Functional genomics of mosquito vector/malaria parasite interactions
The Laboratory Insect Immunogenomics focuses on the study of the major vector of malaria in Africa, Anopheles gambiae and particularly how its innate immune system manipulates and is manipulated by the malaria parasite during its passage through the mosquito. Malaria is one of the main infectious causes of human mortality world-wide, mainly among young children in sub-Saharan Africa. Transmission of the malaria agent, the Plasmodium parasite, requires its cyclical development in two organisms: the human host and the Anopheles mosquito vector. Current disease control methods that aim to either cure the disease in the human body or to control the vector populations are hampered due to increase in drug resistance of the parasite and insecticide resistance of the mosquito. Our underlying conviction is that bringing the power of functional genomics into the study of this biological system will rapidly advance our understanding of mosquito immunity and parasite development. This will be crucial for the development of novel approaches to malaria control that are urgently needed to reinforce and complement the ongoing research into drug, vaccine and insecticide development.
Completion of the genome sequence of A. gambiae two years ago, together with the development of DNA microarrays in this species and adaptation of the RNAi technique to adult mosquitoes, has allowed comparative and functional genomic approaches to understanding the mosquito innate immune system, and its interactions with parasites. Using the rodent model system, P. berghei, we have identified a variety of factors that negatively affect the development of parasites in the mosquito (antagonists), in some cases leading to complete transmission blockage. In addition, mosquito molecules have been identified that play positive roles and are required for successful parasite transmission (agonists). Research is continuing to identify new factors involved in these interactions and to decipher the interplay of these molecules and their regulation. Importantly, as our findings indicate a highly complex interplay between parasite and vector, we are currently extending our studies of parasite-vector interactions towards the human malaria parasite, P. falciparum.
To investigate the genetic armory of the A. gambiae mosquito we have initially used an EST library derived from two immune competent mosquito cell lines to construct DNA microarrays containing approximately 2,500 genes . These arrays permitted for the first time an understanding of the global innate immune responses in adult mosquitoes and cultured cells, as well as the mosquito reaction to Plasmodium infection and mechanisms for refractoriness to the parasite. To increase further the resolution of this powerful technique, we recently constructed DNA microarrays containing approximately 20,000 Anopheles ESTs representing over 9,000 mosquito genes, and work is ongoing to develop an amplicon-based microarray platform encompassing all predicted genes in the mosquito genome. This resource also allows for high throughput production of dsRNAs that can be used for RNAi gene silencing as well as expression of fragments of all predicted mosquito proteins, for use in antibody production.
In order to achieve full annotation of this new microarray platform, enhancing the utilityof the microarray data, we developed an EST informatics resource named AnoEST. This resource provides analysis of over 200,000 A. gambiae EST and cDNA sequences available to date in public sequence repositories. AnoEST aligns all ESTs on the mosquito genome, grouping them into expressed genomic loci that are assigned comprehensive functional annotation. Thus, it also allows the analysis of alternative splicing events, refinement of gene models, and prediction of additional genes that were missed by automatic gene prediction pipelines. AnoEST is publicly available and accessible as a Distributed Annotation Service (DAS). To facilitate information exchange and promote availability of the produced genomic data for the scientific community working on invertebrate vectors of human pathogens, we participate in an international collaborative project to develop the VectorBase. This is an integrated federation of databases and associated bioinformatics analysis resources, among them AnoEST, which will host and manage upcoming data from genome projects and functional genomics studies, including data from gene and protein expression and RNAi gene silencing.
The development of these functional genomic and bioinformatic tools has set the basis for a more complete understanding of the biology of the mosquito. Using the 20K DNA microarrays, we explored the genome transcriptional programmes during the mosquito’s life cycle and identified adult tissue-specific transcripts.
Currently, we are performing a comparative transcriptome analysis of the life stages of Anopheles and Drosophila, which diverged some 250 million years ago. These arrays are also utilized to understand the mosquito immune responses, responses against the malaria parasite, and immune transcriptional networks. In collaboration with other laboratories, we use these microarrays to study additional important aspects of vector biology such as insecticide resistance and responses to O’nyong nyong viral infections.
One focus in the laboratory is the study the Anopheles midgut responses to Plasmodium invasion. Plasmodium undergoes its major developmental bottleneck in the mosquito  during ookinete invasion of the mosquito midgut epithelium, and before it develops into its highly multiplicative stage (oocyst) as it resides on the basal lamina (Figure 1). We believe that parasite losses at that time are due to killing mechanisms of the mosquito. To elucidate aspects of the cellular and molecular mechanisms that regulate this invasion process we monitored the ookinete and epithelial cell responses at the cell biological and genome-wide transcriptional level, followed by functional analysis of identified genes by reverse genetics (RNAi). This research has revealed complex reciprocal interactions between mosquito cells and the parasite, and identified a repertoire of ookinete and epithelial cell motile behaviors that determine successful parasite transmission . Transcriptional profiling of A. gambiae midgut epithelia during invasion by P. berghei revealed that 6.5% of the mosquito transcriptome is differentially regulated during parasite invasion. Many of the genes that are consistently up or down regulated belong to known immunity gene families (see below) or have been shown previously to respond to bacterial challenges. However, the most prominent regulated cluster comprises genes involved in regulation of cytoskeleton reorganization and epithelial restitution, two biological processes which we have shown previously to occur during ookinete penetration of the mosquito midgut .
- Dimopoulos G, Christophides GK, Meister S, Schultz J, White KP, Barillas-Mury C, and Kafatos FC. . pmid:12077297.
- Sinden RE and Billingsley PF. . pmid:11323288.
- Vlachou D, Zimmermann T, Cantera R, Janse CJ, Waters AP, and Kafatos FC. . pmid:15186403.
Dissecting the mosquito immune pathways. Pioneering studies in the fly Drosophila melanogaster contributed to the detailed understanding of innate immunity and showed that the underlying mechanisms have largely been conserved through the course of metazoan evolution . These studies revealed most of the components of two conserved immune signaling pathways, Toll and Imd, that are utilized by the fly to respond to bacterial and fungal infections (Figure 2A). The availability of the A. gambiae genome sequence  previously allowed us to perform a comparative genomic analysis of putative immunity genes between Anopheles and Drosophila . Although the majority of the intracellular components of the Toll and Imd pathways are conserved between the two organisms (Figure 2B), a number of differences – which may have significant impact on immunity mechanisms – were also identified. The most important of these is the absence of the NF-kB-like transcription factor Dif in Anopheles. This suggested that in a functional Toll pathway either REL1 (the ortholog of Dorsal) or REL2 (the ortholog of Relish) might substitute for Dif. We investigated the role of REL2 and other molecules that are possibly implicated in the same signalling pathway in the mosquito immune responses. REL2 regulates the inducible expression of the various antimicrobial peptide genes including CEC1 and the key parasite antagonist, LRIM1 . We showed that the REL2 gene is alternatively spliced, resulting in two protein isoforms that are differentially implicated in defense against Gram-positive or Gram-negative bacteria. Thus, through alternative splicing Anopheles uses a single gene to mediate reactions for which Drosophila employs two genes, Relish and Dif. The REL2 pathway is also involved in the control of Plasmodium parasite infection of the mosquito midgut. Silencing of the pathway drastically increases the parasite numbers that successfully develop into oocysts.
Deciphering the molecular mechanisms affecting ookinete survival in the mosquito. Vector immune responses are responsible, at least in part, for the major parasite losses during parasite development. We now know, largely from functional studies in our laboratory, that even what is called “susceptible” A. gambiae mosquitoes effectively kill a large number of invading ookinetes (80 %), which are most probably cleared by lysis. Two genes, LRIM1 and TEP1, were shown to be strongly involved in ookinete killing in susceptible mosquitoes (G3 strain). LRIM1 expression was strongly induced in mosquito midguts and carcasses in response to the invasion of the mosquito midgut epithelium by Plasmodium ookinetes. The transient KD of LRIM1 in susceptible mosquitoes by RNAi resulted in approximately a four-fold increase in the number of parasites that successfully develop in the mosquito midguts, suggesting that LRIM1 is involved in parasite killing. TEP1 is as a bona fide pattern-recognition receptor that binds to the surface of bacteria and Plasmodium ookinetes. In a collaborative work led by E. Levashina (former staff scientist in our lab and currently an independent scientist at Institut de Biologie Moléculaire et Cellulaire, CNRS in Strasbourg) in vivo RNAi was used to KD TEP1 in adult susceptible mosquitoes and we demonstrated that this binding promotes two distinct immune reactions: phagocytosis of bacteria  and killing of Plasmodium ookinetes in the mosquito midgut . These reactions are reminiscent of functions of complement factors in vertebrates . To complete the functional analysis of TEP1 in vivo, we have established TEP1 gain-of-function transgenic mosquitoes, which overexpress the TEP1 gene under the control of the ubiquitous D. melanogaster heat shock protein 70 promoter. Two transgenic lines were obtained. Surprisingly, in one of the two lines, expression of the endogenous TEP1 and of the transgene was lost in a heat-shock independent manner. The knockout of TEP1 by transgenesis gave the same phenotype as that observed in mosquitoes in which TEP1 was transiently knocked down by RNAi. We are currently studying possible mechanisms of this inactivation. Importantly, overexpression of TEP1 in the second transgenic line lead to a significant decrease in parasite numbers suggesting that TEP1 is not only necessary but also sufficient to promote killing of parasites. Functional studies of this type will be facilitated in the future by conditional transgenic protocols of the type developed recently in a collaboration involving former members of the lab .
In extreme cases, genetic selection of formerly susceptible mosquitoes can give rise to complete refractoriness: An L3-5 strain of A. gambiae melanizes all ookinetes in the basal labyrinth of the midgut epithelium, while another refractory strain lyses the ookinetes in the cytosol of the midgut epithelial cells. Studies from our own and other laboratories, employing mapping of quantitative trait loci (QTL) that affect the L3-5 phenotype, revealed that different QTLs are linked to melanization of two different but related species of Plasmodium, P. cynomolgi B and P. cynomolgi Ceylon. This suggests that different genetic loci may be involved in L3-5 responses to different malaria parasites [12, 13]. We are currently developing fluorescence polarization SNP genotyping methods for fine scale QTL mapping to further pinpoint the genetic basis of the melanization phenotype in the L3-5 strain.
In general, melanization in insects requires the limited proteolytic cleavage of prophenoloxidase (PPO) into active phenoloxidase (PO) by PPO-activating enzymes (PPAEs), a reaction believed to be influenced, both positively and negatively by additional factors including serine protease homologs (SPHs), serpins and C-type lectins (CTL, see below). The identification and biochemical study of components of the PPO-activating cascade is an important goal for understanding how melanization is regulated in A. gambiae. Nine PPO genes have been identified in the A. gambiae genome, in contrast to merely 3 PPO genes in D. melanogaster. These 9 PPO genes are transcribed at different developmental stages, showing partly overlapping expression profiles. Expression of several PPOs is also induced during blood meal digestion. We are currently analysing the function of PPO genes in the melanization process using in vivo RNAi. Our first results indicate that a combination of several PPO is involved in parasite melanization. PPOs are activated by PPAEs, which are trypsin-like serine proteases containing an amino-terminal CLIP-domain with an unknown function. Of the 41 CLIP-domain encoding genes identified in A. gambiae, subfamily CLIPB, containing 17 genes, is of special importance due to significant structural identity with other insect PPAEs involved in melanization. Functional studies using RNAi identified seven CLIPBs involved in ookinete melanization and two in ookinete killing through a distinct mechanism, probably lysis. These results suggest the involvement of a serine protease cascade in parasite killing. The putative interactions of these CLIPBs with each other and/or with other identified immune proteins are under investigation in order to understand how the cascade regulates downstream effector mechanisms.
A group of proteins that may interact with CLIPBs in the melanization response are a group of serine-protease inhibitors, called serpins. Over the last few years several serpins have been shown to negatively regulate the PPO cascade, by directly inhibiting PPAE. One of these serpins is Spn27A from D. melanogaster [14, 15] which forms an orthologous group with three A. gambiae genes (SPRN1, 2 and 3). Reverse genetic analysis revealed that SRPN2 is the functional Spn27A homolog in adult female mosquitoes, causing spontaneous melanization in the mosquito. Depletion of SRPN2 from the hemolymph also negatively affects vector competence towards P. berghei infection, by strongly reducing the number of developing oocysts. Interestingly, the mosquito SRPN2 gene cannot rescue the melanization phenotype of the Spn27A mutant fly strain, suggesting differences in the downstream targets that these serpins inhibit.
On the other hand, previous studies from several insect species revealed that melanization is positively regulated by members of the CTL and SPH subfamilies. The picture seems to be reversed in A. gambiae, in which functional analysis of the CTL genes using RNAi allowed the identification of two CTLs, CTL4 and CTLMA2 that block parasite melanization, consequently protecting Plasmodium ookinetes from this potent immune response . The mechanism(s), by which these CTLs block melanization, is not yet elucidated. However, by analogy to published data, we hypothesize that CTL4 and CTLMA2 might associate with SPH of the CLIPA subfamily to form functional complexes which inhibit the PPO cascade. To this purpose, we functionally analysed eight out of ten CLIPA genes using RNAi in adult susceptible mosquitoes and scored the effect of the different gene KD on Plasmodium survival, with the aim of identifying CLIPA genes whose knockdown phenotype is similar to that previously described for CTL4 and CTLMA2. The preliminary results were intriguing and revealed that different CLIPA proteins have opposing functions in Plasmodium melanization. Work is in progress to investigate the potential interactions between CTL4, CTLMA2 and the CLIPA proteins that inhibit melanization. The aim is to decipher the role of these lectins in melanization and to shed light on the mechanisms that govern, at least in part, successful parasite survival in its vector.
- Hoffmann JA and Reichhart JM. . pmid:11812988.
- pmid= 12364791
- Christophides GK, Zdobnov E, Barillas-Mury C, Birney E, Blandin S, Blass C, Brey PT, Collins FH, Danielli A, Dimopoulos G, Hetru C, Hoa NT, Hoffmann JA, Kanzok SM, Letunic I, Levashina EA, Loukeris TG, Lycett G, Meister S, Michel K, Moita LF, Müller HM, Osta MA, Paskewitz SM, Reichhart JM, Rzhetsky A, Troxler L, Vernick KD, Vlachou D, Volz J, von Mering C, Xu J, Zheng L, Bork P, and Kafatos FC. . pmid:12364793.
- Osta MA, Christophides GK, and Kafatos FC. . pmid:15044804.
- Levashina EA, Moita LF, Blandin S, Vriend G, Lagueux M, and Kafatos FC. . pmid:11257225.
- Blandin S, Shiao SH, Moita LF, Janse CJ, Waters AP, Kafatos FC, and Levashina EA. . pmid:15006349.
- Blandin S and Levashina EA. . pmid:14698229.
- Lycett GJ, Kafatos FC, and Loukeris TG. . pmid:15342516.
- Zheng L, Cornel AJ, Wang R, Erfle H, Voss H, Ansorge W, Kafatos FC, and Collins FH. . pmid:9103203.
- Zheng L, Wang S, Romans P, Zhao H, Luna C, and Benedict MQ. . pmid:14577840.
- De Gregorio E, Han SJ, Lee WJ, Baek MJ, Osaki T, Kawabata S, Lee BL, Iwanaga S, Lemaitre B, and Brey PT. . pmid:12408809.
- Ligoxygakis P, Pelte N, Ji C, Leclerc V, Duvic B, Belvin M, Jiang H, Hoffmann JA, and Reichhart JM. . pmid:12456640.
Malaria and mosquito population dynamics
Transmission of the Plasmodium parasite to a human host completely depends on the availability of a competent mosquito vector. A. gambiae is the most important and efficient vector for transmission of human malaria in Africa. In some cases, A. gambiae kills the Plasmodium parasite, thus blocking the transmission cycle. The interactions between the vector and the parasite involve the mosquito’s innate immunity. Sequence polymorphisms occurring in immune-related genes (as already documented for TEP1) may reflect phenotypic variations in vector competence. Moreover polymorphism may be indicative of adaptation in a co-evolving mosquito/parasite system. We therefore screened four strains of A. gambiae (susceptible and refractory to Plasmodium parasites) and wild populations for their polymorphism. Sixty immune-related genes were sequenced, and Single Nucleotide Polymorphisms (SNPs) were identified by alignment. The diversity in immune-related genes was high compared to other parts of the genome, suggesting a diversifying selection acting on these sequences. The identified SNPs will be used to investigate the potential association between genotypes and the susceptibility/refractoriness of A. gambiae to the parasite under field conditions.