Findings

Research hypothesis

Factors that influence mosquito fitness, especially host seeking and mate finding are complex and modulated by multiple cues, of which olfactory cues are most important [1–4]. Detection of odor molecules requires odorant binding proteins (OBPs) that are abundant in antennal chemosensilla [5, 6]. OBPs are low molecular weight soluble proteins that bind and transport odor molecules from sensillae to G-protein-coupled receptors in olfactory sensory neurons [6]. The finding of receptor AgamOBP1 binding to its ligand indole demonstrated the significance of OBPs in odor recognition [7]. Understanding olfactory function could lead to development of malaria control strategies based on repelling Plasmodium sp. carrying Anopheles mosquitoes or attracting them into a fatal trap. A first step is assessment of expression of olfactory system associated genes [7–10]. There is sexually dimorphic expression of OBPs in Anopheles mosquitoes and Drosophila melanogaster[11–13]. We are focusing on identifying OBPs in antennae of Anopheles gambiae, because in Africa A, gambiae is the most important vector of Plasmodium falciparum[14], a major vector of Wuchereria bancrofti, which causes lymphatic filariasis [15], and a vector of O'nyong-nyong virus [16]. In this study, based on results of a screening microarray (unpublished) and previous microarray studies [9, 11], we hypothesized that the OBP, OBP2 (AGAP003306), would have increased transcript accumulation by quantitative reverse transcription PCR (qRT-PCR) in female as compared to male A. gambiae antennae.

Methods

Collection and Processing of RNA

We studied AGAP003306, which had 2 fold greater expression in RNA isolated from antennae of 4 day old A. gambiae (Keele strain from Johns Hopkins) females than males in a microarray experiment (unpublished). In another microarray study of RNA from antennae of 5-7 day old Pink-eye A. gambiae, expression of AGAP003306 (OBP2) was 1.4 times higher in females as compared to males, but no qRT-PCR was done [9]. In yet another microarray study of RNA isolated from 3 day old whole mosquitoes (Pink-eye strain A. gambiae) there was approximately 3 fold increased expression in females vs males [17].

In this study, A. gambiae sensu stricto (G3 strain) were from the same batch of eggs (each batch giving rise to a mosquito lot) and were raised to adulthood under standard insectary conditions, and fed ad libitum with 10% sugar water [18]. We studied the antennae under controlled conditions of age and exposure to food. Adults of both sexes were collected exactly 4 days after emergence. The mosquitoes were immobilized by exposure to -20°C for 15 minutes, males and females separated, and antennae removed by manual dissection over dry ice, placed into separate 1.5 mL centrifuge tubes and homogenized using a pestle, each in 300 μL of Trizol reagent (Invitrogen, CA). RNA was isolated following manufacturer's instructions and purified using RNeasy mini column (Qiagen). The RNA was then assessed for quality and quantity using NanoDrop (ND-1000). The mosquito antennae that generated the RNA for the qRT-PCR experiments were isolated in July 2009 (mosquitoes from the University of Maryland), and January 2010 and June 2010 (mosquitoes from the National Institutes of Health).

qRT-PCR Assay

As an endogenous control, and foundation for the qRT-PCR analysis, we used the S7 ribosomal RNA gene of A. gambiae[19]. As another control we analyzed AGAP009629, which did not have differential expression in antennae of females vs. males by microarray, but had increased expression in antennae of unfed vs. blood-fed females (unpublished).

The primer pairs synthesized and used (Table 1) were designed using Primer 3[20]. The QuantiTect SYBR Green RT-PCR kit (Qiagen) was used and reactions conducted in 96 well plates for 40 cycles (Applied Biosystems StepOnePlus™ System). Each reaction contained 12.5 μL Qiagen 2x Master Mix, 0.25 μL forward and reverse primers (0.5 μM final concentration), 0.25 μL RT Mix (containing reverse transcriptase), 10.75 μL DEPC water and 1.0 μL of RNA (0.05-1 μg/μL). The 96 well plate was sealed with adhesive film and centrifuged at 3700 rpm for 1 minute at 4°C. The ribosomal S7 gene was used as an endogenous control and a reaction without RT Mix (reverse transcriptase) was included for all reactions as a negative control. Changes in threshold cycles (ΔΔCT) analysis was done to assess the ratio of RNA expression in females vs. males using previously published methods for analysis (Step One Software, v2.2, Applied Biosystems [21, 22]).

Table 1 Primers used in qRT-PCR
Full size table

Phylogenetic analysis

Multiple sequence alignments were built using the KALIGN program [23], followed by manual adjustments on the basis of profile-profile and structural alignments. Phylogenetic analysis was conducted using an approximately-maximum-likelihood method implemented in the FastTree 2.1 program under default parameters [24].

Results

qRT-PCR

RNA was extracted from paired antennae of approximately 100 female and 100 male A. gambiae derived from the same batch of eggs and raised in the same cage that had never been exposed to a blood meal. 80 ng RNA/mosquito to 560 ng RNA/mosquito (mean of 270 ng RNA/mosquito) was obtained. Approximately double the RNA was obtained from female as compared to male antennae. We isolated antennae on three separate occasions from three separate batches of mosquitoes over a period of 2 years. The results of three assays using RNA isolated from antennae (from three separate lots of mosquitoes) at three different times and normalized to the expression of the S7 ribosomal RNA gene are shown in Table 2. In all experiments there was an increase in accumulation of transcripts of AGAP003306 in female vs. male A. gambiae (5 to 25 fold). When these data were integrated (Figure 1), there was a statistically significant increase in transcript accumulation of AGAP003306 in females vs. males (p = 0.037, paired student's t test). Furthermore, there was no significant difference in transcript accumulation between females and males of the control gene, AGAP009629 (p = 0.246, paired student's t test).

Table 2 Expression of AGAP003306 (OBP2) and AGAP0099629 (control) relative to expression of S7 in antennae of female and male A. gambiae in three biological replicates (experiments 1, 2 and 3).
Full size table
Figure 1
figure 1

Comparative expression of AGAP003306 in female and male A. gambiae. Columns show mean of ΔCт Mean values (normalized amount of AGAP003306 RNA present) of the three experimental replicates shown in Table 1 multiplied by -1, error bars are standard errors. Data are significantly different in a pair student's t-test (p = 0.037, n = 3).

Full size image

The relative increased accumulation of OBP2 transcripts in females vs. males in the three experimental (biological) replicates varied (Table 2). The mosquito antennae that generated the RNA for the qRT-PCR experiments were isolated in July 2009 and were from mosquitoes from the University of Maryland (experiment 1), and in January 2010 (experiment 2) and in June 2010 (experiment 3) both from the National Institutes of Health, Bethesda. Thus, the differences in relative transcript accumulation were likely due to biological variability in gene expression from mosquitoes from different laboratories studied at different times. To determine if this was the case rather than variability of the assay, we repeated the qRT-PCR assays using the RNA of experiments 2 and 3. These technical replicates showed similar results as the original experiments (Table 3). In a recent publication, similar results for the OBP2 gene were found in a single biological replicate of RNA from 4-6 day old A. gambiae by RNA-seq [13].

Table 3 Expression of AGAP003306 (OBP2) relative to expression of S7 in antennae of female and male A. gambiae in two technical replicates.
Full size table

Phylogenetic Analysis

OBP2 belongs to an OBP super family that includes the insect pheromone binding proteins [6]. Another member of this family, Agam OBP1, mediates indole recognition in antennae of female A. gambiae[7]. The olfactory receptors of terrestrial animals exist in an aqueous environment; yet detect odorants that are primarily hydrophobic. The aqueous solubility of hydrophobic odorants is thought to be greatly enhanced via OBPs, which exist in the extracellular fluid surrounding odorant receptors. This family includes proteins that specialize in binding insect pheromones (PBPs) and others that bind general odorants (GOBPs) [6]. Prior phylogenetic analysis has suggested that evolution of the OBP superfamily has evolved primarily through the process of lineage-specific expansion [25]. Thus, the majority of the OBPs in a given lineage such as Diptera, Hymenoptera, Lepidoptera or Coleoptera tend to cluster with others from the same lineage to the exclusion of those from other lineages. The genome of A. gambiae itself contains about 72 members of the OBP family.

We performed a phylogenetic analysis using over 100 representative OBPs from dipterans, hymenopterans and coleopterans with completely sequenced genomes. OBP2 is lodged within a predominantly dipteran lineage-specific expansion of OBPs that are particularly well represented in the culicid mosquitoes (Figure 2). This clade of OBPs includes the Culex (e.g. CquiOBP1) OBP that binds the oviposition kairomones (5R, 6S)-6-acetoxy-5-hexadecanolide [26] (marked red, Figure 2). This analysis also showed that orthologs of A. gambiae OBP2 are conserved across Culex, Aedes, and Anopheles genera but are absent in Drosophila (blue box, Figure 2), pointing to a function for this protein in potentially binding a conserved odorant molecule in culicid mosquitoes. The up regulation of OBP2 observed in females as compared to males, suggests it could possibly bind a molecule comparable to the oviposition kairomone bound by OBP1. However, such a kairomone could also have an alternative role in guiding female feeding behavior. On the other hand it is also possible that OBP2 binds a pheromone that males express. The focus of future studies would be to pinpoint the role of this protein by determining the impact of knocking it down vis-a-vis feeding behavior and fitness of females.

Figure 2
figure 2

Phylogenetic analysis. Phylogenetic analysis was done using over 100 representative OBPs from dipterans (magenta), hymenopterans (turquoise) and coleopterans (yellow). The light blue box shows that orthologs of A. gambiae OBP2 are conserved across Culex, Aedes, and Anopheles genera, but are absent in Drosophila. The orange box shows that this clade of OBPs includes the Culex (e.g. CquiOBP1) OBP that binds the oviposition kairomones (5R, 6S)-6-acetoxy-5-hexadecanolide. The lineage-specfic expansions in various insect lineages other than the OBP1 and OBP2 clades, which are discussed in the paper have been collapsed and the branch lengths equalized for simplicity of viewing.

Full size image

New interventions are needed to control the mosquitoes that transmit the parasites that cause malaria [27, 28] and lymphatic filariasis. Despite exciting scientific advances during the past few decades, no new approaches to mosquito vector control have been translated into widely used effective interventions. Sequencing the A. gambiae genome [29] and transcriptomics have provided a foundation for an approach to developing new interventions based on identifying genes and gene products that are important in transmission and mate-seeking. Stable genetic knockouts have not been generated in A. gambiae. However, transient knockdown by injection of sRNAi can be done and used to confirm the functional importance of OBP2 and other genes. This will be one of the next steps in our work.