Figures
Abstract
During a blood meal, female Anopheles mosquitoes are potentially exposed to diverse microbes in addition to the malaria parasite, Plasmodium. Human and animal African trypanosomiases are frequently co-endemic with malaria in Africa. It is not known whether exposure of Anopheles to trypanosomes influences their fitness or ability to transmit Plasmodium. Using cell and molecular biology approaches, we found that Trypanosoma brucei brucei parasites survive for at least 48h after infectious blood meal in the midgut of the major malaria vector, Anopheles coluzzii before being cleared. This transient survival of trypanosomes in the midgut is correlated with a dysbiosis, an alteration in the abundance of the enteric bacterial flora in Anopheles coluzzii. Using a developmental biology approach, we found that the presence of live trypanosomes in mosquito midguts also reduces their reproductive fitness, as it impairs the viability of laid eggs by affecting their hatching. Furthermore, we found that Anopheles exposure to trypanosomes enhances their vector competence for Plasmodium, as it increases their infection prevalence. A transcriptomic analysis revealed that expression of only two Anopheles immune genes are modulated during trypanosome exposure and that the increased susceptibility to Plasmodium was microbiome-dependent, while the reproductive fitness cost was dependent only on the presence of live trypanosomes but was microbiome independent. Taken together, these results demonstrate multiple effects upon Anopheles vector competence for Plasmodium caused by eukaryotic microbes interacting with the host and its microbiome, which may in turn have implications for malaria control strategies in co-endemic areas.
Author summary
In nature, females Anopheles mosquitoes that transmit the malaria parasites Plasmodium, take successive blood meals to maximize their offspring. During these blood meals, mosquitoes are exposed to a variety of microbes present in the host blood in addition to Plasmodium, the obligate parasite that causes malaria. The Trypanosoma parasites, causing trypanosomiases, are sympatric with the malaria parasites in numerous African regions, therefore, a single female mosquito could be in contact with both pathogens concurrently or through successive blood meals. In this work, we showed that exposure of females Anopheles mosquitoes to Trypanosoma enhanced their susceptibility to malaria parasites, reduced their reproductive fitness and modulated their bacterial gut flora. While the effect of trypanosomes ingestion on Plasmodium infection is microbiome dependent, the phenotype on the reproductive fitness is microbiome independent. These results highlight the need for considering the effect of eukaryotic microbes on Anopheles biology for malaria control strategies.
Citation: Dieme C, Zmarlak NM, Brito-Fravallo E, Travaillé C, Pain A, Cherrier F, et al. (2020) Exposure of Anopheles mosquitoes to trypanosomes reduces reproductive fitness and enhances susceptibility to Plasmodium. PLoS Negl Trop Dis 14(2): e0008059. https://doi.org/10.1371/journal.pntd.0008059
Editor: José M. C. Ribeiro, National Institute of Allergy and Infectious Diseases, UNITED STATES
Received: August 23, 2019; Accepted: January 15, 2020; Published: February 7, 2020
Copyright: © 2020 Dieme et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This work received financial support from the Institut Pasteur “Projet Transversaux de Recherche” grant (PTR_542) to CM, the European Commission, Horizon 2020 Infrastructures 731060 Infravec2 to KDV, the European Research Council, Support for frontier research, Advanced Grant #323173 to KDV, and the Agence Nationale de la Recherche Laboratoire d'Excellence “Integrative Biology of Emerging Infectious Diseases” #ANR-10-LABX-62-IBEID to KDV. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Vector borne diseases such as malaria, trypanosomiases, and others are often sympatric in Africa [1]. Therefore, in these areas, successive blood meals taken by adult Anopheles female mosquitoes during their life could increase their exposure to different microorganisms, including parasites, bacteria and viruses [1].
In areas with the highest malaria burden, such as in Sub-Saharan Africa, Plasmodium falciparum transmission depends on the complex ecological determinants (biotic and abiotic factors) that drive population dynamics of the primary African vectors Anopheles gambiae s.l. and A. funestus but also mosquito intrinsic factors for pathogen development [2–7]. Eukaryotic microbes, including microsporidia, gregarines and trypanosomatids, have been identified in mosquitoes. However, relatively little attention has been paid to Anopheles vector interactions with eukaryotic microbes and their potential impact on Plasmodium development in the mosquito [3].
African trypanosomiases (Human and Animal African Trypanosomiases) are neglected tropical diseases caused by Trypanosoma species most exclusively transmitted to humans and their livestock by the bite of both male and female tsetse flies of the genus Glossina. These diseases have devastating socio-economic consequences for Sub-Saharan Africa with 13 million people and about 50 million cattle at risk [8–10]. Between 1990 and 2015, almost 440,000 Human African Trypanosomiases (HAT) cases were reported [11]. At the same time, sustained international control efforts have reduced the number of new HAT cases to only 2,163 in 2016 [12]. Nevertheless, probably more cases remain undetected given that sleeping sickness tends to occur in remote rural areas and that latent infections in humans or animals could represent a significant proportion of the infected populations that would act as reservoirs for the parasites [13] [14].
Although Trypanosoma and Plasmodium infections have been extensively examined separately in their respective vectors, little is known about the possible effects of Anopheles mosquito co-exposures to both parasite species, successively or concurrently. Patients in endemic countries can be co-infected with Plasmodium and Trypanosoma parasites [15]. Therefore, concomitant exposures to Plasmodium and Trypanosoma in Anopheles vectors are theoretically possible in Africa. Moreover, the epidemiological roles of latent human cases and animal reservoirs in HAT endemic foci [9,13,16] are not well-documented and the current burden of Animal African Trypanosomiases (AAT) is a concern. In addition, recent studies on A. gambiae feeding behavior have shown high proportions of (i) blood meals actually taken on a single non-human host and (ii) mixed blood meals taken from both animals and humans [17,18]. Therefore, successive exposures of Anopheles vectors to Trypanosoma ingested from animals and Plasmodium ingested from infected humans are likely to occur. Based on all of this information, we hypothesized that in co-endemic areas where Anopheles mosquitoes are exposed to Trypanosoma brucei brucei in bloodmeals from animal hosts, these parasites could modulate the micro-environment of the mosquito midgut, potentially influencing the development of Plasmodium in the midgut [1].
Here, using Anopheles coluzzii, a member of the A. gambiae complex, we show that T. b. brucei bloodstream forms survive for at least 48h post-feeding in the A. coluzzii midgut before being cleared. This survival was sufficient to impact the abundance of the bacterial gut flora as well as mosquito reproductive fitness. In addition, T. b. brucei ingestion significantly increased Plasmodium infection prevalence in a microbiota-dependent manner. These findings suggest that T. b. brucei parasites affect a potential interplay between the gut microbiota and reproduction of A. coluzzii and, more importantly, that it could directly increase the risk of malaria parasite transmission.
Methods
Ethic statement
This study was conducted in strict accordance with the recommendations from the Guide for the Care and Use of Laboratory Animals of the European Union (European Directive 2010/63/UE) and the French Government. The protocol was approved by the “Comité d’éthique en expérimentation animale de l’Institut Pasteur” CETEA 89 (Permit number: 2013–0129), by the French Ministry of Scientific Research (Permit number: 202195.02) and undertaken in compliance with Institut Pasteur Biosafety Committee (protocol CHSCT 14.114).
Mosquitoes
A. coluzzii colony Fd03 initiated in Mali [19] was reared at 26°C and 80% humidity, on a 12 h light/dark cycle with access to cotton soaked in 10% sucrose solution, in insectaries of the Unit of Genetics and Genomics of Insect Vectors of the Institut Pasteur Paris, France.
Mouse infections with Trypanosoma brucei brucei
All experiments were performed with three-week-old female Swiss Mus musculus mice (Janvier, France). T. b. brucei AnTat 1.1E fluorescent bloodstream forms expressing a cytosolic chimeric reporter including a red fluorescent marker and a red-shifted bioluminescent marker (PpyREH9/TY1/TdTomato) were cultivated in HMI9 medium supplemented with 10% fetal calf serum at 37°C in 5% CO2 [20]. Parasitemia was assayed by automated fluorescent cell counting with a Muse cytometer (Merck-Millipore, detection limit 5.102 parasites/ml) according to the manufacturer’s recommendations. Parasites were counted, centrifuged and resuspended at 107cells/ml. Then, 106 parasites of this suspension were inoculated by intra-peritoneal (IP) injection to each mouse. Mice parasitemia was assessed under an inverted light microscope Leica DMIL (Leica) with standardized single-use hemocytometers (Hycor Kova, detection limit 104 parasites/ml) according to the manufacturer’s recommendations or by automated fluorescent cell counting with a Muse cytometer (Merck-Millipore).
Mouse infections with Plasmodium yoelii
Mice were inoculated with 105 red blood cells (RBCs) infected with GFP-transgenic P. yoelii strain, GFP@HSP70-GOMO [21]. Four days post-injection, blood samples were taken from the tail, and parasitemia was determined by flow cytometry. Furthermore, male gametocyte maturity was verified by performing an exflagellation test as previously described [22].
T. b. brucei survival in mosquito midgut
To assess T. b. brucei survival in mosquito midgut, mosquitoes were allowed to feed on cultured fluorescent T. b. brucei AnTat1.1E bloodstream forms. 119 and 68 mosquitoes were dissected at 2 and 5 days post-feeding, respectively, and entire midguts were examined by fluorescence microscopy to detect living red fluorescent parasites. Dissected midguts were scored twice for the presence of trypanosomes either 2 or 5 days post-feeding on mice.
Mosquito infection with the rodent malaria parasite P. yoelii
For all experiments, mice infected with P. yoelii, strain GFP@HSP70-GOMO at 5–6% parasitemia with mature gametocytes, were used. Mice were first anaesthetized by intraperitoneal (IP) injection of ketamine (Imalgene 1000 at 125 mg/kg) and xylazine (Rompun 2% at 12.5 mg/kg) before mosquitoes were allowed to feed for 30 min. Unfed mosquitoes were discarded and fed mosquitoes were maintained at 24°C (P. yoelii) and 70% relative humidity on 10% sucrose solution as previously described [23]. 3 independent biological experiments were performed.
Concomitant A. coluzzii co-infections with T. b. brucei and P. yoelii
Mice were infected either with P. yoelii or co-infected with both P. yoelii and T. b. brucei. For co-infected mice, T. b. brucei were inoculated 1 day after P. yoelii injection. 4 days post P. yoelii infections in mice, 2 groups of A. coluzzii mosquitoes were fed either on a P. yoelii mono-infected mouse that served as control or on a co-infected mouse. All mosquitoes that were not visibly engorged were removed. Mosquito midguts were dissected and infection status was checked by fluorescence microscopy looking at the oocyst stage 8 days post-feeding.
Successive A. coluzzii infections with T. b. brucei and P. yoelii with and without antibiotics
A. coluzzii females were allowed to feed either on mice mono-infected with T. b. brucei or on naive mice (as negative control). Five days later, all mosquitoes were given a second blood meal on mice infected with P. yoelii. For each feeding, all mosquitoes that were not visibly engorged were removed. Samples were dissected at 8 days after P. yoelii challenge to confirm infection status at the oocyst stage. Throughout the experiment mosquitoes were maintained under antibiotic pressure as previously described [24]. Briefly, immediately following adult emergence, mosquitoes were maintained on a 10% sucrose solution complemented with Penicillin 62.5 μg/mL, Streptomycin 100 μg/mL and gentamicin 50 μg/mL, and this solution was changed every day. Successive co-infections were performed as described above.
Successive A. coluzzii co-infections with T. b. brucei and P. falciparum
Gametocytes from cultured P. falciparum isolate NF54 were produced by the CEPIA mosquito infection facility of the Institut Pasteur, as described previously [23]. A. coluzzii females were allowed to feed either on mice mono-infected with T. b. brucei or on naive mice (as negative control). Five days after feeding, all mosquitoes were given a second blood meal on cultured P. falciparum gametocytes, as described previously [25]. Unfed females were discarded and only fully engorged females were maintained at 26°C and at 70% relative humidity on 10% sucrose solution supplemented with 0.05% para-amino benzoic acid.
Assessing the direct effect of T. b. brucei on the gut microbiome and the reproductive fitness with cultured parasites
The pleomorphic T.b. brucei AnTat1.1E bloodstream forms were cultured in HMI9 medium as described above. In order to obtain stumpy forms, the parasite stage infective to the insect host, an in vitro induction was performed in slender forms cultured at 1.105 parasites/mL with the cAMP analogue 8-pCPT-2′-O-Me-cAMP (Biolog, Germany) at 5μM for 48h [26]. The presence of stumpy forms was verified by using the anti-PAD1 rabbit polyclonal antibody (1:250, Keith Matthews, Edinburgh, UK) targeting the carboxylate-transporter Proteins Associated with Differentiation 1 (PAD1) [27]. Differentiated parasites were centrifuged and resuspended in commercial mechanically defibrinated sheep blood (BCL, France) at 108 parasites/mL in order to feed Anopheles females through a membrane feeding system in which the blood mixture is maintained at 37°C. Mosquito guts were collected 48h and 5 days post-trypanosome ingestion. These experiments were performed to remove any potential confounding factors from the blood of infected mice and be able to attribute results to the direct effect of trypanosomes on the mosquitoes.
Plasmodium infection phenotype
Midguts were dissected 8 days after infection. For P. falciparum, midguts were stained in 0.4% mercurochrome and the number of oocysts was counted under a contrast light microscope (Nikon Eclipse Ni). For GFP-fluorescent P. yoelii (strain GFP@HSP70-GOMO), midgut oocysts were directly counted under a fluorescent binocular stereo microscope (Nikon SMZ18), using a GFP filter (450-500nm for absorbance spectra; 510 nm for emission). Prevalence of infection is defined as the proportion of infected mosquitoes among the total number of dissected mosquitoes. Infection intensity is defined as the number of oocysts per mosquito, determined using only those mosquitoes harbouring at least one oocyst. Uninfected mosquitoes were excluded from the analysis of infection intensity.
Quantitative polymerase chain reaction (qPCR)
Using cDNA or genomic DNA, all qPCRs were performed as described in [25], using SYBR green supermix (KAPA SYBR FAST ABI, from Sigma-Aldrich) and the CFX96 Touch Real-Time PCR Detection System (from Biorad). Ribosomal protein AgrpS7 gene (F_5’-CACCGCCGTGTACGATGCCA-3’ and R_5’-ATGGTGGTCTGCTGGTTCTT-3’) was used as an internal control and the spliced-leader (SL) RNA (F_5’-CAATATAGTACAGAAACTG-3’ and R_ 5’-AACTAACGCTATTATTAGAA-3’) was used to confirm the infection status of each sample by T. b. brucei. The quantification of each gene was obtained as a ratio of the AgrpS7. Analysis of the expression of transcript relative to rpS7 was performed according to the 2−ΔΔCt method [28]. PCR condition run are: 95°C for 5min, then 40 cycle of (95°C for 15 sec, 60°C for 1min (plateread)), 60°C for 30 sec.
Microbiota analysis by qPCR
Dissection was performed as described previously [29]. Briefly, before dissection, mosquitoes were washed in 75% ethanol for 5 min, and washed three times in sterile PBS to wash out non-attached bacteria, thus preventing sample contamination with cuticle bacteria during dissection. For each biological replicate, 20 midguts were collected from each mosquito group (-Tryp) and (+Tryp), frozen immediately on dry ice and stored at −80°C until processing. To assess antibiotic effectiveness, 20 mosquito midguts from each group (-Tryp) and (+Tryp) were collected 48h post-feeding. Midguts were excluded from the analysis when they appeared to have burst, resulting in a substantial loss of the gut content. DNA was extracted with DNeasy PowerSoil Kit (QIAGEN). The V4 region of the 16S rDNA (16S_V4q_F: 5’-GTGCCAGCMGCCGCGGTAA-3’ and 16S_V4q_R: 5’-GGACTACHVGGGTWTCTAAT -3’) was used for measuring the total bacterial abundance by quantitative PCR. For Enterobacteriaceae family detection, we used two different primer pairs: 16SrDNA [30] (Entero_16S_F: 5’ CGTCGCAAGMMCAAAGAG 3’- and Entero_16S_R: 5’TTACCGCGGCTGCTGGCAC3’) and 23SrDNA (-Entero_23S_F: 5’-TGCCGTAACTTCGGGAGAAGGCA-3’ and Entero_23S_R: 5’ -TCAAGGACCAGTGTTCAGTGTC- 3’) [31,32]. DNA samples from each independent biological replicate were used to perform distinct qPCR in triplicate and fold changes obtained between (-Tryp) and (+Tryp) were combined as a mean and illustrated graphically.
Effect of trypanosome ingestion on the reproductive fitness
A. coluzzii vitellogenin expression with and without antibiotics.
Mosquitoes were collected at day 2 post feeding on Trypanosoma-infected or naive mice. For mosquitoes treated with antibiotics, the protocol described above was used. Total RNA was extracted with TRIzol reagent (Invitrogen) from pools of 10 mosquitoes from each batch. cDNAs were generated using M-MLV reverse transcriptase (Invitrogen) from total RNAs. qPCR was performed to quantify differences in vitellogenin gene expression between A. coluzzii mosquitoes fed on Trypanosoma-infected mice and naive mice or between mosquitoes co-infected with T. b. brucei/P. yoelii and those infected only with P. yoelii. The primers Vg_qF: 5’-CCGACTACGACCAGGACTTCC-3’ and Vg_qR: 5’ CACTGGACGACACGTACGGGC-3’ were used as target for qPCR.
Effect of trypanosome ingestion on fecundity and egg hatching.
Egg laying was assessed for both individual females and for pools of 20 females exposed (+Tryp) or not exposed (-Tryp) to T. brucei brucei. For the pools of females, the number of laid eggs was counted 72h post-blood feeding and divided by the number of females (20 gravid females) to get an average number of eggs laid per female, as a proportion. The differences in these proportions between -Tryp and +Tryp groups were tested using Chi-Squared tests in 3 independent experiments. For the eggs laid by individual females that were placed in individual pot with a single male, the number of eggs was counted individually and differences between the two groups were tested using a non-parametric Wilcoxon signed-rank non-parametric test.
To measure the egg hatching rate without bias, we have randomly collected eggs freshly laid by about 10 individual female “mothers” (<5 eggs per individual female) of both the (-Tryp) and the (+Tryp) groups, put these eggs in multiplate wells containing tap water, stored them at 26°C, and monitored their hatching daily. The numbers of hatched eggs (into larvae) were counted from day 2 to day 4 post-egg laying. The number of hatched eggs from each group (-Tryp or +Tryp) was counted and divided by the number of initial eggs put in water, as a proportion. Differences in these proportions between (-Tryp) and (+Tryp) groups were tested using a Chi-Squared test.
RNA sequencing and analysis
Two batches of Anopheles mosquitoes have been fed on mice, non-infected (control) or mono-infected with red fluorescent T. b. brucei. For each condition, a pool of 10 infected-mosquitoes was collected at 24h and 48h post-feeding. Total RNA extractions from intact mosquitoes were performed on each pool using TRIzol reagent (Invitrogen), yielding a total of 12 samples. All RNA-seq was performed at the University of Minnesota Genomics Center (genomics.umn.edu).
Details for RNA sequencing experiment and analysis
Three independent biological replicates were performed, and for each experimental replicate, two batches of Anopheles mosquitoes have been fed on mice, non-infected (naive feeding, control) or mono-infected with red fluorescent T. b. brucei (Trypanosoma-infectious feeding). For each condition, a pool of 10 infected-mosquitoes was collected at 24h and 48h post-feeding. Total RNA extractions from intact mosquitoes were performed on each pool, yielding 12 samples (see Table 1 above).
Two groups of mosquitoes were fed for one on naive mouse (uninfected control) and for the other on trypanosome-infected mouse. 10 mosquitoes were collected at 24h and 48h post-feeding and used for total RNA extraction. The experiment as reproduced three times (“Experiment 1”, “Experiment 2” and “Experiment 3”).
Briefly, using Illumina’s Truseq RNA Sample Preparation Kit (Cat. # RS-122-2001), 1 microgram of total RNA was oligo-dT purified using oligo-dT coated magnetic beads, fragmented and then reverse transcribed into cDNA. The cDNA was fragmented, blunt-ended, and ligated to indexed (barcoded) adaptors and amplified using 15 cycles of PCR. Final library size distribution was validated using capillary electrophoresis and quantified using fluorimetry (PicoGreen) and via q-PCR. Indexed libraries were then normalized, pooled and then size selected to 320bp +/- 5% using Caliper’s XT instrument. Truseq libraries were hybridized to a paired end flow cell and individual fragments were clonally amplified by bridge amplification on the Illumina cBot. Once clustering was complete, the flow cell was loaded on the HiSeq 2000 and sequenced using Illumina’s SBS chemistry. Primary analysis of sequence reads and demultiplexing were done using CASAVA 1.8.2 and de-multiplexed FASTQ files were used for downstream analyses.
The quality of the raw reads was checked with FastQC version 0.11.5 (http://www.bioinformatics.babraham.ac.uk/projects/fastqc) and multiqc version 0.7 [33] BWA-mem version 0.7.7-r441 (https://arxiv.org/abs/1303.3997) with default parameters was used for alignment against the reference genome of Anopheles gambiae str. PEST version AgamP4 (from VectorBase). Genes were counted using featureCounts [34] version 1.4.6-p3 with the annotation version AgamP4.7 and the parameters–t mRNA–g ID.
Counts data were analyzed using R version 3.3.1 (R Core Team. R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing (2016)) and the Bioconductor package DESeq2 version 1.14.1 [35] using default parameters. A generalized linear model including time (24h and 48h), treatment (non-infected, T. b. brucei infected), hatching effect and the interaction term between time and treatment was applied in order to test for (i) inter-condition differences and (ii) the time-treatment interaction. For each pairwise comparison, raw p-values were adjusted for multiple testing using the Benjamini and Hochberg procedure [36]. Genes with adjusted p-values below 0.05 were considered differentially expressed. Sequence data related to analysis have been submitted in ArrayExpress under this link:
https://www.ebi.ac.uk/arrayexpress/experiments/E-MTAB-7469
ArrayExpress accession: E-MTAB-7469
Username: adrien.pain@pasteur.fr
Password: xwceYPkk
Immunofluorescence assays
Trypanosomes were isolated from mosquito midguts in phosphate buffered saline (PBS) 1X and placed on poly-L-lysine coated slides and fixed in methanol at -20°C for 5 seconds before being re-hydrated in PBS for 10 minutes as previously described [37]. Trypanosoma parasites were co-stained in PBS containing 0.1% bovine serum albumin for 45 minutes at 37°C with the two following antibodies: (1) the anti-CRD rabbit polyclonal antibody (1:300) targeting the cross-reactive determinant of the glycosylphosphatidylinositol anchor of surface membrane proteins, predominantly the variant surface glycoproteins of trypanosome bloodstream forms [38], and (2) the anti-EP mouse IgG1 monoclonal antibody (1:500) (CLP001A, Cedarlane, Canada) targeting the procyclin surface coat of procyclic trypomastigotes. Species-specific secondary antibodies coupled to AlexaFluor594 or 488 (Jackson ImmunoResearch, USA) were then used in PBS containing 0.1% bovine serum albumin for 30 minutes at 37°C. DNA was stained with 4,6-diamidino-2-phenylindole (DAPI) and slides were mounted under cover slips with ProLong antifade reagent (Invitrogen) as previously described [37]. Image acquisition was carried out under a Leica DMI 4000B automated inverted epifluorescence microscope (LEICA, Germany) with a 100x objective and equipped for bright field and phase contrast imaging, using a Prime95B CMOS camera (PhotoMetrix) controlled by Micro-manager 1.4 (NIH). Image intensities were standardized (image displays were set to the same minimum and maximum intensities in order to homogenize their presentation and allow visual comparisons) and analyzed with ImageJ 1.49 (NIH).
Statistical analyses
Differences in infection prevalence and egg hatching rates were statistically tested using Chi-Square tests. For differences in oocyst load (infection intensity) and in the number of eggs per individual females, we used Wilcoxon signed-rank non-parametric tests. Statistical differences in prevalence and intensity were first tested independently for each independent replicate as described above, and p-values were empirically determined using 105 Monte-Carlo permutations. Following independent statistical tests, the p-values from independent tests of significance were combined using the meta-analytical approach of Fisher [39] when the direction of change for each independent replicate was concordant (e.g., each independent replicate displayed higher infection prevalence than their paired -Tryp controls). Statistical analyses were done using R [40]. For qPCR analysis of the expression of transcripts relative to rpS7, the 2−ΔΔCt method was used. Difference in deltaCt distribution across the independent biological replicates between (-Tryp) and (+Tryp) samples was statistically tested using Student t-test.
Results
T. b. brucei differentiates into procyclic forms and survives at least 48h in A. coluzzii midgut
To investigate the survival of T. b. brucei bloodstream forms in mosquitoes, A. coluzzii mosquitoes were allowed to feed on cultured fluorescent T. b. brucei bloodstream forms. Mosquitoes were dissected at day 2 and 5 post-feeding. Results revealed that ingested T. b. brucei parasites survived at least 48h in A. coluzzii midgut for 81.5% of fed females (Table 2). Interestingly, fluorescent midguts were also observed in 2.9% of the mosquitoes five days after trypanosome ingestion (Table 2).
Mosquito midguts were extracted at day 2 and day 5 post-feeding. Results of three independent experiments are shown in the table, where the numbers (No.) of positive midguts (carrying fluorescent T. b. brucei parasites) over the total number of dissected mosquitoes were counted. The prevalence column shows the proportion (in %) of mosquito midguts carrying fluorescent trypanosomes, at day 2 and day 5 post-trypanosomes feeding. The mean of prevalence is shown in bold for both day 2 and day 5 time points. The lower total numbers of mosquitoes at day 5 are not due to increased mosquito mortality but to the sampling design.
Knowing this, we assessed whether wild-type bloodstream stumpy trypanosomes can differentiate into procyclic forms in the mosquitoes. A double immunofluorescence assay was performed on trypanosomes extracted from mosquito midguts at 24h post-ingestion using an antibody specifically targeting the bloodstream Variant Surface Glycoprotein (VSG) coat (anti-CRD) [38] and an anti-procyclin antibody labelling the procyclic surface coat [41] on trypanosomes extracted from mosquito midguts at 24h post-ingestion. While only few parasites were presenting a positive signal for VSGs, we observed a strong procyclin signal on a majority of parasites from midguts collected at 24h (Fig 1). We also performed videos of mosquito midguts dissected 48h post-trypanosome ingestion (S1 Movie) and, in all midguts where the red fluorescence was detected, trypanosomes with the morphology and size of procyclic trypomastigote were observed actively swimming. These results indicate that trypanosomes readily differentiate from stumpy to procyclic stages during the first 24h in the mosquito midgut.
A double immunofluorescence assay was performed using an antibody specifically targeting the bloodstream form VSG’s cross-reactive determinant (anti-CRD in green) and an anti-procyclin antibody labelling the procyclin surface coat (anti-EP in red) on trypanosomes extracted from mosquito midguts at 24h post-ingestion. Each image line illustrates a distinct microscopic field. Whereas only a few parasites were positive for VSGs alone, a strong procyclin signal was observed for a majority of parasites. Some parasites were positive for both antibodies, suggesting an ongoing surface coat switching. Scale bars indicate 10 micrometers.
We hypothesized that a microorganism viable for at least 48h and with such motility in the mosquito midgut might significantly affect the midgut micro-environment and possibly impact i) the gut bacterial flora, ii) the mosquito immune response and/or iii) the mosquito fitness. Therefore, we successively tested these three hypotheses.
Trypanosoma ingestion increases the mosquito gut microbiota abundance
We monitored the bacterial load in the mosquito midgut at day 2 and day 5 post-feeding on Trypanosoma-infected mice or naive mice (control group). Quantitative PCR (qPCR) targeting the 16S ribosomal RNA gene (16S rDNA) revealed a large proliferation of bacteria in mosquitoes exposed to T. b. brucei, with a 5-fold increase of the total enteric bacterial population abundance at day 2 and day 5 after the blood meal (Fig 2A). Similar increases in the microbiome at both times points were observed when mosquitoes were allowed to feed on cultured trypanosomes added to commercial sheep blood as compared to sheep blood without trypanosomes (Fig 2B). This indicates that the effect we observed on the microbiome when using the mouse infection system (Fig 2A) was not the result of potential mice factors, but instead is due to a direct effect of trypanosome parasites.
The graphs in A and B show median fold change of the total bacteria load in the mosquito midgut using 16S rDNA detection by qPCR and the ribosomal protein rps7 gene as the internal calibrator. qPCR detection was performed at day 2 (D2) and day 5 (D5) post-blood meal. The graph in A shows the effect on the mosquito microbiome from females fed on the mouse infection system. The graph in B shows the effect on the mosquito microbiome from females fed on cultured trypanosomes mixed with sheep blood. -Tryp = mosquitoes fed on blood without trypanosomes; +Tryp = mosquitoes fed on blood containing trypanosomes. The dotted line represents the level of 16S rDNA from -Tryp mosquitoes. The ratio of the normalized 16S rDNA detection in “+Tryp” versus “-Tryp” was calculated using triplicates from the same cDNA dilution. Error bars show median absolute deviation computed by permutation from 3 experiments. *: statistically significant p-value (p<0.05) related to the deltaCt distribution between “+Tryp” and “-Tryp”.
Trypanosoma ingestion weakly modulates Anopheles immune factors
To query the effects of T. b. brucei survival in the mosquito midgut, RNA sequencing was used to identify host transcriptional response to Trypanosoma ingestion at 24h and 48h post-trypanosome ingestion. While no effect was observed in the transcriptional response to trypanosome ingestion at 24h (S2 Table), only 13 genes were significantly differentially expressed at 48 h after a trypanosome-containing bloodmeal as compared to naive bloodmeal (S1 Table). Most of the candidate genes do not have an annotated or predicted function related to immunity, except for two genes: the leucine-rich repeat protein APL1B (AGAP007035) and the 3-glucan binding protein GNBPB1 (AGAP004455) (S1 Table).
Measurement by RT-qPCR in independent RNA samples confirmed that only APL1B expression was increased by two-fold after trypanosome ingestion (S1A Fig), however this expression increase was not statistically significant between (-Tryp) and (+ Tryp) groups. Nevertheless, although this was not statistically significant, the weak induction of APL1B expression after trypanosome exposure was abolished when mosquitoes were treated with antibiotics (S1B Fig), indicating that APL1B expression responds to augmented bacterial abundance rather than to the presence of trypanosomes. Thus, ingestion of trypanosomes modulated very weakly and indirectly the expression of one immune factor through an influence on the enteric flora.
Trypanosoma ingestion affects the reproductive fitness in A. coluzzii
Mosquito egg maturation induces the expression of the gene encoding for the precursor of the major yolk protein, vitellogenin (Vg) through the activation of the 20-hydroxyecdysone (20E) pathway [42–44]. In order to measure the effect of trypanosome ingestion on mosquito reproductive fitness, we measured expression of Vg 48h after feeding on blood with or without trypanosomes. Vg expression decreased significantly in mosquitoes exposed to trypanosomes but was not modified by another eukaryotic microorganism, such as P. yoelii, when it was ingested alone (Fig 3A). To strengthen this finding, we also tested the expression of another lipid transporter, the Lipophorin (Lp), another yolk protein precursor in mosquitoes [45,46], and also controlled via the 20E pathway [44]. We observed that, as for Vg expression, Lp expression is decreased 48h post trypanosome ingestion (S2A Fig).
(A) The graph shows median fold change of vitellogenin expression in mosquitoes fed on mice infected by Trypanosoma (Tryp) or by P. yoelii (Py) or by both Trypanosoma and P. yoelii (Py+Tryp) as compared to those fed on naive mice (doted line). The ribosomal protein rps7 gene was used as an internal calibrator. Vitellogenin expression decreased in mosquitoes that fed on mice mono-infected by T. b. brucei (Tryp) and in those that fed on mice co-infected by both T. b. brucei and P. yoelii (Py+Tryp), as compared to those fed on a naive mouse. The ratio of the normalized vitellogenin expression in Tryp, Py, Py+Tryp versus naive control was computed using triplicates from the same cDNA dilution. Error bars show median absolute deviation computed by permutation from 3 independent replicates. *: Statistically significant p-value (p<0.05) related to the deltaCt distribution between “+Tryp” and “-Tryp”. NS: Non-significant p-value. (B) The graph shows the average number of eggs per female using pools of females exposed (+Tryp) or not (-Tryp) to T. b. brucei in three independent replicates. The table below the graph shows, for each experiment, the number of counted eggs and the number of gravid females exposed (+Tryp) or not (-Tryp) to trypanosome parasites. The p-values obtained from a Chi-square test show that there is no statistical significant effect on egg laying. (C) A comparable, yet refined experiment, was performed with the number of eggs counted from each individual gravid female and the difference between the two groups of females was analysed using a Wilcoxon signed-rank non-parametric tests; n = number of individual females from each group. (D) The graph shows the hatching rate of eggs collected from females exposed (grey bar) or not exposed (black bar) to trypanosome parasites at Day 2 to Day 4 post-egg laying. The p-values obtained from a Chi-squared test show that there is a statistical significant impact on the hatching rates between exposed (grey bar) versus non-exposed (black bar) females; n = number of eggs from each group that were placed in individual wells containing water.
Then we collected eggs from both pooled and individual A. coluzzii females fed on either trypanosome-infected (+Tryp) or naive mice (-Tryp), and we observed that trypanosome ingestion did not affect Anopheles fecundity, defined as the number of laid eggs per females, for either pooled females (Fig 3B) or individual ones (Fig 3C).
Despite no difference in the number of eggs laid, due to the fact that trypanosome ingestion significantly affects Vg and Lp expression (Fig 3A), we also compared the egg-hatching rate between the two mosquito populations, which represents the capacity of laid eggs to develop into larvae, i.e. the female fertility. Eggs from each female group (-Tryp and +Tryp) were collected at 72h post-blood feeding and placed individually in well containing tap water. The number of viable larvae was counted from day 2 to day 4 post-egg laying. The egg-hatching rate was significantly lower in eggs from +Tryp females as compared to eggs from -Tryp females (Fig 3D). This indicates that, while trypanosome ingestion does not affect female fecundity (i.e. the number of eggs laid), it affects the viability of the laid eggs (i.e. female fertility) by altering their capacity to develop into larvae. This is likely due to a functional failure related to the observed decrease of both Vg and Lp expression.
In order to remove any potential confounding factors from the blood of infected mice and be able to attribute results to the direct effect of trypanosomes on the mosquito reproductive fitness, cultured T. b. brucei parasites were mixed with commercial sheep blood (for control, commercial sheep blood without trypanosomes was used) and used to feed mosquitoes. We first assessed Vg and Lp expressions at 48h, and we observed a strong decrease in expression of both genes in the +Tryp group as compared to the -Tryp group (S2B Fig). Then, we collected eggs from individual females to assess their fecundity (number of laid eggs/female) and similarly to the mouse infection system, we did not observe any difference between +Tryp and -Tryp groups in the number of laid eggs (S2C Fig). These results indicate that the effects on both Vg and Lp decreased expressions, and more widely on the mosquito reproductive fitness, are neither due to potential immune or metabolic factors nor to anaemia from trypanosome-infected mice.
Trypanosoma ingestion increases A. coluzzii susceptibility to P. yoelii and P. falciparum
As Trypanosoma ingestion strongly affects the abundance of the enteric microbiome and the expression of Vg in A. coluzzii, which have both been described to modulate Anopheles competence for Plasmodium [47–49], we hypothesized that trypanosome parasites could also impact the development of Plasmodium during either subsequent or concomitant exposure.
Feeding mosquitoes with an initial T. b. brucei infectious blood meal significantly enhanced mosquito susceptibility to P. yoelii delivered in a second bloodmeal five days later, as compared to control mosquitoes that first fed on a naive mouse. The effect was significant for both infection prevalence (p = 0.006) and infection intensity (p = 0.001) (Fig 4A & 4B). When P. falciparum gametocytes were used for the second blood meal, infection prevalence but not intensity was increased in the mosquitoes exposed to trypanosomes in the first bloodmeal (p = 0.007) (Fig 4C).
Panels (A) & (B) show results of infection prevalence and infection intensity for P. yoelii, respectively. Panels (C) & (D) show results of infection prevalence and infection intensity for P. falciparum, respectively. -Tryp = group of mosquitoes previously fed on a naive mouse (without Trypanosoma parasites); +Tryp = group of mosquitoes previously fed on a Trypanosoma-infected mouse. **: Combined p-value <0.01 (Fisher method) from the 3 independent biological replicates obtained for the infection prevalence (p = 0.006 for P. yoelii; p = 0.007 for P. falciparum) and for infection intensity (p = 0.001 for P. yoelii only). n = Total number of dissected mosquitoes. N = number of biological replicates.
Simultaneous exposure of mosquitoes to both P. yoelii and T. b. brucei also influenced the development of Plasmodium, similarly, although weakly, to successive exposure (infection prevalence, p = 0.07 and infection intensity, p = 0.010) (Fig 5).
The simultaneous presence of T. b. brucei and P. yoelii in A. coluzzii does not promote Plasmodium infection prevalence (A) but promotes infection intensity (B). In (A), red colour shows proportion of infected and green show proportion of uninfected individuals. -Tryp = group of mosquitoes fed on a P. yoelii infected mouse (but without Trypanosoma parasites); +Tryp = group of mosquitoes fed on a mouse co-infected with T. b. brucei and P. yoelii parasites. *: Combined p-value <0.05 (Fisher method) from the 3 independent biological replicates obtained for the infection intensity (p = 0.010). n = Total number of dissected mosquitoes. N = number of biological replicates.
In order to control for the possibility that potential immunomodulatory factors present in the bloodmeal from trypanosome-infected mice could be responsible for the increased mosquito susceptibility to Plasmodium, we fed mosquitoes on in vitro cultured T. b. brucei before challenging with P. yoelii. Sample sizes and statistical power were small because mosquitoes were weakly attracted by the medium; nevertheless, we observed a similar tendency of increased P. yoelii infection prevalence after exposure to trypanosomes (p-value = 0.07, S3A Fig). This result suggests that the agonistic effect of T. b. brucei ingestion on A. coluzzii vector competence was probably not due to mouse serum factors, but rather to the presence of trypanosomes in the mosquito midgut.
Altered microbiome mediates the trypanosome enhancement of Plasmodium infection, but not vitellogenin decrease
An increase in midgut of Enterobacteriaceae has been positively correlated with Anopheles vector competence for Plasmodium [47]. We assessed by qPCR the level of Enterobacteriaceae in mosquitoes exposed to trypanosomes by targeting both the 16SrDNA and the 23SrDNA to increase the robustness of the measure. Consistent with the enhancement of Plasmodium infection phenotype, we found an increased abundance of Enterobacteriaceae as measured by both 16S and 23S in midguts of trypanosome-carrying mosquitoes (S4A Fig).
As the relevant bacteria were antibiotic sensitive in treated mosquitoes (S4B Fig), we queried the effect of such treatment on the phenotype for Plasmodium susceptibility. Treatment of mosquitoes with antibiotics abolished the increased susceptibility of Anopheles to Plasmodium observed after trypanosome ingestion (Fig 6), indicating that the increased Plasmodium susceptibility was dependent upon the bacterial expansion.
-Tryp = group of mosquitoes previously fed on a naive mouse (without Trypanosoma parasites); +Tryp = group of mosquitoes previously fed on a mouse infected with Trypanosoma parasites. Normal sugar = sugar without antibiotics; AB_sugar = sugar supplemented with antibiotics; **: Combined p-value <0.01 (Fisher method) from the 2 independent biological replicates obtained for the infection intensity (p = 0.007). NS: statistically not significant p-value (p = 0.201). n = Total number of dissected mosquitoes. Prev: infection prevalence in %.
However, the observed decrease of vitellogenin expression in mosquitoes exposed to trypanosomes was not influenced by antibiotic treatment (Fig 7), indicating that the trypanosome effect on mosquito reproductive fitness was independent of the microbiome.
(A) Vitellogenin (Vg) expression level was quantified by qPCR in “Normal sugar” and “Antibiotic sugar” backgrounds from mosquito samples fed on a naive mouse (-Tryp) or on a Trypanosoma-infected mouse (+Tryp). Each bar of the graph shows median fold change of vitellogenin expression in mosquitoes fed on mice infected by Trypanosoma (+Tryp) as compared to those fed on naive mice (doted line). The ribosomal protein rps7 gene was used as an internal calibrator. The ratio of the normalized Vg expression in “+Tryp” versus “-Tryp” was calculated using triplicates from the same cDNA dilution. *: Statistically significant p-value (p<0.05) related to the deltaCt distribution between “+Tryp” and “-Tryp”. (B) Antibiotic efficiency on the bacterial abundance was verified by measuring the abundance of the bacteria (by qPCR detection of 16S rDNA) between “Antibiotic sugar” and “Normal sugar” (doted line) backgrounds 24h post-blood meal. The ratio of the normalized 16S rDNA detection in “Normal sugar” versus “Antibiotic sugar” backgrounds was computed using triplicates from the same cDNA dilution. In A & B, error bars show median absolute deviation computed by permutation from 3 independent biological experiments.
Discussion
Human and animal trypanosome infections are often sympatric with malaria in Africa, and malaria vector mosquitoes can be exposed to both microbes in the same or successive bloodmeals. Although mosquitoes are not the natural hosts of the T. brucei cyclical development, we found that T. b. brucei parasites can transiently survive in the A. coluzzii midgut for at least 48h post-ingestion and differentiate into procyclic forms, before finally dying and being cleared by the mosquitoes. Strikingly, during this short period of time, the presence of trypanosomes alters the mosquito enteric microbiome, decreases the mosquito reproductive fitness via effects on female fertility, and enhances their susceptibility to Plasmodium infection. While the effect of trypanosomes ingestion on Plasmodium infection is microbiome dependent, the phenotype on their reproductive fitness is microbiome independent.
The vector microbiome is a major factor likely influencing parasite transmission through different mechanisms [3,47,50]. The presence of antibiotics in the mosquito midgut could increase its susceptibility to Plasmodium infection, highlighting beneficial effects of the microbiota on the mosquito resistance to Plasmodium infection [29,51]. In contrast, an imbalance in Enterobacteriaceae can promote Anopheles vector competence for Plasmodium [47]. Here, we show that the presence of live trypanosomes increases the abundance of midgut bacteria in Anopheles, which in turn increases vector competence for Plasmodium. In addition, we also demonstrated that Enterobacteriaceae, which were sensitive to the antibiotics treatment that abolished the Plasmodium susceptibility phenotype, were also elevated.
However, additional biological replicates might be necessary to increase the robustness of these findings related to the effect of trypanosome ingestion on the bacterial microbiome. In addition, some in-depth metagenomic analysis would be necessary to identify the specific bacterial taxa underlying the observed phenotypic effect. The mechanism leading to the gut bacterial expansion is not consistent with a resource competition between trypanosomes and the bacteria, as this would lead to an inverse phenotype. Two hypotheses could be proposed: i) trypanosomes could modify the gut environment in such a way that it would favour the development of some bacteria. ii) the immune factors controlling the gut homeostasis could be diverted towards the trypanosomes, which would lead to the gut bacteria expansion.
Bloodmeals containing trypanosomes reduce mosquito reproductive fitness, as measured both by vitellogenin expression and the egg hatching rate of the offspring. Although the increase in mosquito susceptibility to malaria parasite infection in the presence of trypanosomes was dependent on the midgut bacterial expansion, this was not the case for the decreased Vg expression. Reduction of host fecundity has been reported in other parasitized insects [52]. However, it is not clear whether host fecundity reduction reflects parasite manipulation of the vector in order to increase contacts with new hosts and promote its own transmission or nutrient competition with the insect host, or both [53–55]. The induction of the Vitellogenin and Lipophorin through the 20E pathway is triggered by amino acids and nutrients made during the blood digestion process [42,49,44]. The presence of active trypanosomes in the midgut during at least the first 48h following infection could act as a host competitor, and therefore reduce the amount of nutrients available for Vg and Lp induction through the 20E pathway. In other words, Trypanosoma survival in the mosquito midgut might alter the activation of the 20E pathway and thus the expression of downstream factors such as Vg and Lp, which in turn would impact on the reproductive fitness. Nevertheless, although an alteration of the 20E pathway has been previously correlated with a decreased susceptibility to Plasmodium [44], we found that the presence of trypanosomes induces a decrease of Vg and Lp expressions but an increase of the Anopheles vector competence for Plasmodium. Therefore, the effect of trypanosomes on Anopheles vector competence would be mostly related to their impact on the microbiome rather than on the 20E pathway. In Glossina, the activation of immune responses by immunogenic wild type trypanosome strains reduces reproductive output as well as the milk gland protein levels (required for larval development), while infections with non-immunogenic trypanosomes do not [56], revealing a balanced interplay between immune activity and reproduction in tsetse flies. In Anopheles, few studies have highlighted a potential trade-off between immunity and reproduction. In the mosquitoes Anopheles stephensi and Anopheles gambiae, Plasmodium yoelii nigeriensis was shown to influence the balance between reproductive fitness and immune defences, which was linked to an immune pathway inducing apoptosis in ovarian follicle cells [57,58]. Furthermore, in A. gambiae, the low Plasmodium survival phenotype associated with the Lp/Vg knockdowns relies on an anti-plasmodial TEP1-related mechanism [49]. Considering our current data, it is tempting to postulate that T. b. brucei could modulate the interplay between immunity and reproduction in A. coluzzii. However, our results showed that T. b. brucei only weakly modulated immune factors and that the increased susceptibility to Plasmodium was rather dependent on the gut bacterial flora than on a direct effect of trypanosome cells. However, in contrast to the vector competence phenotype, the effects on the mosquito reproductive fitness triggered by T. b. brucei were not dependent on the gut microbiota. Hence these two phenotypes are likely the result of distinct, as yet undescribed, mechanisms.
Given the sympatric distribution of malaria and trypanosomiases, it would be interesting to assess whether these experimentally trypanosome-induced effects on laboratory mosquito strains could also occur in nature, with some potential consequences for the epidemiology of malaria. Although T. brucei is a common parasite and sympatric with P. falciparum in a number of African regions, mosquitoes do not successfully transmit it; nevertheless, they could be exposed to it. More information would be necessary to determine the complex relationships between multiple parasite infections and interpret transmission dynamics in nature. Other less studied protists that are not necessarily human pathogens but rather insect parasites, such as the monoxenic parasite Crithidia fasciculata for instance, are probably as abundant in nature, and may also affect mosquito biology in a similar manner. In total, adult Anopheles female mosquitoes are likely to be exposed to a large panel of microorganisms, bacteria and viruses that would alter their vector competence in distinct ways [1]. Overall, our results emphasize the potential influence of eukaryotic organisms in combination with the bacterial microbiome on Anopheles vector competence for Plasmodium.
Supporting information
S1 Fig. APL1B overexpression 48h post-trypanosome ingestion depends on the midgut bacterial expansion.
(A) Relative quantification of APL1B and GNBPB1 gene expression in females fed on trypanosome-infected mouse (+Tryp) or on naive mouse (-Tryp) and maintained on sucrose, using expression of the ribosomal protein rps7 gene as an internal calibrator. The dotted line represents the median expression in the control (-Tryp). (B) Relative quantification of APL1B gene expression in females fed on trypanosome-infected mouse (+Tryp) or on naive mouse (-Tryp) and maintained on sucrose supplemented with an antibiotic cocktail, using expression of the ribosomal protein rps7 gene as the internal calibrator. In A and B, the ratio of the normalized gene of interest in (+Tryp) versus (-Tryp) control was computed using triplicates from the same cDNA dilution. Error bars show median absolute deviation computed by permutation from 3 technical replicates for each independent biological replicate (EXP1 and EXP2).
https://doi.org/10.1371/journal.pntd.0008059.s001
(TIF)
S2 Fig. Trypanosome parasites displayed a direct effect on Anopheles reproductive fitness.
A & B. Relative quantification of Lp and Vg gene expression in females fed on trypanosome-containing blood (+Tryp) or on blood without trypanosomes (-Tryp), using expression of the ribosomal protein rps7 gene as the internal calibrator. The dotted line represents the median expression in the “-Tryp” control group. The graph in A shows qPCR results from the mouse infection system, while the graph in B shows qPCR from cultured trypanosomes mixed with sheep blood. *: Statistically significant p-value (p<0.05) related to the deltaCt distribution between “+Tryp” and “-Tryp” across 3 independent biological replicates. C. The graph shows the number of laid eggs per individual females fed on cultured trypanosomes mixed with sheep blood. The differences between the two groups of females (+Tryp) versus (-Tryp) was analysed using a Wilcoxon signed-rank non-parametric test; n = number of individual females from each group.
https://doi.org/10.1371/journal.pntd.0008059.s002
(TIF)
S3 Fig. Ingestion of cultured T. b. brucei increases the infection prevalence of A. coluzzii to P. yoelii.
Panel A shows results of infection prevalence. Red colour shows proportion of infected and green shows proportion of uninfected individuals. Panel B shows result of infection intensity. -Tryp = group of mosquitoes previously fed with culture medium only (without Trypanosoma parasites); +Tryp = Group of mosquitoes fed with culture medium containing T. b. brucei. N = number of biological replicates. Combined p-value was done using Fisher method from the 3 independent biological replicates. n = Total number of dissected mosquitoes.
https://doi.org/10.1371/journal.pntd.0008059.s003
(TIF)
S4 Fig. Trypanosoma ingestion increases the abundance of Enterobacteriaceae family in A. coluzzii at day 5 post-feeding.
(A) 16S and 23S rDNA detection of Enterobacteriaceae was performed by qPCR at day 5 (D5) post-blood meal using the expression of the ribosomal protein rps7 gene as the internal calibrator. The two couple of primers (16S and 23S were both used to increase the robustness of the results. The graph shows median fold change of the Enterobacteriaceae load in midguts of mosquitoes challenged with trypanosome as compared to mosquito fed on naive mouse (doted line). With Tryp = group of mosquitoes previously fed on a Trypanosoma-infected mouse. The ratio of the normalized 16S (or 23S) rDNA detection in “With Tryp” versus “Naive” was computed using triplicates from the same cDNA dilution. Error bars show median absolute deviation computed by permutation from 3 experiments. *: Statistically significant p-value (p<0.05) related to the deltaCt distribution between “+Tryp” and “Naive”. NS: Non-significant p-value. (B) Antibiotic efficiency on Enterobacteriaceae family. 16S and 23S rDNA detection of Enterobateriaceae was performed by qPCR at day 5 (D5) post-naive blood meal using sample from mosquitoes treated or not with antibiotics. Expression of the ribosomal protein rps7 gene was used as the internal calibrator. The dotted line represents the level of 16S and 23S rDNA in the normal sugar background (without antibiotic). The ratio of the normalized 16S (or 23S) rDNA detection in “AB sugar” (with antibiotic) versus “Normal sugar” (without antibiotics) was computed using triplicates from the same cDNA dilution. Error bars show median absolute deviation computed by permutation from 3 independent biological experiments.
https://doi.org/10.1371/journal.pntd.0008059.s004
(TIF)
S1 Movie. Procyclic-like trypanosomes swimming out of a mosquito midgut 48h post-ingestion.
Anopheles midguts were dissected in PBS 48h post-trypanosome ingestion and scrutinized under a microscope at the 100x magnification. Trypanosomes with a procyclic trypomastigote shape were found to be highly motile in all midguts.
https://doi.org/10.1371/journal.pntd.0008059.s005
(MPEG)
S1 Table. RNAseq identification of 13 genes, which expressions are modulated 48h after T. b. brucei ingestion.
Among these 13 genes, two are immune–like genes: 3-Glucan binding protein and APL1B (highlighted in pink). Results come from three independent biological experiments. For each gene, differential expressions between T. b. brucei-infected and naive feeding backgrounds are expressed in Log2 fold change, p-values and adjusted p-values (padj) to highlight statistical difference.
https://doi.org/10.1371/journal.pntd.0008059.s006
(XLSX)
S2 Table. RNAseq analysis at 24h after T. b. brucei ingestion.
For each gene, differential expressions between T. b. brucei-infected and naive feeding backgrounds are expressed in Log2 fold change, p-values and adjusted p-values (padj) to highlight statistical difference. The count values are noted for each gene and for each three biological replicates: NaiveMouse_1 = mosquito fed on naive mouse for the first biological replicate; TrypaMouse_1 = mosquito fed on T. b. brucei-infected mouse for the first biological replicate.
https://doi.org/10.1371/journal.pntd.0008059.s007
(XLS)
S3 Table. RNAseq analysis at 48h after T. b. brucei ingestion.
For each gene, differential expressions between T. b. brucei-infected and naive feeding backgrounds are expressed in Log2 fold change, p-values and adjusted p-values (padj) to highlight statistical difference. The count values are noted for each gene and for each three biological replicates: NaiveMouse_1 = mosquito fed on naive mouse for the first biological replicate; TrypaMouse_1 = mosquito fed on T. b. brucei-infected mouse for the first biological replicate.
https://doi.org/10.1371/journal.pntd.0008059.s008
(XLS)
Acknowledgments
We thank the Institut Pasteur core facility, the CEnter for the Production and Infection of Anopheles (CEPIA), for providing Plasmodium falciparum infected mosquitoes and the facilities for mosquito infection experiments with rodent parasites. We are grateful to Dr. Olivier Sylvie from the "Molecular biology and immunology of malaria liver infection" unit (Inserm, France) for the GFP-transgenic P. yoelii strain, GFP@HSP70-GOMO. We acknowledge J. Bangs and K. Matthews for providing various antibodies.
References
- 1. Dieme C, Rotureau B, Mitri C (2017) Microbial Pre-exposure and Vectorial Competence of Anopheles Mosquitoes. Front Cell Infect Microbiol 7: 508. pmid:29376030
- 2. Kramer LD (2016) Complexity of virus-vector interactions. Curr Opin Virol 21: 81–86. pmid:27580489
- 3. Belda E, Coulibaly B, Fofana A, Beavogui AH, Traore SF, et al. (2017) Preferential suppression of Anopheles gambiae host sequences allows detection of the mosquito eukaryotic microbiome. Sci Rep 7: 3241. pmid:28607435
- 4. Medeiros MC, Anderson TK, Higashiguchi JM, Kitron UD, Walker ED, et al. (2014) An inverse association between West Nile virus serostatus and avian malaria infection status. Parasit Vectors 7: 415. pmid:25178911
- 5. Lefevre T, Vantaux A, Dabire KR, Mouline K, Cohuet A (2013) Non-genetic determinants of mosquito competence for malaria parasites. PLoS Pathog 9: e1003365. pmid:23818841
- 6. Mitri C, Vernick KD (2012) Anopheles gambiae pathogen susceptibility: the intersection of genetics, immunity and ecology. Curr Opin Microbiol 15: 285–291. pmid:22538050
- 7. Mitchell SN, Kakani EG, South A, Howell PI, Waterhouse RM, et al. (2015) Mosquito biology. Evolution of sexual traits influencing vectorial capacity in anopheline mosquitoes. Science 347: 985–988. pmid:25722409
- 8. Buscher P, Cecchi G, Jamonneau V, Priotto G (2017) Human African trypanosomiasis. Lancet.
- 9. Stone CM, Chitnis N (2015) Implications of Heterogeneous Biting Exposure and Animal Hosts on Trypanosomiasis brucei gambiense Transmission and Control. PLoS Comput Biol 11: e1004514. pmid:26426854
- 10. Rotureau B, Van Den Abbeele J (2013) Through the dark continent: African trypanosome development in the tsetse fly. Front Cell Infect Microbiol 3: 53. pmid:24066283
- 11. Franco JR, Simarro PP, Diarra A, Jannin JG (2014) Epidemiology of human African trypanosomiasis. Clin Epidemiol 6: 257–275. pmid:25125985
- 12.
WHO_Fact_Sheet (February 2018) http://www.who.int/en/news-room/fact-sheets/detail/trypanosomiasis-human-african-(sleeping-sickness).
- 13. Informal Expert Group on Gambiense HATR, Buscher P, Bart JM, Boelaert M, Bucheton B, et al. (2018) Do Cryptic Reservoirs Threaten Gambiense-Sleeping Sickness Elimination? Trends Parasitol 34: 197–207. pmid:29396200
- 14. Capewell P, Cren-Travaille C, Marchesi F, Johnston P, Clucas C, et al. (2016) The skin is a significant but overlooked anatomical reservoir for vector-borne African trypanosomes. Elife 5.
- 15. Kagira JM, Maina N, Njenga J, Karanja SM, Karori SM, et al. (2011) Prevalence and types of coinfections in sleeping sickness patients in kenya (2000/2009). J Trop Med 2011: 248914. pmid:21915184
- 16. N'Djetchi MK, Ilboudo H, Koffi M, Kabore J, Kabore JW, et al. (2017) The study of trypanosome species circulating in domestic animals in two human African trypanosomiasis foci of Cote d'Ivoire identifies pigs and cattle as potential reservoirs of Trypanosoma brucei gambiense. PLoS Negl Trop Dis 11: e0005993. pmid:29045405
- 17. Diatta M, Spiegel A, Lochouarn L, Fontenille D (1998) Similar feeding preferences of Anopheles gambiae and A. arabiensis in Senegal. Trans R Soc Trop Med Hyg 92: 270–272. pmid:9861393
- 18. Ngom el HM NJ, Ba Y, Konaté L, Faye O, Diallo M, Dia I. (2013 ) Spatio-temporal analysis of host preferences and feeding patterns of malaria vectors in the sylvo-pastoral area of Senegal: impact of landscape classes. Parasit Vectors 19;6(1):332.
- 19. Redmond SN, Eiglmeier K, Mitri C, Markianos K, Guelbeogo WM, et al. (2015) Association mapping by pooled sequencing identifies TOLL 11 as a protective factor against Plasmodium falciparum in Anopheles gambiae. BMC Genomics 16: 779. pmid:26462916
- 20. Calvo-Alvarez E, Cren-Travaille C, Crouzols A, Rotureau B (2018) A new chimeric triple reporter fusion protein as a tool for in vitro and in vivo multimodal imaging to monitor the development of African trypanosomes and Leishmania parasites. Infect Genet Evol.
- 21. Manzoni G, Briquet S, Risco-Castillo V, Gaultier C, Topcu S, et al. (2014) A rapid and robust selection procedure for generating drug-selectable marker-free recombinant malaria parasites. Sci Rep 4: 4760. pmid:24755823
- 22. Bhattacharyya MK, Kumar N (2001) Effect of xanthurenic acid on infectivity of Plasmodium falciparum to Anopheles stephensi. Int J Parasitol 31: 1129–1133. pmid:11429178
- 23. Mitri C, Jacques JC, Thiery I, Riehle MM, Xu J, et al. (2009) Fine pathogen discrimination within the APL1 gene family protects Anopheles gambiae against human and rodent malaria species. PLoS Pathog 5: e1000576. pmid:19750215
- 24. Carissimo G, Pondeville E, McFarlane M, Dietrich I, Mitri C, et al. (2015) Antiviral immunity of Anopheles gambiae is highly compartmentalized, with distinct roles for RNA interference and gut microbiota. Proc Natl Acad Sci U S A 112: E176–185. pmid:25548172
- 25. Mitri C, Bischoff E, Takashima E, Williams M, Eiglmeier K, et al. (2015) An Evolution-Based Screen for Genetic Differentiation between Anopheles Sister Taxa Enriches for Detection of Functional Immune Factors. PLoS Pathog 11: e1005306. pmid:26633695
- 26. Barquilla A, Saldivia M, Diaz R, Bart JM, Vidal I, et al. (2012) Third target of rapamycin complex negatively regulates development of quiescence in Trypanosoma brucei. Proc Natl Acad Sci U S A 109: 14399–14404. pmid:22908264
- 27. Dean S, Marchetti R, Kirk K, Matthews KR (2009) A surface transporter family conveys the trypanosome differentiation signal. Nature 459: 213–217. pmid:19444208
- 28. Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25: 402–408. pmid:11846609
- 29. Gendrin M, Rodgers FH, Yerbanga RS, Ouedraogo JB, Basanez MG, et al. (2015) Antibiotics in ingested human blood affect the mosquito microbiota and capacity to transmit malaria. Nat Commun 6: 5921. pmid:25562286
- 30. Hermann-Bank ML, Skovgaard K, Stockmarr A, Larsen N, Molbak L (2013) The Gut Microbiotassay: a high-throughput qPCR approach combinable with next generation sequencing to study gut microbial diversity. BMC Genomics 14: 788. pmid:24225361
- 31. Matsuda K, Tsuji H, Asahara T, Kado Y, Nomoto K (2007) Sensitive quantitative detection of commensal bacteria by rRNA-targeted reverse transcription-PCR. Appl Environ Microbiol 73: 32–39. pmid:17071791
- 32. Kurakawa T, Kubota H, Tsuji H, Matsuda K, Takahashi T, et al. (2013) Intestinal Enterobacteriaceae and Escherichia coli populations in Japanese adults demonstrated by the reverse transcription-quantitative PCR and the clone library analyses. J Microbiol Methods 92: 213–219. pmid:23262033
- 33. Ewels P, Magnusson M, Lundin S, Kaller M (2016) MultiQC: summarize analysis results for multiple tools and samples in a single report. Bioinformatics 32: 3047–3048. pmid:27312411
- 34. Liao Y, Smyth GK, Shi W (2014) featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30: 923–930. pmid:24227677
- 35. Love MI, Huber W, Anders S (2014) Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15: 550. pmid:25516281
- 36. Benjamini Y, Hochberg Y (1995) Controlling the False Discovery Rate—a Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society Series B-Methodological 57: 289–300.
- 37. Rotureau B, Subota I, Buisson J, Bastin P (2012) A new asymmetric division contributes to the continuous production of infective trypanosomes in the tsetse fly. Development 139: 1842–1850. pmid:22491946
- 38. Zamze SE, Ferguson MA, Collins R, Dwek RA, Rademacher TW (1988) Characterization of the cross-reacting determinant (CRD) of the glycosyl-phosphatidylinositol membrane anchor of Trypanosoma brucei variant surface glycoprotein. Eur J Biochem 176: 527–534. pmid:2458923
- 39.
Fisher RA (1925) Statistical Methods for Research Workers. Edinburgh: Oliver & Boyd. 356 p.
- 40.
R Core Team (2012) R: A language and environment for statistical computing. In: R Foundation for Statistical Computing, editor. Vienna, Austria.
- 41. Richardson JP, Jenni L, Beecroft RP, Pearson TW (1986) Procyclic tsetse fly midgut forms and culture forms of African trypanosomes share stage- and species-specific surface antigens identified by monoclonal antibodies. J Immunol 136: 2259–2264. pmid:3512712
- 42. Hansen IA, Attardo GM, Park JH, Peng Q, Raikhel AS (2004) Target of rapamycin-mediated amino acid signaling in mosquito anautogeny. Proceedings of the National Academy of Sciences of the United States of America 101: 10626–10631. pmid:15229322
- 43. Bai H, Gelman DB, Palli SR (2010) Mode of action of methoprene in affecting female reproduction in the African malaria mosquito, Anopheles gambiae. Pest Manag Sci 66: 936–943. pmid:20730984
- 44. Werling K, Shaw WR, Itoe MA, Westervelt KA, Marcenac P, et al. (2019) Steroid Hormone Function Controls Non-competitive Plasmodium Development in Anopheles. Cell 177: 315–325 e314. pmid:30929905
- 45. Raikhel AS, Dhadialla TS (1992) Accumulation of yolk proteins in insect oocytes. Annu Rev Entomol 37: 217–251. pmid:1311540
- 46. Sun J, Hiraoka T, Dittmer NT, Cho KH, Raikhel AS (2000) Lipophorin as a yolk protein precursor in the mosquito, Aedes aegypti. Insect Biochem Mol Biol 30: 1161–1171. pmid:11044662
- 47. Boissiere A, Tchioffo MT, Bachar D, Abate L, Marie A, et al. (2012) Midgut microbiota of the malaria mosquito vector Anopheles gambiae and interactions with Plasmodium falciparum infection. PLoS Pathog 8: e1002742. pmid:22693451
- 48. Tchioffo MT, Boissiere A, Churcher TS, Abate L, Gimonneau G, et al. (2013) Modulation of malaria infection in Anopheles gambiae mosquitoes exposed to natural midgut bacteria. PLoS One 8: e81663. pmid:24324714
- 49. Rono MK, Whitten MM, Oulad-Abdelghani M, Levashina EA, Marois E (2010) The major yolk protein vitellogenin interferes with the anti-plasmodium response in the malaria mosquito Anopheles gambiae. PLoS Biol 8: e1000434. pmid:20652016
- 50. Dennison NJ, Jupatanakul N, Dimopoulos G (2014) The mosquito microbiota influences vector competence for human pathogens. Curr Opin Insect Sci 3: 6–13. pmid:25584199
- 51. Dong Y, Manfredini F, Dimopoulos G (2009) Implication of the mosquito midgut microbiota in the defense against malaria parasites. PLoS Pathog 5: e1000423. pmid:19424427
- 52. Hurd H (2003) Manipulation of medically important insect vectors by their parasites. Annu Rev Entomol 48: 141–161. pmid:12414739
- 53. Brandt SM, Schneider DS (2007) Bacterial infection of fly ovaries reduces egg production and induces local hemocyte activation. Dev Comp Immunol 31: 1121–1130. pmid:17400292
- 54. Mitchell SN, Catteruccia F (2017) Anopheline Reproductive Biology: Impacts on Vectorial Capacity and Potential Avenues for Malaria Control. Cold Spring Harb Perspect Med 7.
- 55. Rogers ME, Bates PA (2007) Leishmania manipulation of sand fly feeding behavior results in enhanced transmission. PLoS Pathog 3: e91. pmid:17604451
- 56. Hu C, Rio RV, Medlock J, Haines LR, Nayduch D, et al. (2008) Infections with immunogenic trypanosomes reduce tsetse reproductive fitness: potential impact of different parasite strains on vector population structure. PLoS Negl Trop Dis 2: e192. pmid:18335067
- 57. Ahmed AM, Hurd H (2006) Immune stimulation and malaria infection impose reproductive costs in Anopheles gambiae via follicular apoptosis. Microbes Infect 8: 308–315. pmid:16213176
- 58. Hopwood JA, Ahmed AM, Polwart A, Williams GT, Hurd H (2001) Malaria-induced apoptosis in mosquito ovaries: a mechanism to control vector egg production. J Exp Biol 204: 2773–2780. pmid:11683433