Recent updates in Trends in Parasitology [1] and Trends in Ecology and Evolution 2 R.H. ffrench-Constant, Something old, something transgenic, or something fungal for mosquito control?, Trends Ecol. Evol. 20 (2005), pp. 577-579. Article | PDF (122 K) | View Record in Scopus | Cited By in Scopus (3)[2] highlighted the back-to-back articles in Science 3 and 4 that demonstrated the potential biocontrol of malaria by targeting mosquitoes with entomopathogenic fungi (Metarhizium and Beauveria spp.). The wide impact of the original articles and the need to find alternatives to pesticidal control are likely to encourage the incorporation of these fungi into biocontrol programs, although several concerns have been raised 1, 2 and 5. Here, we detail some of these and advocate an inclusive approach to malarial biocontrol that proceeds with a full appreciation of the complicated biology of the pathogenic fungi concerned.

There is a long history of using Metarhizium and Beauveria in insect biocontrol [6] with ‘Green Muscle', a broad consortium set up to control locusts using Metarhizium, as an excellent example (for more information, see https://www.sciencedirect.com/science?_ob=RedirectURL&_method=externObjLink&_locator=url&_cdi=6081&_plusSign=%2B&_targetURL=http%253A%252F%252Fwww.lubilosa.org%252F). Green Muscle developed from the initial conception that spores could be suspended in oil to facilitate germination in arid regions [7], which made infecting mosquitoes feasible [4]. However, failures in biocontrol have been more common than successes and are often due to the ‘ephemeral and amateurish activities' [6] of biologists failing to appreciate the complexity of the task at hand. Such damning assessments by mycologists are not new (Petch, 1925 cited in [6]) and reflect, in particular, a lack of attention to the basic biology of the fungal biocontrol agents (Harry Evans, pers. commun.).

The Science papers did pay attention to the biology of Metarhizium [8], which is understandable given that a senior author of one (Matt Thomas) participated in Green Muscle [9], which should now, ideally, serve as model for developing a commercial mycopesticide against Anopheles mosquitoes. However, the Green Muscle programme took ten years to develop a product and had a total price tag of US$15 million [10], thus it could tempting to be less rigorous this time around. Given the sensitive political situation in many areas where malaria is endemic, humanity cannot afford shortcuts, because any failures owing to poor management or premature implementation will reduce local governmental support rather than enhance it (Andrew Read, pers. commun.). Therefore, if we are to ‘muscle out malaria', well-coordinated interdisciplinary approaches are necessary and basic mycological research will have to be a key component of these joint efforts.

Both in the original papers, popular media and the Trends updates, much was made of an apparent lack of resistance. However, it is premature to award ‘silver bullet' status to these pathogens, as this notion is based mainly on studies in temperate and depauperate agricultural ecosystems. Recent experiments showed that significant variation for resistance against Metarhizium anisopliae exists in leafcutter ants that are not known to suffer much from this pathogen in the field [11]. In addition, suggestive examples of behavioural resistance [12] add to these concerns because avoidance of fungal-treated sites by locusts occurred during the development of Green Muscle and is apparently strain specific (Harry Evans, pers. commun.). Knowing how insects recognise fungal spores is clearly of key importance, and the new focus on fungal biocontrol of malaria should therefore act as a catalyst for further research on the basic biology of fungal pathogens.

Understanding morphological, biochemical or immune system-based resistance to insect pathogenic fungi will be easier if we know their biology in ancestral habitats. Metarhizium and Beauveria spp. evolved in Southeast Asian rainforests, escaping to become specialised associates of agricultural systems^*. Insects killed by these two fungal diseases are extremely rare in tropical rainforests, but when they are found they display host specificity (Nigel Hywel-Jones, unpublished data); consistent with a large variety of fungal strains that can be isolated from tropical soil samples [13]. Again, the contrast with ‘common knowledge' from temperate agricultural systems is striking, which underscores the need for coordinated biodiversity surveying as part of interdisciplinary programs, comparable to expeditions organised by the British Mycological Society (https://www.sciencedirect.com/science?_ob=RedirectURL&_method=externObjLink&_locator=url&_cdi=6081&_plusSign=%2B&_targetURL=http%253A%252F%252Fwww.britmycolsoc.org.uk%252F [14]).

Finally, both Metarhizium and Beauveria are ‘anamorphic' asexual states of fungi that are confusingly known as Cordyceps when they express their ‘teleomorphic' sexual state. Cordyceps is common in rainforests and might regulate insect populations [6]. The nature of sexual spore transmission in rainforest environments means that behavioural avoidance by insects is more difficult. Therefore, immune system-based resistance in insects might have evolved towards Cordyceps and, given that Metarhizium and Beauveria are genetically identical to Cordyceps, this would be important when considering the potential evolution of resistance in mosquitoes.

In conclusion, we agree that insect pathogenic fungi such Metarhizium could well become important for the future control of malaria, but the road towards muscling out malaria in a sustainable way will be narrow, winding and littered with obstacles that only integrated research with major mycological components can remove [8].

Acknowledgements

We thank Andrew Read, Matthew Thomas and our mycologist colleagues Jorgen Eilenberg, Harry Evans, Nigel Hywel-Jones and Thomas Læssoe for valuable input and discussion.