidal activity is retained after removal of live C.sp_P from breeding water, indicating that the bacteria are producing a factor or factors that act(s) independently of live bacteria. This activity is derived primarily from C.sp_P and not from the larval carcasses, as larvicidal activity was present in breeding water that contained C.sp_P and no larvae. The larvicidal factor(s) is/are less than 3 kDa in size, potentially suggesting a short peptide or small molecule. Activity persisted after heat treatment as high as 90 °C for one hour, and it was lost after vacuum centrifuging the breeding water for 30 minutes. We also found that C.sp_P produced hydrogen cyanide in larval breeding water at concentrations sufficient to result in larval mortality, and vacuum centrifugation of C.sp_P-treated larval breeding water resulted in near elimination of HCN from the water and a concurrent loss of larvicidal activity. Treatment with hydroxocobalamin (OHCbl), a cyanide antidote, also eliminated larvicidal activity from C.sp_P-treated breeding water. Taken together, these results indicate that larval mortality is at least in part a result of HCN accumulation in the larval breeding water, eventually reaching environmental levels that are toxic to larvae and then gradually evaporating over time. It is not clear from these data, however, that HCN is the only source of larvicidal activity produced by C.sp_P or what other factors may also be involved.
The fact that exposure to cell-free C.sp_P-treated water resulted in larval mortality is consistent with the hypothesis that HCN is a source of larvicidal activity, as HCN synthase is a membrane bound protein and cyanide accumulates in the supernatant during bacterial growth28,33. HCN has also been shown to be heat resistant above 90 °C and is volatile34,35, suggesting that it would persist in breeding water after heat treatment and would be lost after vacuum centrifugation. OHCbl acts by binding cyanide directly to form non-toxic cyanocobalamin, also known as vitamin B1236. Accordingly, we show that addition of an excess of OHCbl to the filtered C.sp_P-treated larval breeding water abolished larvicidal activity. OHCbl can bind a variety of other molecules besides HCN, including many small anionic molecules (e.g. nitric oxide and sulfite) as well as larger molecules like glutathione and other thiols37,38,39,40, so we cannot rule out the possibility that OHCbl may have bound other molecules that contribute to larval mortality. However, OHCbl has a high affinity for HCN37 suggesting it would readily bind cyanide in the larval water, and the results of this experiment as well as the additional findings in the manuscript are consistent with the hypothesis that HCN production by C.sp_P is a source of larval mortality.
HCN production by Pseuodmonas aeruginosa has been shown to cause mortality in Caenorhabditis elegans, and to also play a role in virulence against Drosophila melanogaster30,31. To our knowledge, this is the first evidence of bacteria-produced HCN causing mortality in a mosquito.
Larvicidal activity was persistently lower in filtered (i.e. bacteria-free) C.sp_P-treated water relative to larval water containing live C.sp_P. It is possible that when live C.sp_P are in the breeding water it contains more HCN over longer periods of time, whereas HCN is expected to be continually evaporating from filtered water during exposure, resulting in gradually reduced levels over time. It is also possible that live C.sp_P are ingested by the larvae, and that internal exposure to the bacteria contributes to mortality, either through HCN production or an alternative mechanism. We note that hypoxia has been reported in the larval gut, which is a necessary environmental trigger for HCN production7,41. A third possibility is that C.sp_P produces additional factors that are not maintained in filtered preparations that either act to compound the effects of HCN or to induce larvicidal activity through an unrelated mechanism. Further investigation into whether C.sp_P produces larvicidal compounds in addition to HCN is warranted.
We found that larval mortality induced by C.sp_P was dependent on larval diet concentration. In a “low nutrient” environment (1 mg diet/5 ml water) vs. “high nutrient” (3 mg diet/5 ml water), the larvicidal activity of C.sp_P was substantially reduced. C. violaceum, C. vaccinii, and several species in the genus Pseudomonas have been observed to produce HCN and this process is highly influenced by nutritional availability22,33,42,43,44. HCN is a secondary metabolite produced as a byproduct of glycine metabolism, and is therefore highly dependent on glycine availability33,45,46,47. While glycine is the precursor of HCN formation, other amino acids influence HCN production, such as threonine which can be converted to glycine, and methionine which synergistically enhances the effect of glycine availability on HCN formation33,45. HCN production is also influenced by environmental levels of iron. Increased concentrations of iron in growth media causes higher HCN production in C. violaceum and P. aeruginosa42,48. Given the multiple nutritional requirements for HCN production, the “low nutrient” condition may not have contained enough of one of the molecules mentioned above to support lethal levels of HCN production.
In Pseudomonas, HCN production has also been shown to be regulated by quorum sensing49 and oxygen availability; microaerophilic (but not anaerobic) conditions promote HCN production through regulation of oxygen sensitive transcription factor complexes28,41. HCN production is therefore highest in late log phase, after a period of rapid oxygen consumption and when high cell densities enable quorum sensing49. We found this to also be the case for C.sp_P; the highest concentration of HCN in culture was during late log phase. In larval breeding water, in a “high nutrient” environment over the course of four days, C.sp_P induced little mortality when added at 1 × 105 CFU/5 mL, but 100% mortality when added at 1 × 107 CFU/5 mL. HCN production would not be predicted to occur at low bacterial densities, and the fact that mortality did not occur by four days in the treatments with fewer initial bacteria suggests that the bacteria could not replicate enough in the breeding water to reach densities sufficient to trigger quorum sensing. In a “low nutrient” environment, even addition of 1 × 107 CFU/5 mL was not sufficient to induce 100% mortality. As discussed above, this may be due to a lack of nutrients essential for HCN production. Alternatively, it is possible that reduced nutrient availability may slow metabolic activity of C.sp_P, preventing bacterial replication and/or oxygen consumption from reaching levels high enough to induce HCN production.
We found that exposure to C. violaceum, C. aquaticum, and C. vaccinii also resulted in larval mortality, suggesting the effect is widespread throughout the Chromobacterium genus. C. subtsugae did not have any observable larvicidal activity, despite being added to larval breeding water at higher concentrations than the other species. These findings are consistent with previous work showing that C. vaccinii has larvicidal activity against A. aegypti but C. subtsuage does not against C. pipiens22,23. It has been previously shown that C. violaceum produces HCN in culture, as does C. vaccinii.22,29,33, and this may be the mechanism by which these species cause larval mortality. C. subtsugae does not produce HCN above background levels when cultured in King’s Medium B, suggesting that this species does not produce HCN even under optimal growth conditions22. It remains unclear why C. subtsugae fails to produce HCN, though we speculate that its lack of HCN production may underlie its inability to induce mortality in mosquito larvae.
Cyanide binds to cytochrome c oxidase, the final enzyme in the electron transport chain, and inhibits cellular respiration, leading to anoxia and cell death50. The nervous system is particularly susceptible to cyanide poisoning and collapse of nervous system function leads to subsequent respiratory system failure and death51. Because cyanide acts on such a common and critical cellular process, it is a broad-spectrum toxin, capable of inducing mortality in organisms ranging from bacteria to mammals. Cyanide’s mode of action may also greatly limit the mosquito’s ability to evolve resistance to the poison. Cyanide resistant respiration is not uncommon in microorganisms and plants, but it is rare among animals52,53,54. Notably, however, there is at least one instance of cyanide resistance arising in a scale insect population that was regularly treated with hydrogen cyanide55, and some species of herbivorous insects have evolved the ability to tolerate ingestion of cyanogenic plants56,57,58. Cyanide induces mortality in aquatic invertebrates at concentrations less than 100 µg/L59. In fish, mortality is induced at environmental cyanide concentrations as low as 24 µg/L in Atlantic salmon and as low as 57 µg/L in rainbow trout59,60. Cyanide is rapidly detoxified and does not bioaccumulate, but sub-lethal doses can have long-term effects on reproduction and development in freshwater fish51. In humans, sublethal doses through consumption of cyanogenic foods such as cassava can cause multiple health problems such as spastic paraparesis and thyroid disease61,62,63,64. The EPA’s recommended maximum contamination level for HCN in drinking water is 200 µg/L and according to the WHO a tolerable daily intake for adults is 12 µg/kg of body weight65,66. At peak HCN production, consumption of 1.5 L of C.sp_P-treated water by a 70 kg individual would exceed this recommended daily intake limit.
Sub-lethal exposure of An. gambaie larvae to C.sp_P results in significant delays in the rate of eclosion and reduced survival over time, with increased mortality observed in larval and pupal stages and a lower overall proportion of eclosing adults. HCN produced by C.sp_P during the sub-lethal exposure may cause delayed mortality, or it may be due to another factor produced by the bacteria during this time. Another, non-mutually exclusive explanation is that C.sp_P may colonize the larval midgut after initial exposure, and chronic internal exposure to the bacteria over time could lead to delayed mortality and slowed development time. Regardless of the mechanism, our study suggests that C.sp_P could have long-term negative effects on fitness of vector mosquitoes even when given at a sub-lethal dose.
In conclusion, we showed that C.sp_P is a cyanogenic bacterium that induces death in larval An. gambiae mosquitoes at least in part via production of hydrogen cyanide in larval breeding water. At sub-lethal doses, it also results in reduced fitness potential in An. gambiae. We also found that larvicidal activity is widespread throughout the Chromobacterium genus; additional study of the genus may reveal whether this is due to cyanogenesis or another mechanism. Because HCN is a general toxin affecting oxidative respiration, the potential of C.sp_P as a biocontrol agent may be limited. However, further work to develop methods to avoid off-target effects and target mosquito larvae specifically (e.g. through use of larval attractants or through treatment of breeding sites) is warranted.
Methods
Ethics statement
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Animal Care and Use Committee of the Johns Hopkins University (permit number MO15H144).
Mosquito strain maintenance
Anopheles gambiae Keele strain mosquitoes were reared at 27 °C and 80% residual humidity with a 14/10 hour light/dark cycle. Adult females were fed on anaesthetized mice for egg collection, larvae were reared on TetraMin® fish food and cat food pellets, and adults were provided 10% sucrose ad libitum.
Bacterial culture and growth curve measurements
All bacterial stocks were maintained in 25% glycerol stocks at −80 °C. To generate these stocks, liquid LB media was inoculated with a single bacterial colony and grown overnight at 30 °C with constant shaking. Cultures were pelleted and washed two times with 1× PBS and then mixed 1:1 with sterile 50% glycerol solution and stored at −80 °C. To culture Chromobacterium species Panama (C. sp_P), Comamonas sp.10, C. violaceum (ATCC 12472)25, C. aquaticum (DSM 19852)67, C. vaccinii (DSM 25150)68, and C. subtsugae (DSM 17043)23 for experiments, 1 µL of pure bacterial freezer stock was added to 5 mL LB and grown at 30 °C with shaking for 16–18 hours. For most larval survival experiments, overnight liquid culture was washed twice with sterile 1X PBS and resuspended in additional 1X PBS to 1.0 (±0.1) OD600. For the experiment testing multiple species of Chromobacterium in parallel, overnight liquid cultures were washed twice with 1X PBS. C. subtsugae overnight cultures routinely have lower CFU/mL than the other species, therefore C. subtsugae was further concentrated 5X and the other cultures were used without concentration. This step was performed in an attempt to standardize CFU/mL within an order of magnitude for all species. 100uL of each culture were then added to 5 mL larval breeding water, resulting in an approximate bacterial density of 107 CFU/5 mL larval water for all species except C. subtsugae, for which it was approximately 108 CFU/5 mL larval water. Liquid cultures were serially diluted and grown on LB agar to estimate CFU/mL at the time of treatment. For growth curve measurements, 10 mL of an overnight C.sp_P culture was used to inoculate 1 L liquid LB in triplicate. Cultures were then grown for eight hours at 30 °C with shaking and OD600 was measured every 1–1.5 hours during the incubation.
Larval survival after exposure to live C.sp_P or C.sp_P-treated breeding water
Larvae for all experiments were a mix of 1st and 2nd instars. For experiments using live bacteria, ten An. gambiae larvae were placed in individual wells of a 6-well cell culture dish containing 5 mL water and approximately 3 mg ground larval food (unless stated otherwise). Ground larval food was used for all larval experiments, and consisted of liver powder, tropical fish flake food, and rabbit food pellets mixed in a 2:1:1 ratio. 100uL of prepared bacterial culture was added to the larval water and survival was monitored for four days. For the experiment measuring larval survival and HCN production in concert, ten An. gambiae larvae were placed in 18 separate 150 mm petri dishes containing 75 mL water and approximately 45 mg ground larval food. To three petri dishes, 500 µL 1X PBS was added. To fifteen petri dishes, 500 µL 1.0 OD600 C.sp_P bacterial culture was added. At 4, 6.5, 7.5, 24, and 48 hours after treatment, all 75 mL was harvested from three of the fifteen C.sp_P-treated petri dishes and used for HCN analysis. At 3, 4, 5, 6, 6.5, 7.5, 24, and 48 hours after treatment, survival from all remaining petri dishes was recorded.
Larval survival after exposure to filtered (i.e. cell-free) C.sp_P-treated water
Larvae were treated with C.sp_P or PBS as described above. After death (6–8 hours post exposure), larval water from PBS and C.sp_P -treated wells was vortexed, centrifuged at 5000 rpm for 3 minutes, and passed through a 0.2 µm syringe filter, which removes live bacterial cells. Filtrate sterility was confirmed in two ways: in an initial experiment, filtrate was spread on LB agar and was confirmed to yield zero colonies after multiple days of growth. In subsequent experiments, filtrate was left at room temperature for multiple days and was confirmed to never become turbid after filtration. Any additional treatments to the water were performed at this point (e.g. size-fractionation, heating, vacuum centrifugation) and 1 mL aliquots of the filtered water were then used to treat pools of 5 An. gambiae larvae (1st and 2nd instars) in 24-well cell culture dishes. No food was added, and survival was recorded after an overnight (16–17 hours) or 24-hour incubation. When heating the filtered water, 1 mL aliquots were placed in a heat block in sealed 1.5 mL microcentrifuge tubes for one hour. Samples were mixed by hand and allowed to return to room temperature before use.
Size fractionation of filtered larval breeding water
To size-fractionate filtered PBS and C.sp_P-treated larval water, an equal volume of each was loaded onto an Amicon Ultra-4 Centrifugal Filter Unit with Ultracel-10 membrane and spun at 7917 × g for 15 minutes. The filtrate was then loaded onto an Amicon Ultra-4 Centrifugal Filter Unit with Ultracel-3 membrane and spun at 7917 × g for 25 minutes. The concentrate from the Ultracel-10 filtration unit contains molecules greater than 10 kDa, while the concentrate from the Ultracel-3 filtration unit contains molecules less than 10 kDa but greater than 3 kDa. The final filtrate contains molecules less than 3 kDa. Sterile water was added to both concentrates to raise their volumes to equal that of the final filtrate.
Vacuum centrifugation of filtered larval breeding water
C.sp_P- and PBS-treated larval water that had been previously isolated (for size fractionation experiments and stored at 4 °C) was used for survival analysis after vacuum centrifugation. C.sp_P-treated water used for HCN quantification after vacuum centrifugation was generated fresh as described above. Aliquots of these water samples were vacuum centrifuged in open 1.5 mL microcentrifuge tubes for 30 minutes at room temperature. Control aliquots were incubated at room temperature in sealed microcentrifuge tubes during centrifugation. Sterile water was added to centrifuged samples to compensate for loss of volume.
Hydroxocobalamin treatment of filtered larval breeding water
C.sp_P- and PBS-treated larval water used for this experiment was isolated and filtered on two independent days as described above. Larvae were treated with these water samples twice, once on the day they were isolated and then on a later date. In the time between treatment days, water samples were stored at 4 °C. 1 mL of filtered C.sp_P- or PBS-treated water was applied to pools of 5 An. gambiae larvae, and to half the pools from each treatment, filter sterilized OHCbl in water was added to the larval water at a final concentration of 0.2 mg/mL. Filter sterilized water was added to the remaining pools as a control.
Treatment of larvae with NaCN
An 18 mg/mL stock solution of NaCN dissolved in water was serially diluted 1:5, 1:10, 1:20, 1:40, and 1:80, resulting in NaCN concentrations of 3.6, 1.8, 0.9, 0.45, and 0.23 mg/mL, respectively. One milliliter of each dilution (plus water as a negative control) was applied to five An. gambiae larvae (a mixture of 1st and 2nd instars) in three wells of a 24-well cell culture plate. No larval food was added. After 24 hours, mortality in each well was recorded.
Hydrogen cyanide measurements
For experiments measuring HCN concentration in bacterial culture, in NaCN solution, and after vacuum centrifugation, HCN was measured using the Hach Cyanide Test Kit CYN-3 (Lot A5316) according to the manufacturer’s instructions. For the experiment measuring larval survival and HCN produced by C.sp_P in larval breeding water over time, EPA Method 335.4 was used. This method utilizes semi-automated colorimetry to assess total cyanide and was performed by SGS Accutest (Dayton, NJ). Both methods were used to measure HCN concentration in the same NaCN stock, and the result using the Hach Cyanide Test Kit was 30.75 mg/L, while that from semi-automated colorimetry was 18 mg/L. Semi-automated colorimetry is more sensitive and is less subject to user error than the Hach Cyanide Test Kit, and the different results from the two methods indicate that the Hach test kit overestimated cyanide concentrations by a factor of 1.7083. The following correction was therefore applied to all data obtained using the Hach Kit: y = x * 1/1.7083 where \(y\) = corrected HCN value and \(x\) = original HCN value.
Larval development and survival after non-lethal exposure to C.sp_P
A mix of first and second instar An. gambiae larvae were treated with C.sp_P or 1X PBS (negative control) in groups of 350 in 175 mL of water containing 105 mg ground larval food. C.sp_P overnight liquid culture was washed 2X with 1X PBS and approximately 3.6 × 106 CFU (rep 1: 2.4 × 106, rep 2: 4.8 × 106) of C.sp_P were added to the larval breeding water. After a two-day exposure, larval survival was recorded and C.sp_P- and PBS-treated larvae were rinsed twice with distilled water and transferred to large larval pans containing 2 L distilled water supplemented with 2 pellets of cat food and 15 mg of TetraMin® fish food. Every 2–3 days, an additional 15 mg of fish food was added. Pupation events were recorded daily beginning 3 days post exposure and continuing until all pupae eclosed or died. Larval cadavers were removed daily from larval pans and the total number of larvae that either died as larvae or were still alive (but never pupated) was recorded at the conclusion of the experiment.
Statistical analysis
Survival over time was assessed using Cox proportional hazards models with treatment, experimental replicate, and replicate pool of larvae as predictor variables in the model. Models were fit using “coxph” in the Survival package in R69. Survival at a single time point was analyzed using binomial logistic regression with treatment(s), experimental replicate, replicate pool of larvae, and date of larval water harvest as predictor variables where appropriate. Summary results and p-values from each model can be found in Table S1. Binomial logistic regressions were performed in R using “glm” in the stats package, and pairwise comparisons were conducted using “glht” in the multcomp package. Rate of eclosion was assessed using a log rank test in R implemented with “survdiff” in the survival package.
Data Availability Statement
Raw data for all experiments can be found in the Supplementary Dataset.
References
Wang, Y., Gilbreath, T. M., Kukutla, P., Yan, G. & Xu, J. Dynamic gut microbiome across life history of the malaria mosquito Anopheles gambiae in Kenya. PLoS One 6, e24767 (2011).
Gimonneau, G. et al. Composition of Anopheles coluzzii and Anopheles gambiae microbiota from larval to adult stages. Infect. Genet. Evol. 28, 715–724 (2014).
Akorli, J. et al. Seasonality and locality affect the diversity of Anopheles gambiae and Anopheles coluzzii midgut microbiota from Ghana. PLoS One 11, e0157529 (2016).
Coon, K. L., Brown, M. R. & Strand, M. R. Mosquitoes host communities of bacteria that are essential for development but vary greatly between local habitats. Mol. Ecol. 25, 5806–5826 (2016).
Coon, K. L., Vogel, K. J., Brown, M. R. & Strand, M. R. Mosquitoes rely on their gut microbiota for development. Mol. Ecol. 23, 2727–39 (2014).
Merritt, R. W., Dadd, R. H. & Walker, E. D. Feeding behavior, natural food, and nutritional relationships of larval mosquitoes. Annu. Rev. Entomol. 37, 349–376 (1992).
Coon, K. L. et al. Bacteria-mediated hypoxia functions as a signal for mosquito development. Proc. Natl. Acad. Sci. USA 114, E5362–E5369 (2017).
Minard, G., Mavingui, P. & Moro, C. V. Diversity and function of bacterial microbiota in the mosquito holobiont. Parasit. Vectors 6, 146 (2013).
Gendrin, M. & Christophides, G. K. In Anopheles mosquitoes - New insights into malaria vectors (ed. Manguin, S.) 525–548 (InTech, 2013). doi:10.5772/3392.
Cirimotich, C. M. et al. Natural microbe-mediated refractoriness to Plasmodium infection in Anopheles gambiae. Science 332, 855–8 (2011).
Bahia, A. C. et al. Exploring Anopheles gut bacteria for Plasmodium blocking activity. Environ. Microbiol. https://doi.org/10.1111/1462-2920.12381 (2014).
Ramirez, J. L. et al. Chromobacterium Csp_P Reduces Malaria and Dengue Infection in Vector Mosquitoes and Has Entomopathogenic and In Vitro Anti-pathogen Activities. PLoS Pathog. 10, e1004398 (2014).
Ramirez, J. L. et al. Reciprocal tripartite interactions between the Aedes aegypti midgut microbiota, innate immune system and dengue virus influences vector competence. PLoS Negl. Trop. Dis. 6, e1561 (2012).
Apte-Deshpande, A., Paingankar, M., Gokhale, M. D. & Deobagkar, D. N. Serratia odorifera a midgut inhabitant of Aedes aegypti mosquito enhances its susceptibility to dengue-2 virus. PLoS One 7, e40401 (2012).
Floore, T. G. Mosquito larval control practices: past and present. J. Am. Mosq. Control Assoc. 22, 527–533 (2006).
Lacey, L. A. Bacillus Thuringiensis Serovariety Israelensis and Bacillus Sphaericus for Mosquito Control. J. Am. Mosq. Control Assoc. 23, 133–163 (2007).
Goldberg, L. & Margalit, J. A Bacterial Spore Demonstrating Rapid Larvicidal Activity Against Anopheles sergentii, Uranotaenia unguiculata, Culex univitattus, Aedes aegypti and Culex pipiens. Mosq. News 37, 355–358 (1977).
Paul, A., Harrington, L. C., Zhang, L. & Scott, J. G. Insecticide resistance in Culex pipiens from New York. J Am Mosq Cont Assoc 21, 305–309 (2005).
Jones, G. W. et al. A new Cry toxin with a unique two-component dependency from Bacillus sphaericus. FASEB J. 21, 4112–4120 (2007).
Carroll, J. & Ellar, D. J. Analysis of the large aqueous pores produced by a Bacillus thuringiensis protein insecticide in Manduca sexta midgut-brush-border-membrane vesicles. Eur. J. Biochem. 245, 797–804 (1997).
Zhang, Q., Hua, G. & Adang, M. J. Effects and mechanisms of Bacillus thuringiensis crystal toxins for mosquito larvae. Insect Sci. 24, 714–729 (2017).
Martin, P. A. & Soby, S. Insecticidal strains of Chromobacterium vaccinii sp. nov. for control of insects (2016).
Martin, P. A. W., Gundersen-Rindal, D., Blackburn, M. & Buyer, J. Chromobacterium subtsugae sp. nov., a betaproteobacterium toxic to Colorado potato beetle and other insect pests. Int. J. Syst. Evol. Microbiol. 57, 993–999 (2007).
Liu, R., Li, J. & Ge, Z. Review on Chromobacterium Violaceum for Gold Bioleaching from E-waste. Procedia Environ. Sci. 31, 947–953 (2016).
Brazilian National Genome Project Consortium. The complete genome sequence of Chromobacterium violaceum reveals remarkable and exploitable bacterial adaptability. Proc. Natl. Acad. Sci. USA 100, 11660–5 (2003).
Lopes, S. C. P. et al. Violacein extracted from Chromobacterium violaceum inhibits Plasmodium growth in vitro and in vivo. Antimicrob. Agents Chemother. 53, 2149–52 (2009).
Shin, D., Jeong, J., Lee, S., Pandey, B. D. & Lee, J. C. Evaluation of bioleaching factors on gold recovery from ore by cyanide-producing bacteria. Miner. Eng. 48, 20–24 (2013).
Laville, J. et al. Characterization of the hcnABC gene cluster encoding hydrogen cyanide synthase and anaerobic regulation by ANR in the strictly aerobic biocontrol agent Pseudomonas fluorescens CHA0. J. Bacteriol. 180, 3187–96 (1998).
Vöing, K., Harrison, A. & Soby, S. D. Draft Genome Sequences of Three Chromobacterium subtsugae Isolates from Wild and Cultivated Cranberry Bogs in Southeastern Massachusetts. 3, 1–2 (2015).
Broderick, K. E. et al. Cyanide produced by human isolates of Pseudomonas aeruginosa contributes to lethality in Drosophila melanogaster. J. Infect. Dis. 197, 457–64 (2008).
Gallagher, L. A. & Manoil, C. Pseudomonas aeruginosa PAO1 kills Caenorhabditis elegans by cyanide poisoning. J. Bacteriol. 183, 6207–14 (2001).
Blackburn, M. B., Sparks, M. E. & Gundersen-Rindal, D. E. The genome of the insecticidal Chromobacterium subtsugae PRAA4-1 and its comparison with that of Chromobacterium violaceum ATCC 12472. Genomics data 10, 1–3 (2016).
Michaels, R. & Corpe, W. A. Cyanide Formation by Chromobacterium violaceum. J. Bacteriol. 89, 106–12 (1965).
Gause, E. H. & Montgomery, P. D. Hydrogen Cyanide Stability and Heat of Polymerization. J. Chem. Eng. Data 5, 351–354 (1960).
Blumer, C. & Haas, D. Mechanism, regulation, and ecological role of bacterial cyanide biosynthesis. Arch. Microbiol. 173, 170–177 (2000).
Hall, A. H., Isom, G. E. & Rockwood, G. A. Toxicology of Cyanides and Cyanogens: Experimental, Applied and Clinical Aspects. (John Wiley & Sons, Incorporated, 2015).
Firth, R. A., Hill, H. A. O., Pratt, J. M., Thorp, R. G. & Williams, R. J. P. The Chemistry of Vitamin B12. Part XI. Some Further Formation Constants. J. Chem. Soc. A 381–386 (1969).
Adler, N., Medwick, T. & Poznanski, T. J. Reaction of Hydroxocobalamin with Thiols. J. Am. Chem. Soc. 88, 5018–5020 (1966).
Marques, H. M. Activation parameters for the reaction of aquocobalamin (Vitamin B12a) with small anionic and neutral ligands. J. Chem. Soc. Dalt. Trans. 339–341, https://doi.org/10.1039/DT9910000339 (1991).
Wolak, M., Zahl, A., Schneppensieper, T., Stochel, G. & Van Eldik, R. Kinetics and mechanism of the reversible binding of nitric oxide to reduced cobalamin B12r (Cob(II)alamin. J. Am. Chem. Soc. 123, 9780–9791 (2001).
Castric, P. Influence of Oxygen on the Pseudomonas aeruginosa Hydrogen Cyanide Synthase. Curr. Microbiol. 29, 19–21 (1994).
Castric, P. A. Hydrogen cyanide, a secondary metabolite of Pseudomonas aeruginosa. Can. J. Microbiol. 21, 613–18 (1975).
Ryall, B., Mitchell, H., Mossialos, D. & Williams, H. D. Cyanogenesis by the entomopathogenic bacterium Pseudomonas entomophila. Lett. Appl. Microbiol. 49, 131–135 (2009).
Askeland, R. A. & Morrison, S. M. Cyanide production by Pseudomonas fluorescens and Pseudomonas aeruginosa. Appl. Environ. Microbiol. 45, 1802–7 (1983).
Castric, P. A. Glycine metabolism by Pseudomonas aeruginosa: hydrogen cyanide biosynthesis. J. Bacteriol. 130, 826–31 (1977).
Brysk, M. M., Lauinger, C. & Ressler, C. Biosynthesis of cyanide from [2-14C15N]glycine in Chromobacterium violaceum. Biochim. Biophys. Acta 184, 583–588 (1969).
Wissing, F. Cyanide formation from oxidation of glycine by a Pseudomonas species. J. Bacteriol. 117, 1289–94 (1974).
Rodgers, P. B. & Knowles, C. J. Cyanide Production and Degradation During Growth of Chromobacterium violaceum. J. Gen. Microbiol. 108, 261–267 (1978).
Pessi, G. & Haas, D. Transcriptional control of the hydrogen cyanide biosynthetic genes hcnABC by the anaerobic regulator ANR and the quorum-sensing regulators LasR and RhlR in Pseudomonas aeruginosa. J. Bacteriol. 182, 6940–6949 (2000).
Egekeze, J. O. & Oehme, F. W. Cyanides and their toxicity: a literature review. Vet. Q. 2, 104–114 (1980).
Eisler, R. Cyanide Hazards to Fish, Wildlife, and Invertabrates: A Synoptic Review. U.S. Fish Wildl. Serv., Biol. Rep. 85, 1–65 (1991).
Solomos, T. Cyanide-Resistant Respiration in Higher Plants. Annu. Rev. Plant Physiol. 28, 279–297 (1977).
Knowles, C. J. Microorganisms and cyanide. Bacteriol. Rev. 40, 652–80 (1976).
Veiga, A., Arrabaca, J. D. & Loureiro-Dias, M. C. Cyanide-resistant respiration, a very frequent metabolic pathway in yeasts. FEMS Yeast Res. 3, 239–245 (2003).
Dickson, R. C. Inheritance of resistance to hydrocyanic acid fumigation in the California red scale. Hilgardia 13, 515–522 (1941).
Van Ohlen, M., Herfurth, A. M., Kerbstadt, H. & Wittstock, U. Cyanide detoxification in an insect herbivore: Molecular identification of β-cyanoalanine synthases from Pieris rapae. Insect Biochem. Mol. Biol. 70, 99–110 (2016).
Meyers, D. M. & Ahmad, S. Link between l-3-cyanoalanine synthase activity and differential cyanide sensitivity of insects. Biochim. Biophys. Acta 1075, 195–197 (1991).
Wybouw, N. et al. A gene horizontally transferred from bacteria protects arthropods from host plant cyanide poisoning. Elife 3, e02365 (2014).
Smith, L. L. et al. Acute and chronic toxicity of HCN to fish and invertebrates. U.S. Environ. Prot. Agency (1979).
Alabaster, J. S., Shurben, D. G. & Mallett, M. J. The acute lethal toxicity of mixtures of cyanide and ammonia to smolts of salmon, Salmo salar L. at low concentrations of dissolved oxygen. J. Fish Biol. 22, 215–222 (1983).
Teles, F. F. F. Chronic poisoning by hydrogen cyanide in cassava and its prevention in Africa and Latin America. Food Nutr. Bull. 23, 407–412 (2002).
Adamolekun, B. Etiology of Konzo, epidemic spastic paraparesis associated with cyanogenic glycosides in cassava: Role of thiamine deficiency? J. Neurol. Sci. 296, 30–33 (2010).
Ermans, A. M., Mbulamoko, N. M., Delange, F. & Ahluwalia, R. Role of cassava in the etiology of endemic goitre and cretinism. (International Development Research Centre, 1980).
Casadei, E., Jansen, P., Rodrigues, A., Molin, A. & Rosling, H. Mantakassa: an epidemic of spastic paraparesis associated with chronic cyanide intoxication in a cassava staple area of Mozambique. 2. Nutritional factors and hydrocyanic acid content of cassava products. Bull. World Health Organ. 62, 485–492 (1984).
U.S. Department of Health and Human Services. Toxicological Profile for Cyanide. https://doi.org/10.1201/9781420061888_ch68 (2006).
World Health Organization. Cyanide in Drinking-water. doi:WHO/SDE/WSH/03.04/05 (2003).
Young, C. C. et al. Chromobacterium aquaticum sp. nov., isolated from spring water samples. Int. J. Syst. Evol. Microbiol. 58, 877–880 (2008).
Soby, S. D., Gadagkar, S. R., Contreras, C. & Caruso, F. L. Chromobacterium vaccinii sp. nov., isolated from native and cultivated cranberry (Vaccinium macrocarpon Ait.) bogs and irrigation ponds. Int. J. Syst. Evol. Microbiol. 63, 1840–1846 (2013).
R Core Team (2013). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. www.R-project.org/.
Acknowledgements
We wish to thank the insectary personnel at the Johns Hopkins Malaria Research Institute for assistance with mosquito rearing, and the Bloomberg Philanthropies for their support. This work was funded by the National Institutes of Health, National Institute for Allergy and Infectious Disease, grants R01AI101431 and R01 AI081877. SMS was supported by a Ruth L. Kirschstein National Research Service Award F32AI112208-01A1. HJM was supported by a Johns Hopkins Malaria Research Institute predoctoral fellowship.
Author information
Authors and Affiliations
Contributions
S.M.S. designed and conducted experiments, analyzed data, and wrote the manuscript. S.V.T. designed and conducted experiments, analyzed data, and wrote the manuscript. H.J.M. conducted experiments and wrote the manuscript, G.D. designed experiments and wrote the manuscript.
Corresponding author
Ethics declarations
Competing Interests
GD and SMS have filed a patent application related to this work, Publication number US20170280730 A1, “Compositions and Methods useful for the prevention of malaria and dengue virus transmission”.
Additional information
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Short, S.M., van Tol, S., MacLeod, H.J. et al. Hydrogen cyanide produced by the soil bacterium Chromobacterium sp. Panama contributes to mortality in Anopheles gambiae mosquito larvae. Sci Rep 8, 8358 (2018). https://doi.org/10.1038/s41598-018-26680-2
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-018-26680-2
- Springer Nature Limited
This article is cited by
-
Contemporary exploitation of natural products for arthropod-borne pathogen transmission-blocking interventions
Parasites & Vectors (2022)
-
Discovery of novel natural products for mosquito control
Parasites & Vectors (2022)
-
Disrupting quorum sensing alters social interactions in Chromobacterium violaceum
npj Biofilms and Microbiomes (2021)