Combining two-dimensional gel electrophoresis and metabolomic data in support of dry-season survival in the two main species of the malarial mosquito Anopheles gambiae☆

In dry savannahs of West-Africa, the malarial mosquitoes of the Anopheles gambiae sensu stricto complex annually survive the harsh desiccating conditions of the dry season. However, the physiological and biochemical mechanisms underlying how these mosquitoes survive such desiccating conditions are still undefined, and controversial. In this context, we provide the first work examining both proteomic and metabolomic changes in the two molecular forms of A. gambiae s.s (M and S forms) experimentally exposed to the rainy and dry season conditions as they experience in the field. Protein abundances of the mosquitoes were measured using a two-dimensional fluorescence difference gel electrophoresis (2D DIGE) coupled with a matrix-assisted laser desorption/ionisation-time of flight (MALDI-TOF) and tandem mass spectrometry (MS) for protein identification. These assays were conducted by Applied Biomics (http://www.appliedbiomics.com, Applied Biomics, Inc. Hayward, CA, USA), and the mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium (http://proteomecentral.proteomexchange.org) via the PRIDE partner repository with the dataset identifier PXD000294. The metabolomic analysis was conducted using both Acquity UPLC® system (for amino acid identification), and a gas-chromatography-mass spectrometry platform (for sugars identification). Metabolomic fingerprintings were assessed in the University of Rennes 1, UMR CNRS 6553 EcoBio (France). A detailed interpretation of the obtained data can be found in Hidalgo et al. (2014) [1] (Journal of Insect Physiology (2014)).


a b s t r a c t
In dry savannahs of West-Africa, the malarial mosquitoes of the Anopheles gambiae sensu stricto complex annually survive the harsh desiccating conditions of the dry season. However, the physiological and biochemical mechanisms underlying how these mosquitoes survive such desiccating conditions are still undefined, and controversial. In this context, we provide the first work examining both proteomic and metabolomic changes in the two molecular forms of A. gambiae s.s (M and S forms) experimentally exposed to the rainy and dry season conditions as they experience in the field. Protein abundances of the mosquitoes were measured using a two-dimensional fluorescence difference gel electrophoresis (2D DIGE) coupled with a matrix-assisted laser desorption/ionisation-time of flight (MALDI-TOF) and tandem mass spectrometry (MS) for protein identification. These assays were conducted by Applied Biomics (http://www.appliedbiomics.com, Applied Biomics, Inc. Hayward, CA, USA), and the mass spectrometry  proteomics data have been deposited to the ProteomeXchange Consortium (http://proteomecentral.proteomexchange.org) via the PRIDE partner repository with the dataset identifier PXD000294. The metabolomic analysis was conducted using both Acquity UPLC s system (for amino acid identification), and a gas-chromatography-mass spectrometry platform (for sugars identification). Metabolomic fingerprintings were assessed in the University of Rennes 1, UMR CNRS 6553 EcoBio (France). A detailed interpretation of the obtained data can be found in Hidalgo   Proteomic data were assessed using a two-dimensional fluorescence difference gel electrophoresis (2D DIGE) coupled with a matrix-assisted laser desorption/ ionisation-time of flight (MALDI-TOF) and tandem mass spectrometry (MS) Metabolomic data were assessed using a gas-chromatography-mass spectrometry (GC-MS) and an Acquity UPLC s system (Waters Corporation, Milford, MA, USA) platforms Data format Raw, analyzed Experimental factors We compared the protein and metabolites (amino acids, polyols, sugars) in two malarial mosquito species form the Anopheles gambiae species complex (formerly known as the M and S molecular forms). The two species were reared under climatic conditions mimicking the rainy and dry season conditions as they experience naturally.

Data accessibility
The mass spectrometry proteomics data have been deposited to the Proteo-meXchange Consortium (http://proteomecentral.proteomexchange.org) via the PRIDE partner repository with the dataset identifier PXD000294.

Value of the data
The use of 2D gel electrophoresis comparisons identified interesting changes in several protein abundances from one season to the next in mosquitoes; The comparisons of mosquito's metabolic fingerprints highlighted specific changes in their metabolite contents (amino acids and sugars) from one season to the next; By combining both 2D gel electrophoresis and metabolic fingerprint comparisons our approach improves the identification of physiological and biochemical changes in mosquitoes.
Data, experimental design, materials and methods

Data
Mosquitoes from the Anopheles gambiae s.l. complex (both M and S molecular forms of A. gambiae) are widespread in West Africa where they seasonally experienced the harsh desiccating conditions of the dry season. How these mosquitoes survive the dry season and are engaged in exponential population growth as soon as the rainy season starts is still undefined and very controversial [2,3]. This ability to survive the dry season suggests high physiological plasticity in mosquitoes. In order to highlight the general physiological changes involved in mosquitoes at the onset of the dry season, we compared protein abundance and metabolic fingerprints in the two molecular forms of A. gambiae exposed to the climatic conditions of the rainy and the early dry season.
The M and S molecular forms of A. gambiae were recently described as two different species, named Anopheles coluzzii and A. gambiae, respectively [4,5]. Here, we are still referring to the M and S molecular forms, as in our published paper (Hidalgoet al. [1]).

Experimental design
Mosquitoes were reared from eggs to adults into four programmable climatic chambers (Sanyo MLR 315 H, Sanyo Electric Co., Osaka, Japan). Two climatic chambers were used to reproduce the climatic conditions of the rainy season, and two others to reproduce those of the dry season. Climatic conditions were programmed using the temperature and humidity cycles recorded in south-western Burkina-Faso (11°23′N, 04°24′W) with a Vantage Pro2 weather monitoring station (Weatherlink; Davis Instruments, Hayward, CA, U.S.A.). Climatic conditions were recorded in August 2010 (rainy season) and December 2010 (onset of the dry season), then hourly averaged to design 12 step cycles required to programme as closely as possible the natural daily climatic variations inside the climatic chambers [6]. Conditions inside the climatic chambers were tightly monitored throughout experiments using MicroLog Pro monitors (EC750, Davis Instruments, Hayward, CA, U.S.A.).

Proteomic assays
For each mosquito species (M and S molecular forms) and rearing conditions (rainy and dry season), proteomic samples consisted of a pool of 30 teneral (1 h-old adult) female mosquitoes. Samples were sent to the proteomics department of Applied Biomics (http://www.appliedbiomics. com, Applied Biomics, Inc. Hayward, CA, USA) for proteomic assays as described below. ) was used to extract the proteins in each sample of 30 pooled females. Then, the protein concentration of each sample was measured using a Bio-Rad protein assay method, and adjusted to 5 mg.ml À 1 using the 2-D cell-lysis buffer.

Samples labelling
For both species, we used the CyDye DIGE Fluor Cy3 (red) and Cy5 (green) saturation dyes (4 mL, 2 mM) to label samples from the rainy and the dry season rearing conditions, respectively. A Cy2 dye, corresponding to an internal standard for the normalisation of protein abundances, was also used to label a 50:50 mixture of the two rearing conditions samples. The three CyDye DIGE Fluors were incubated with corresponding samples for 30 min on ice, under dark conditions, before adding a 1 mL volume of 10 mM lysine. A 2 Â 2-D sample buffer (8 M urea, 4% CHAPS, 20 mg/mL DTT, 2% pharmalytes and a trace amount of bromophenol blue) was used to terminate the reaction, and the samples were stored at À 80°C until further use for 2D DIGE assays.
The electrophoresis was performed at 15°C on a 12% SDS-PAGE gel. Note that each gel contained the three Cy3, Cy5 and Cy2 labelled samples. For each mosquito species, we performed three-replicate gels to compare the effects of rearing conditions on their protein expression (N ¼6 gels in total). A seventh gel containing the two species exposed to dry conditions and their 50:50 mixture was run. In this last gel, the M molecular form was labelled with Cy5 dye, whereas the S form was labelled with Cy3 dye.

Image analysis
Effect of rearing conditions on protein expression in both species and differences between species were assessed by overlaying images of the seven different gels. Note that the abundance of each protein spot was normalised using the seventh gel. The resulting 2D gels were scanned using a Typhoon Trio scanner (Amersham BioSciences) with excitation and emission wavelengths for Cy2labelled (488/520 nm), Cy3-labelled (548/560 nm) and Cy5-labelled (641/660 nm) proteins using settings that match in similar relative fluorescence intensities for the Cy3-or Cy5-labelled samples (Fig. 1). Image analysis for intensity measurements of the different protein spots was performed using ImageQuantTL and DeCyder softwares (GE Healthcare). Then, images were subjected to in-gel analysis and cross-gel analysis using DeCyder v6.5 (GE-Healthcare) with a P-value o 0.05 according to Student's t-tests. Based on these tests, we retained a total of 109 spots, corresponding to proteins with molecular masses ranging from 14 to 150 kDa, and isoelectric points between 4 and 9. These spots showed significant variation in at least one species and/or one rearing condition.

Protein identification using matrix-assisted laser desorption/ionisation-time of flight (MALDI-TOF) and tandem mass spectrometry (MS)
Identification of the 109 protein spots of interest was performed by Applied Biomics, Inc. Spots were picked up from the gel using an Ettan spot picker (GE Healthcare), and washed twice with 25 mM ammonium bicarbonate and 50% acetonitrile to remove the remaining dyes. Then, proteins were digested in-gel at 37°C using a modified porcine trypsin protease (Trypsin Gold, Promega) according the method described in Rosenfeld et al. [7]. The digested tryptic peptides were desalted using Zip-tip C18 (Millipore, Billerica, MA), mixed with 0.5 mL of α-cyano-4-hydroxycinnamic acid (CHCA) matrix and spotted into wells of a MALDI plate. Mass spectra of the peptides in each digested spot were obtained using a MALDI-TOF (MS) and a TOF/TOF (tandem MS/MS) equipment (AB Sciex). The MALDI-TOF mass spectra data were acquired in reflectron positive ion mode, averaging 2000 laser shots per spectrum. The TOF/TOF tandem MS fragmentation spectra were acquired for each sample, averaging 2000 laser shots per fragmentation spectrum on each of the 5-10 most abundant ions present in each sample (excluding trypsin autolytic peptides and other known background ions). Both the peptide mass and the associated fragmentation spectra were submitted for database search using GPS Explorer software version 3.5 (Applied Biosystems) equipped with the MASCOT search engine (http://www.matrixscience.com, Matrix science). This submission allows the identification of proteins from the National Center for Biotechnology Information non-redundant protein sequence database (NCBInr), restricted to the A. gambiae species complex. The searches were performed without constraining the protein molecular weight or isoelectric point, with carbamidomethylation and oxidation as variable modifications, and with one missed cleavage allowed in the search parameters. The highest protein scoring hit with a protein score confidence interval over 95% from the database search for each 2D gel spot was accepted as positive identification.
Based on an average difference of at least 1.3-fold (absolute value), only 39 spots were selected, which corresponded to 29 distinct proteins, for further identification. The ratio of the 39 spot expressions between the species and/or rearing conditions was obtained from in-gel analysis using DeCyder. Table 1 shows the identification of the 39 spots retained in A. gambiae based on this method.

Data analysis
Average-fold differences of the 39 spot abundances were first normalised (Logarithmic abundance of the protein contents from each 2D DIGE gel), then calculated for the four following comparisons: (i) rainy vs. dry condition in A. gambiae M; (ii) rainy vs. dry condition in A. gambiae S; (iii) A. gambiae M vs. A. gambiae S when exposed to rainy conditions; and (iv) A. gambiae M vs. A. gambiae S when exposed to dry conditions. For each ratio, differences in spot abundance were assessed using Student's t-tests on individual spots (P-value o0.05).
Accordingly, we observed that 15 proteins exhibited significant variations between the two rearing conditions in 1-h old females A. gambiae M, among which 10 were more abundant when females grew under the dry conditions ( Fig. 2A). In contrast, 14 proteins showed significant variations between the two rearing conditions in 1-h old females A. gambiae S, among which 8 were more abundant when females grew under dry conditions (Fig. 2B). Overall, nine proteins exhibited the same pattern of variations in both species: six proteins increased under dry conditions, whereas the three others decreased under these conditions ( Fig. 2A and B).
Finally, 24 proteins showed significant variation between species whatever the rearing conditions. Six of the 24 proteins were always more abundant in females A. gambiae S (Fig. 3). For details about data interpretation and conclusion please report to Hidalgo et al. [1].

Metabolomic assays: amino acid and sugar concentrations
Metabolomic fingerprinting was conducted on both 1 h-old and 24 h-old adult females. For each mosquito species (M and S molecular forms) and rearing conditions (rainy and dry season), metabolomic samples consisted of a pool of five-six female specimens, so that the dry mass of each sample reached at least 1 mg. Five to nine replicates were used for the different experimental conditions.

Extraction of the metabolites
Samples were first freeze-dried (Lyovac™ GT3) for 72 h, before being weighed using a microbalance (Mettler Toledo GmbH © , Greinfense, Switzerland, d¼ 1 mg). Two 3 mm tungsten beads and a 1000 mL volume of methanol-chloroform (2:1) solution were added to each sample. Then, each sample was shaken at 30 Hz for 1.5 min (Retsch™ MM301, bead-beating, Retsch GbmH, Haan, Germany). We added a 500 μL volume of ultrapure water. Samples were vortexed for homogenisation and centrifuged at 4000g for 10 min at 4°C. Finally, we collected and transferred a 600 mL aliquot of the upper aqueous phase (containing amino acids, polyols and sugars) into clean microtubes. The 600 mL aliquot was vacuum-dried (Speed Vac Concentrator, Genevac Ltd., Ipswitch, England), and a volume of 600 mL of ultrapure water was added to the residual. Table 1 List of the 39 protein spots displaying differential abundance (average fold difference41.3, absolute value; Po 0.05) between the two rearing conditions in 1-h old female A. gambiae M and S, or between the two species whatever the rearing conditions. Adapted from Hidalgoet al. [1].

Identification and quantification of amino acids
Twenty mL of each sample previously extracted were diluted with 60 mL of ultrapure water. Then, a 5 μL volume of each diluted sample was transferred into a new microtube for derivatization. Derivatization was conducted according to the AccQTag ultra derivatization kit (Waters Corporation, Milford, MA, USA) procedures. After reconstituting the AccQTag Ultra Reagent Powder, we prepared a Fig. 2. Differential fold-change of the 21 selected spot proteins between rainy and dry conditions in 1-h old females of Anopheles gambiae M (A) and S (B). Blue bars are positive values that represent spots that were more abundant in rainy conditions, whereas red bars are negative values that represent spots that were more abundant in dry conditions. Spots with "n" represent the nine protein spots exhibiting the same pattern of variations in both species. From Hidalgo et al. [1]. 100 pmol/ml calibration standard solution (Acide Hydrolase Standard). Then, a 10 ml volume of the calibration standard solution and a 20 ml volume of AccQTag Ultra Reagent were added to each sample. Samples were vortexed and heated for 10 min at 55°C.
After derivatization, 1 mL of each sample was transferred in an Acquity UPLC s system (Waters Corporation, Milford, MA, USA) equipped with an Acquity UPLC s BEH C18 1.7 μm 2.1 Â 100 mm 2 column heated at 55°C, as described by Renaultet al. [8]. The derivatized amino acids were detected at 260 nm using a photo diode-array detector. Peaks were identified according to their retention time compared with a list of 22 amino acid commercial standards (Table 2). Concentration was calculated by comparison of each amino acid peak area with those of the individual external standards.

Identification and quantification of sugars
A 50 mL volume of each sample previously extracted was transferred into a clean glass vial and vacuum-dried. Then, residuals were resuspended in 50 mL of freshly prepared methoxyamine hydrochloride (Sigma-Aldrich, St. Louis, MO, USA) in pyridine (20 mg/mL). Samples were incubated under orbital shaking at 30°C for 90 min. Following incubation, 50 μl volume of N-methyl-N-(trimethylsilyl) trifluoroacetamide (MSTFA) were added for derivatization. Derivatization was conducted overnight at 37°C.
To identify and quantify sugars, we used gas chromatography coupled with mass spectrometry (GC-MS). The GC-MS system consisted of a TriPlus autosampler, a Trace GC Ultra chromatograph and a Trace DSQII quadrupole mass spectrometer (Thermo Fischer Scientific Inc, Waltham, MA, USA). A 1 ml volume of the derivatized sample was injected into the GC-MS system using the split mode (25:1). The injector temperature was held at 260°C. We programmed the GC-MS in order that the oven temperature remained 4 min at 70°C after the injection, and then increased at a rate of 5°C/min until it reached 300°C, where it remained for 10 min. For elution, we used a 30 m fused silica column (TR5 MS, I.D. 0.25 mm, 95% dimethyl siloxane, 5% phenyl polysilphenylene-siloxane) and helium as the carrier gas at a rate of 1 mL/min. Fig. 3. Differential fold-change of the 24 selected spot proteins between M and S forms reared under rainy (blue bars) and dry (red bars) conditions. Positive values represent spots that were more abundant in Anopheles gambiae M, whereas negative values represent spots that were more abundant in Anopheles gambiae S. Six proteins were more abundant in females A. gambiae S, whatever the experimental conditions, among which prophenoloxidase (spot #76), citrate synthase (spot #84), isocitrate dehydrogenase (spot #85), and glutathione transferase (spot #99) can be mentioned. The RR-2 cuticular protein 123 (spot #101) was also significantly more abundant under the dry conditions (Student t-test, Po 0.001), although its fold change was only 1.27. In addition, a phosphorylase (spot #75) was more abundant in A. gambiae S compared to A. gambiae M when the females were exposed to the dry conditions, but no differential expression was found under rainy conditions. Arginine kinase (spot #96) was always more abundant in A. gambiae M, whatever the experimental conditions. From Hidalgo et al. [1].
Detection of each sugar was conducted by mass spectrometry using the electronic impact (EI) method. Briefly, the temperature of the ion source and mass spectrometry transfer line was set at 260°C. All of the samples were run under the SIM mode (electron energy: À 70 eV). Each sugar peak was annotated using both mass spectra (two specific ions), and the retention index specific to each compound. Randomised sample and standard sequences were established for the sample injection. Chromatograms were deconvoluted using the XCalibur v2.0.7 software (Thermo Fisher Scientific, San Iose, CA, USA). Standard samples consisted of the pure reference compounds (arabinose, fructose, galactose, glucose, ribose and trehalose) at 100, 200, 300, 500 and 1000 mM concentrations. Sugars concentrations were thus quantified using the quadratic calibration curves for each pure reference compound.

Amino acids
Among the 22 amino acids we listed in our external standards, 21 were correctly detected and quantified (i.e. cysteine did not provide reliable measurements, as this amino acid is unstable and is quickly transformed into cystine during derivatization). Significance of amino acid fingerprintings among species and rearing conditions was assessed using multivariate discriminant analyses and MANOVAs. Because false discovery rate can be induced by multiple comparisons, we used a Benjamini-Hochberg procedure to adjust all P-values.
Comparisons of the amino acid fingerprints of each experimental modality showed significant differences in the concentration of at least one amino acid between the two species, their ageing and their rearing conditions. Of particular interest, both species and especially M form, accumulated phenylalanine, tryptophan, tyrosine, and valine 1 h after emergence ( Fig. 4A and B). S form specimens did not show fingerprint variation according to the rearing conditions 1 h after emergence, but significant changes were observed 24 h after emergence, with proline being accumulated under dry conditions (Fig. 4B). For details about data interpretation and conclusion, please report to Hidalgo et al. [1].

Sugars
Among the six sugars listed in our reference standard, only galactose and glucose were reliably quantified in our mosquito samples. Indeed, arabinose, fructose, ribose and trehalose were below the quantification limit of the equipment used. Significance of the galactose and glucose concentrations variation among species and rearing conditions was assessed using Kruskal-Wallis tests, followed by a Bonferroni procedure resulting in a decreased threshold of significance from P-value o0.05 to P-value o0.01. Accordingly, both galactose and glucose concentrations were significantly higher 24 hours after emergence in the two species when mosquitoes were reared under the dry conditions (Fig. 5A-D). Moreover, glucose concentration was lower in M form specimens reared under the dry conditions ( Fig. 5B and C). For details about data interpretation and conclusion please report to Hidalgo et al. [1]. Fig. 5. Mean ( 7 s.e.) galactose (A and B) and glucose (C and D) levels (in n moles mg À 1 dry mass) in 1-h and 24-h old females of Anopheles gambiae M (A and C) and S (B and D). Blue bars correspond to the females reared under rainy season conditions and light red bars correspond to females reared under dry season conditions Letters above the bars report significant differences after Bonferroni correction to account for multiple tests (Po 0.01). From Hidalgo et al. [1].