An insight into the female and male Sabethes cyaneus mosquito salivary glands transcriptome

Abstract

Mosquitoes are responsible for the death and debilitation of millions of people every year due to the pathogens they can transmit while blood feeding. While a handful of mosquitoes, namely those in the Aedes, Anopheles, and Culex genus, are the dominant vectors, many other species belonging to different genus are also involved in various pathogen cycles. Sabethes cyaneus is one of the many poorly understood mosquito species involved in the sylvatic cycle of Yellow Fever Virus. Here, we report the expression profile differences between male and female of Sa. cyaneus salivary glands (SGs). We find that female Sa. cyaneus SGs have 165 up-regulated and 18 down-regulated genes compared to male SGs. Most of the up-regulated genes have unknown functions, however, odorant binding proteins, such as those in the D7 protein family, and mucins were among the top 30 genes. We also performed various in vitro activity assays of female SGs. In the activity analysis we found that female SG extracts inhibit coagulation by blocking factor Xa and has endonuclease activity. Knowledge about mosquitoes and their physiology are important for understanding how different species differ in their ability to feed on and transmits pathogens to humans. These results provide us with an insight into the Sabethes SG activity and gene expression that expands our understanding of mosquito salivary glands.

Graphical Abstract

1. INTRODUCTION

Mosquitoes are regarded as the most dangerous animals on earth because of the many pathogens they transmit which are responsible for the death and debilitation of millions of people every year [1]. Diseases transmitted by mosquitoes are also one of the underlying causes of poverty in developing countries [2]. For these reasons, knowledge about mosquitoes and their physiology are important for understanding how different species differ in their ability to feed on and transmits pathogens to humans. Worldwide, there are 3591 known species of mosquitoes classified in two subfamilies (Anophelinae and Culicinae) and 113 genera [3, 4], however only a handful are known to transmit disease to humans and many are still poorly understood.

Genomic efforts have largely focus on sequencing the primary vectors of the most important human pathogens: Anopheles spp. (malaria); Aedes aegypti and Ae. albopictus (dengue, Zika, Chikungunya and yellow fever); and Culex quinquefasciatus, Cx. tarsalis, and Cx. pipiens (West Nile Virus and St. Louis encephalitis) [5-7]. Transcriptomes have also largely remained withing these mosquito species, however a few additional species have also been sequenced: Aedes fluviatilis [8], Aedes vexans [9], Armigeres subalbatus [10, 11], Ochlerotatus triseriatus [12], Psorophora albipes [13], Toxorhynchites amboinensis, and Wyeomyia smithii [14-16].

Wy. smithii is the only member of the Sabethini tribe to have transcriptome sequences available. No genome sequences are available for this tribe. One study looked at genetic distinctions between biting and non-biting populations of Wy. smithii using microarrays to determine differential expression of head genes of this mosquito species and found an increase in genes involved in blood digestion and reproduction in the blood-feeding populations [15]. Another study used next-generation sequencing to look at the circadian clock-gene divergence in Wy. smithii relative to three other mosquito genera and other insects [16].

Sabethines (Tribe Sabethini) are members of the Culicinae subfamily of mosquitoes and the Aedini (including Ochlerotatus, Aedes, and Haemagogus) is the most closely related tribe to the Sabethini [7]. Members of the Sabethini tribe are forest dwelling, and most are active during the day [3]. There are 435 currently recognized species in 14 genera of the Sabethini tribe. Little is known about most species in this tribe although many species are thought to be potential vectors of several human diseases. Because these are forest mosquitoes, they likely play an important role in maintaining a sylvatic transmission cycle of human pathogens in nonhuman primates [17]. St. Louis encephalitis (SLE) virus has been isolated from different species of Culex, Sabethes, Mansonia, Wyeomyia, among other genera, in Latin America and the Caribbean basin [18]. Sabethes spp. and Haemagogus spp. are known to transmit Yellow Fever Virus (YFV) in the sylvatic cycle, and Haemagogu janthinomys and Sabethes chloropterus in particular, have been implicated as important vectors of YFV in South America [19]. Sabethes cyaneus, another species of this genus, has also been reported to be infected with YFV [20]. Laboratory colonized Sa. cyaneus were also shown to be able to transmit the Zika virus (ZIKV) [21], however no wild-caught mosquitoes have ever been detected with ZIKV in field studies [22].

Sabethes cyaneus can be found in Neo-tropic forests from Belize to Argentina [23]. These mosquitoes develop in tree holes [24] or inside bamboo internodes with small holes [25]. Because adult Sa. cyaneus are rarely found at ground level, they are unlikely to transmit YFV to humans directly [26], however they have been observed to bite humans frequently at that level during the onset of the dry season in eastern Panama [23].

Mosquito saliva plays an active role in blood feeding and pathogen transmission. When a female mosquito takes a blood meal, the resulting tissue injury activates the host’s hemostasis and inflammation responses. Pharmacologically active salivary compounds evolved independently in many blood-feeding arthropods, including between different mosquito species [27]. Saliva helps not only with inhibiting host hemostatic and inflammatory responses, but it also has antimicrobial activity and aids in the ingestion of sugar meals [28, 29].

Salivary glands transcriptomic analyses of several mosquitoes have revealed an array of proteins with known and common functions across hematophagous arthropods as well as unique mosquito proteins or closely related proteins to other Nematocera [30-39]. The mosquito saliva consist of 60–100 secreted proteins, many of which are unique to a specific mosquito species, and most do not have a known function but presumably affect hemostasis, inflammation, and sugar digestion or have antimicrobial activity [40]. Due to the direct involvement of mosquito salivary glands in the transmission of pathogens to human hosts, information on gene transcription and the proteins produced in this organ provides indispensable tools for the systematic and comprehensive analysis of molecules that may play an active role in mosquito blood feeding and pathogen transmission [40].

The mitochondrial genomes of three Sabethes species, Sa. chloropterus, Sa. glaucodaemon and Sa. belisarioi were recently sequenced [41]. However, there are no transcriptome or full genome available for any Sabethes mosquito species, and no characterization of the salivary gland proteins in this genus. In this study we perform a transcriptomic analysis of male and female Sa. cyaneus salivary glands (SG) and characterized the female salivary gland extract (SGE) activity in vitro. We found a total of 5908 genes expressed in the SG, where 137 were exclusively found in males, and 509 were exclusively found in females. A differential expression (DE) analysis revealed 165 up-regulated and 18 down-regulated genes in female SGs compared to male SGs. Most of the up-regulated genes have unknown functions, however, odorant binding proteins, such as those in the D7 protein family, and mucins were among the top 30 genes. Antimicrobial products and cytoskeletal proteins were among the most down-regulated genes in female Sa. cyaneus. In the activity analysis we found that female SGE inhibits coagulation by blocking factor Xa and has endonuclease activity. Female SGE had no hyaluronidase activity, did not inhibit lymphocyte proliferation, nor did it affect nitric oxide production by murine macrophages. This is the first description of a Sabethes SG activity and transcriptomic analysis and our results show an uncommon SG endonuclease activity previously only seen in Culex quinquefasciatus[42].

2. MATERIALS AND METHODS

2.1. Sample preparation

2.1.1. Sabethes cyaneus source

The Maje strain of Sa. cyaneus was isolated by Dr. Woodbridge Foster on Isla de Maje, Lago Bayano in eastern Panama in 1988 and maintained continuously at Ohio State University. In 2017, larvae and pupae from the colony were brought to the Laboratory of Malaria and Vector Research (LMVR) at the National Institute of Allergies and Infectious Diseases in Maryland, USA.

Mosquitoes were reared for a single generation at LMVR in an environmental chamber kept at 28 °C, 80% humidity and a 12 h dark: 12 h light photoperiod and maintained on 10% molasses water.

2.1.2. Dissection of SG and RNA preparation

Salivary glands were dissected from 3-5 days old male and female mosquitoes in PBS under a dissecting microscope. Care was taken to remove as much of the fat body cells as possible with a minuten pin. Due to the small colony size and limiting number of mosquitoes available, only three samples containing 20 SGs each were obtained from males and females for transcriptome analysis. Total RNA was extracted from each sample in 100 μl of TRIzol according to the manufacturer’s instructions (Invitrogen, Carlsbad, CA). Three additional samples (20 SGs each) of female mosquitoes were collected and stored in 20 μl PBS for in vitro functional activity analysis.

2.2. SG transcriptome

2.2.1. Library preparation and sequencing

Tissue samples were submitted to the North Carolina State Genomic Sciences Laboratory (Raleigh, NC, USA) for Illumina RNA library construction and sequencing. Prior to library construction, RNA integrity, purity, and concentration were assessed using an Agilent 2100 Bioanalyzer with an RNA 6000 Nano Chip (Agilent Technologies, USA). Purification of messenger RNA (mRNA) was performed using the oligo-dT beads provided in the NEBNExt Poly(A) mRNA Magnetic Isolation Module (New England Biolabs, USA). Complementary DNA (cDNA) libraries for Illumina sequencing were constructed using the NEBNext Ultra Directional RNA Library Prep Kit (NEB) and NEBNext Mulitplex Oligos for Illumina (NEB) using the manufacturer-specified protocol. Briefly, the mRNA was chemically fragmented and primed with random oligos for first strand cDNA synthesis. Second strand cDNA synthesis was then carried out with dUTPs to preserve strand orientation information. The double-stranded cDNA was then purified, end repaired and “a-tailed” for adaptor ligation. Following ligation, the samples were selected a final library size (adapters included) of 400–550 bp using sequential AMPure XP bead isolation (Beckman Coulter, USA). Library enrichment was performed and specific indexes for each sample were added during the protocol-specified PCR amplification. The amplified library fragments were purified and checked for quality and final concentration using an Agilent 2100 Bioanalyzer with a High Sensitivity DNA chip (Agilent Technologies, USA). The final quantified libraries were pooled for sequencing on one lane of an Illumina HiSeq 2500 DNA sequencer, utilizing 125 bp single end sequencing flow cell with a HiSeq Reagent Kit v4 (Illumina, USA).

2.2.2. RNA-seq data analysis

Raw RNA-seq reads were initially processed by trim galore [43] to filter low quality reads, noisy short fragment, and adapter sequences. There is currently no genome available for Sa. cyaneus, so a de novo transcriptome assembly was created using Abyss [44] and Trinity [45]. Briefly this was done by first creating an Abyss assembly of the reads with k values of 25, 35, 45, 55, 65, 75, 85 and 95, followed by a Trinity assembly of the reads. The assembly of the assemblies used a cap3 + blast n pipeline. The final product is a fasta file with the assembled contigs. This file was used to blast (BLASTx) the peptide translations against several databases (NCBI diptera, swissprot, refseq-invertebrate) and only those with at least 70% coverage of the match were kept. Additionally, all peptides longer than 50 aa were exported and submitted to the signalP program to select those having a signal peptide. This step recovers many peptides that are secreted but do not give good matches to anything known. The blasted and peptides with a signal secretion sequence were merged, and their redundancies removed using the program CD-HIT [46]. The sequence data have been deposited in NCBI (accession number GKCV00000000).

The male and female reads were aligned to the de novo transcriptome file using bowtie2 version 2.4.2 [47] with default parameters. A matrix of raw read counts was generated using the featureCount package from subread version 2.0.0 [48]. Genes with low read counts (≤2 TPM) were filtered out.

Differential expression analysis was done using edgeR version 3.32.1 [49]. The counts were normalized using the edgeR function calcNormFactors which adjusts for difference in library size and uses a TMM (Trimmed mean of M-value) method to eliminate RNA composition effect. Mean Difference (MD) plots were generated to check composition bias between male and female samples. The Log fold change (Log FC), p-value and false discovery rate (FDR) were calculated using the exactTest function. Because we had only one replicate per sample, dispersion could not be calculated for DE analysis, and so dispersion was manually set to 0.2. Only genes with FDR < 0.05 were evaluated for differential expression. A heatmap of the most differentially expressed genes was generated using gplots (version 3.1.1) heatmap.2 and a volcano plot was made using ggplot2 (version 3.3.5) functions in R.

Homology to known transcripts was assessed by generating BLASTp and BLASTx reports against a Diptera database extracted from NCBI’s nr and transcriptome shotgun assembly (TSA) databases and those that had at least 50% coverage and an e-value of <1e-15 were included in our protein description.

2.3. SGE activity in vitro

2.3.1. Factor Xa inhibition

Inhibition of coagulation process through blocking of FXa was determined by a colorimetric assay. Briefly, human FXa (Haematologic Technologies) was diluted to 1 nM in FXa buffer (10 mM HEPES, 150 mM NaCl, 5 mM CaCl₂, 0.15% BSA, pH 7.3). In a flat bottom microplate (Costar), 75 μl of 1 nM FXa was incubated with serial dilutions of SGE from Sa. cyaneus for 15 min at 37 °C. Reactions were adjusted to a final volume of 100 μl with FXa buffer. As a positive control, 1 μl of 0.1 mg/ml of recombinant serpin from Aedes albopictus was added [50]. As the negative control, no sample was added to the reaction, only PBS. Substrate Chromogenix S-2222 (Diapharma) was added to a final concentration of 5 μM and reactions were run at 37 °C for 1 h. Plates were read at 405 nm using a VersaMax microplate reader (Molecular Devices). Samples were run in triplicates.

2.3.2. Hyaluronidase assay

The hyaluronidase activity was determined by the turbidimetric method. Briefly, 0.1 mg/ml hyaluronic acid (HA) solution was prepared in 25 mM HEPES, 100 mM NaCl, 0.1% BSA, pH 7.3. Four micrograms of HA were incubated with different dilutions of SGE of Sa. cyaneus As a positive control, 10 nM final concentration of recombinant hyaluronidase from the sand fly Lutzomyia longipalpis was added [51]. Reactions were run at 37 °C for 1 h and stopped by the addition of 100 μl of 10 mg/ml cetylpiridinium chloride, which reacts with HA resulting in turbidity. Plates were read at 405 nm using a VersaMax microplate reader (Molecular Devices, San Jose, CA). Maximum turbidity corresponds with intact HA. Blanks lacking samples as well as HA were included in all tests. Results are expressed as the % of remaining HA, taking as 100% remaining HA as the negative control, where no sample was added to the reaction, only PBS.

2.3.3. Endonuclease activity

Endonuclease reactions contained 400 ng double-stranded circular plasmid DNA (VR2001; Vical Incorporated) diluted in 20 μl of 150 mM NaCl, 5 mM MgCl₂, 50 mM Tris-HCl, 0.5% BSA, pH 8.0. The reaction mixtures were incubated with different dilutions of SGE and as positive control, 1 nM of recombinant sand fly endonuclease Lundep [52] was used. After 10 min at 37 °C, reactions were stopped by adding 5 nM EDTA. 15 μl of reaction was electrophoresed in 1.2% agarose gel (Invitrogen) and visualized under UV light.

2.3.4. Lymphocyte proliferation

Lymphocytes were isolated from the spleen of BALB/c mice. Spleens were macerated through a 40 μm cell strainer (Falcon) and washed with RPMI. The cells were spun down at 1500 rpm for 5 min at 10 °C and the pellet was lysed with 2 ml of Ammonium-Chloride-Potassium (ACK) lysis buffer (Thermo Fisher, Waltham, MA) for 2 min at room temperature. Lysed cells were transferred to a new tube and topped with complete RPMI (supplemented with 10% Bovine Fetal Serum and 1X of PenStrept, Thermo Fisher, Waltham, MA) and strained again with 10 ml RPMI. Cells were spun at 1500 rpm for 5 min at 10 °C and the pellet was resuspended in 2 ml of complete RPMI. The final count was performed using a Neubauer chamber.

Samples were incubated with the isolated cells in sterile flat well plates for 2 days at 37 °C with 5% CO2. Serial dilutions of Sa. cyaneous SGE from a single biological replicate (pool of SGs) were used with 4 technical replicates. Cells incubated with Ae. aegypti SGE were used as a positive control, as it has shown to inhibit cell proliferation [53] and cells with Concanavalin A (0.5 μg/mL final concentration) were used as negative control, as it stimulates cell proliferation. The lectin Concanavalin A (Con A), a is a well-known T cell mitogen that can activate the immune system, recruit lymphocytes, and elicit cytokine production. Concanavalin A irreversibly binds to glycoproteins on the cell surface and commits cells to proliferation. Con A is widely used as a surrogate for antigen-presenting cells in T cell in splenocyte stimulation and proliferation experiments. After incubation, 10 μl of AlamarBlue (ThermoFisher, Waltham, MA) was added to each well and incubated for 72 h. Absorbance was read 570 and 600 nm.

3. RESULTS AND DISCUSSION

The evolutionary arms race between hematophagous insects and their hosts gives rise to diverse insect salivary proteins important for combating the host’s hemostasis, inflammation and immunity [27]. Saliva is injected while the mosquito mouth parts are inside the host skin searching for blood and during blood ingestion, thus saliva ends up both inside the host and inside the mosquito gut as a result of saliva re-ingestion along with the blood meal [27, 54]. Sabethes cyaneus mosquitoes feed primarily on non-human primates and occasionally on human hosts. This species, like other Sabethini mosquitoes are involved in maintaining the sylvatic cycle of ZIKV and YFV and are potential human vectors of these viruses [21]. This study is the first attempt at characterizing the salivary proteins important for blood feeding in this species. We compared transcript abundance levels between male and female Sa. cyaneus and looked at in vitro hemostasis and immunological activity by salivary gland extracts (SGE).

Our RNA-seq libraries returned a total of 68,980,988 million reads for female, and 61,281,958 million reads for male samples. Both samples had equal distribution of read counts (S Fig. 1). After removal of genes with no counts in one strain and the genes with a TPM < 2 in at least one sample, 5908 genes remained for differential expression (DE) analysis. Of the 5908 genes, 5771 were found in females, 5399 were found in males, 5262 were found in both males and females, and 646 were found in only one (male or female) (Fig. 1, S Table 1). There were 509 transcripts found in females that were not present in males and 137 transcripts found exclusively in males (Fig. 1). The Mean Difference plot indicated no composition bias between male and female samples (S Fig. 2).

Figure 1:

Venn diagram of the number of total transcripts in male and female salivary glands.

Our Differential Expression analysis identified genes differentially expressed between female and male Sa. cyaneus salivary glands (SGs). A total of 183 transcripts were differentially expressed between male and female samples. Female SGs had more (165) upregulated genes than male SGs (18) (Fig. 2A), an effect that has been observed in other Culicine species [55]. A volcano plot showing the significantly up- or downregulated genes are shown in Fig. 2B. We found that the gene upregulation profile of female Sa. cyaneus was similar to Aedes aegypti. A study of the differentially expressed salivary gland transcripts of female and male Ae. aegypti found 207 upregulated, 68 downregulated, and 85 not differentially expressed genes in female SGs [55]. Of the upregulated transcripts in female Sa. cyaneus, the top approximately one third (43) are classified as unknown, unknown conserved, and unknown conserved secreted proteins/products (Fig. 3). The remainder two thirds are potential enzymes (42), immunity related proteins (8), potential housekeeping genes (55), and 17 other proteins with known functions (including Aegyptins (2), serpins (2), mucins (6), and odorant binding proteins (OBPs) (6)). The most differentially expressed gene was Sc-14366 (Log FC = 13.9), a putative larval cuticle protein lcp-30 belonging to the structure protein superfamily. When looking at the top 50 most upregulated genes in the female sample, the most abundantly identified categories were unknown proteins (7), odorant binding proteins (i.e. D7 proteins) (6), unknown conserved proteins (4), and mucin (4) (Table 1 and Fig. 4).

Figure 2:

Transcripts significantly differential expressed between male and female Sa. cyaneus. (A) Venn diagram showing the number of down-regulated (left) and up-regulated (right) transcripts in females compared to males. (B) Volcano plot showing gene expression profile of male and female salivary glands. Red dots represent up-regulated and blue dots represent down-regulated in females. Grey dots represent genes that are not differentially expressed.

Figure 3:

Total number of up-regulated genes in female Sa. cyaneus by protein superfamily. Colors represent the number of proteins in each superfamily.

Table 1:

Statistically significant up-regulated transcripts in female Sa. cyaneus salivary gland.

Gene ID	Total Reads Males	Total Reads Females	Log FC	P-Value	Protein Classification
Sc-14366	0	1387	13.26	1.08E-11	Structural protein
ScSigP-25427_FR5_5-169	8	73210	12.77	5.44E-17	Odorant binding protein (putative long D7)
ScSigP-24884_FR5_7-589	278	2297046	12.64	2.46E-17	Apyrase
ScSigP-24644_FR6_6-344	86	690653	12.60	3.18E-17	Odorant binding protein (putative long D7)
ScSigP-23471_FR5_17-296	33	265071	12.60	4.20E-17	Allergen
ScSigP-24499_FR2_25-278	157	1221110	12.56	3.50E-17	Aegyptin
ScSigP-24700_FR4_1-657	33	254084	12.54	5.19E-17	Lipases (putative acid sphingomyelinase)
ScSigP-24696_FR6_4-423	93	665567	12.44	5.61E-17	Serpins
ScSigP-24500_FR6_5-326	174	1183340	12.36	6.73E-17	Allergen
Sc-23125	68	454782	12.34	8.01E-17	Odorant binding protein
Sc-24497	33	205393	12.23	1.50E-16	Aegyptin
ScSigP-24506_FR1_1-607	46	280841	12.20	1.37E-16	Mucins
Sc-911	0	565	11.97	9.21E-10	Nuclear protein
ScSigP-3840_FR2_501-617	0	547	11.92	1.08E-09	Unknown protein
ScSigP-24500_FR2_15-84	20	92199	11.79	8.42E-16	Unknown protein
ScSigP-24896_FR6_114-180	7	27820	11.57	4.40E-15	Zinc finger protein
ScSigP-23741_FR6_263-345	0	402	11.48	4.93E-09	Unknown protein
ScSigP-21810_FR2_1078-1167	82	293096	11.43	1.79E-15	Unknown conserved protein
ScSigP-24994	0	377	11.39	6.75E-09	Proteases (aminopeptidase N-like)
ScSigP-17482_FR5_1-58	0	374	11.38	6.96E-09	Unknown protein
ScSigP-388	2	6799	11.29	8.20E-14	Endoenzyme (hyaluronidase A isoform X1)
Sc-24675	79	235351	11.17	4.57E-15	Mucins
Sc-17014	0	281	10.96	2.80E-08	Unknown protein
Sc-1612	0	268	10.90	3.57E-08	Unknown conserved protein
ScSigP-7463	0	255	10.82	4.50E-08	Protein export
ScSigP-5649_FR5_1-76	18	41024	10.78	3.22E-14	Unknown conserved protein
Sc-16371	0	196	10.44	1.60E-07	Antimicrobial products
ScSigP-3840_FR4_2-626	7	12524	10.41	2.37E-13	Glycosidases and amylases
ScSigP-24327_FR3_391-474	0	170	10.24	3.17E-07	Unknown conserved secreted product
ScSigP-25088_FR6_100-300	100	142598	10.11	1.76E-13	Mucins
ScSigP-24748_FR6_10-339	2560	2965546	9.81	4.29E-13	Odorant binding protein (putative long D7)
ScSigP-24748_FR3_347-457	386	433321	9.76	5.22E-13	Unknown protein
Sc-24695	473	411683	9.40	1.85E-12	Odorant binding protein (putative long D7)
ScSigP-8096_FR5_1-64	16	12438	9.22	6.36E-12	Unknown protein
Sc-24818	6	4565	9.18	2.23E-11	Carbohydrate metabolism
ScSigP-4147_FR1_147-292	1	824	9.17	8.48E-10	Mucins
ScSigP-21410_FR4_42-282	128	78926	8.90	1.12E-11	Small molecule binding proteins
Sc-21201	2	1038	8.58	9.41E-10	Lipid metabolism
Sc-24214	34	16325	8.53	5.23E-11	Protein modification
ScSigP-24529_FR4_36-536	1705	715083	8.34	6.79E-11	Odorant binding protein
ScSigP-4147_FR6_108-390	7	2516	8.10	6.97E-10	Small molecule binding proteins
ScSigP-10749_FR3_1-288	3	1006	7.97	6.46E-09	Extracellular matrix
Sc-11917	5	1562	7.89	2.67E-09	Lipases (phospholipase A1)
Sc-3449	2	599	7.78	1.40E-08	Unknown conserved protein
ScSigP-17978_FR1_1-1396	63	14471	7.47	1.70E-09	Nuclear protein
ScSigP-16283_FR1_1-148	1	238	7.38	3.63E-07	Proteases
Sc-17415	15	3070	7.30	5.17E-09	Immunity
Sc-16543	1	205	7.16	7.39E-07	Unknown conserved secreted product
Sc-24435	246	44039	7.12	4.92E-09	Basic tail-containing product
Sc-24327	15	2659	7.09	1.05E-08	Signal transduction

Figure 4:

Top 50 up-regulated genes in female Sa. cyaneus by protein superfamily. Colors represent the number of proteins in each superfamily.

Hemostatic mediators that inhibit blood loss and promotes wound healing are important deterrents to blood feeding by hematophagous arthropods and contribute the fast evolutions of diverse anti-hemostatic molecules in the arthropod saliva. Blood clotting is an important step in hemostasis, where coagulated blood reinforces the platelet plug to seal the wound. Blood feeding by mosquitoes can be detrimentally impacted if blood flow is reduced or stopped due to blood clotting. Blood clotting can be activated by the intrinsic or extrinsic pathways which converge with the activation of factor X (FX) to factor Xa (FXa). Factor Xa converts prothrombin to thrombin which then converts fibrinogen to fibrin that forms the blood clot [56]. FXa also plays a role in inflammation through activation of protease-activated receptors (PAR) [57].

In culicine mosquitoes, proteins in the serpin (Serine Protease Inhibitor) family have been associated with factor Xa inhibition [50, 58, 59] and is the main salivary anti-clotting in Ae. aegypti mosquitoes [58, 59]. Alboserpin, the Aedes albopictus salivary serpin protein, was shown to strongly inhibit Factor Xa and bind heparin, interact with phosphatidylcholine (PC) and phosphatidylethanolamine (PE) but not with phosphatidylserine (PS), and display potent antithrombotic properties in vivo [50]. Serpins are commonly found in mosquitoes [12, 13, 34, 39, 40, 55, 60, 61]. In Sa. cyaneus females we found ten transcripts that match serpins from various species of Diptera including Ae. aegypti (2) and Ae. albopictus (2). Only two of the ten were upregulated in females Sa. cyaneus and these were a best match to serpins from Psorophora albipes (Log FC = 12.4) and Culex tarsalis (Log FC = 4.8). To evaluate whether female Sabethes SGE had factor Xa inhibitory activity, we measured the cleavage of the chromogenic substrate S-2222 for FXa. As shown in Fig. 5, Sabethes SGE inhibited generation of FXa in a dose-dependent manner starting at dilution 1/32. While the serpin activity of Psorophora and Culex SGE have not been evaluated, the SGEs of Ae. aegypti and Ae. albopictus also produced a dose dependent inhibition of FXa, with strong inhibition starting at approximately 2 μg of female SGE protein from Ae. aegypti and 2 nM Alboserpin from Ae. albopictus, and nearly complete inhibition at 10 μg and 14 nM respectively [50, 58]. The 2 upregulated serpins, especially transcript ScSigP-24696_FR6_4-423 that was highly upregulated (Log FC = 12.4) in female Sa. cyaneus SGs are potential candidates for the FXa inhibition effect seen here, however the molecule responsible for this activity remains to be identified.

Figure 5:

Inhibition of coagulation process through blocking of factor Xa determined by a colorimetric assay. Phosphate buffered saline (PBS) was used as a negative control and, serpin from Aedes albopictus was used as positive control (C+). Numbers above Sabethes SG represent the different dilutions used.

While most proteins found in mosquito salivary glands have no known functions, many play known roles in the mosquito blood feeding process. Several enzymes, have known or presumed functions in mosquito saliva [62]. The D7 family of odorant binding proteins are found in the sialome of all blood feeding mosquitoes and are one of the most abundantly expressed and well characterized mosquito SG proteins [62-64]. The second and fourth most differentially expressed genes in female Sa. cyaneus were ScSigP-25427_FR5_5-169 (Log FC = 12.8) and ScSigP-24644_FR6_6-344 (Log FC = 12.6), both putative long d7 salivary protein. The two other putative long d7 salivary proteins in the Sa. cyaneus transcriptome were also upregulated with Log FC values > 9 (Table 1). The D7s are strong binders of biogenic amines and leukotrienes and have been biochemically shown to antagonize inflammation and hemostasis agonists [65-69]. D7 proteins were found upregulated in female mosquitoes of all species sequenced so far except Toxorhynchites amboinensis, a species that does not feed on blood [14].

An apyrase (ScSigP-24884_FR5_7-589) was the third most differentially expressed gene (Log FC = 12.6) in the Sa. cyaneus SG. Apyrases hydrolyze ATP and ADP which are important hemostasis and inflammatory agonists released by platelets and broken cells following tissue injury [70]. Apyrases are also commonly found in the salivary glands of other mosquito species [13, 32, 34, 35, 38-40, 55, 60].

Sabethes cyaneus has many of the ubiquitous proteins found in the SGs of other mosquitoes. Twelve mosquito species, namely, Ae. aegypti [39, 55], Ae. albopictus [60], Ochlerotatus triseriatus [12], Cx. tarsalis [61], Culex pipiens quinquefasciatus [37], Ps. albipes [13], Anopheles darlingi [40], Anopheles funestus [34], Anopheles gambiae [32, 35], Anopheles stephensi [38], and Tx. amboinensis [14] have a salivary gland transcriptome (sialotranscriptome) described. Of the putative secreted proteins overexpressed in Sa. cyaneus female SGs, antigen 5 and mucins have been detected in all mosquito species [12-14, 32, 34, 35, 38-40, 55, 60, 61]. Antigen 5 family are widespread in insect and tick sialomes suggesting a unique association to hematophagy, including the inhibition of platelet aggregation in a tabanid salivary protein [71], immunoglobulin binding [72], and putative inhibitor of the classic pathway of complement in the stable fly Stomoxys calcitrans [73]. Mucins probably function by coating and lubricating the salivary channels and food canals, and may also have antimicrobial properties [40, 62].

Maltases [12, 32, 34, 35, 38-40, 55, 60, 61], proteases [12-14, 32, 35, 39, 40, 55, 60, 61], and allergens [12, 34, 38-40, 55, 60, 61], which were upregulated in female Sa. cyaneus, are also commonly found in other mosquito species. Maltases are commonly present on both male and female mosquito SG and are presumed to be involved in sugar feeding functions [62]. The function of serine proteases in mosquito saliva has not been biochemically characterized so far, but it is presumed to function in immunity, serving as prophenoloxidase activators, or in blood feeding, by digesting the host matrix and blood proteins, and neutralizing host defense proteins [62, 74].

Other Sa. cyaneus upregulated genes found in several species of mosquitoes are: Lipases [13, 37, 38, 55], such as phospholipase A2, which has not been biochemically characterized, however the Culex quinquefasciatus phospholipase C has a potent activity that hydrolyzes Platelet Aggregation Factor (PAF) [75]; Aegyptin [12, 13, 40, 61], which inhibits platelet aggregation by interfering with collagen recognition [76, 77]; and endonucleases [12-14, 37, 61], which decrease the skin matrix viscosity around the feeding site [62].

Endonucleases are a broad group of enzymes capable of hydrolyzing nucleic acids and can serve in a wide variety of functions such as nucleotide salvage, repair, recombination, transposition, and degradation as well as in cell defense by promoting the degradation of foreign nucleic acids [42]. Most biochemically characterized endonucleases have the conserved R(K)GH triad. Two upregulated genes in female Sa. cyaneus have completely (Sc-1612) or highly conserved (Sc-3553) motifs with the active center of the Cx. quinquefasciatus endonuclease gene (CuquEndo: AAR18449) contain the RGH and other amino acid residues implicated in DNA hydrolysis and stabilization of the active site (Fig 6). It’s been proposed that a mosquito salivary endonuclease could act as a spreading factor for other salivary activities by reducing the local skin matrix viscosity at the biting site [42]. In the sand fly, Lutzomyia longipalpis, secreted salivary endonuclease (Lundep) is capable of reducing local inflammation induced by vertebrate hosts, and destroying neutrophil extracellular traps (NETs) produced by activated human neutrophils, an activity that exacerbates Leishmania major infection in vivo [52]. We observed a high dose dependent endonuclease activity in the Sa. cyaneus SGE as it was found in SGE of L. longipalpis and Cx. quinquefasciatus. Activity was seen up until the 1/16 dilution sample (Fig 7). More diluted SGE samples (1/64 – 1/512) had no endonuclease activity equivalent to the no salivary gland (NSG) control. CuquEndo, Sc-1612, and Sc-3553 also contain other amino acid residues implicated in the nucleophilic attack of DNA substrate and stabilization of the active site [42]. However, the function of these endonucleases in blood feeding in Sa. cyaneus remains to be elucidated.

Figure 6:

Two putative Sa. cyaneus endonucleases genes were aligned to the C. quinquefasciatus (CuquEndo: AAR18449) and the sand fly, Lutzomyia longipalpis (Lundep: AAS16916.1) endonuclease genes. Conserved motifs are shown in magenta. Most known endonucleases have the conserved RGH triad.

Figure 7:

Sa. cyaneus SGE has endonuclease activity. Numbers above Sabethes SGE represent the different dilutions used. Positive control (C+) is the sand fly endonuclease gene, Lundep (1 nM). Negative control (NSG) is no salivary gland added.

The degradation of hyaluronic acid, an important component of dermal extracellular matrix, is mediated via free chemical radicals and different hyaluronidases, and is important in wound healing [78]. Hyaluronidases have been found in salivary glands (sialomes) of various bloodsucking insects [13, 37, 71, 79, 80]. While hyaluronidase genes have been identified in the transcriptomes and genomes of many mosquitoes, such as in the oocytes of Ae. albopictus females reared under different diapause conditions [81], in the neurotranscriptome and genome of Ae. aegypti [82, 83], and in the genomes of Culex pipiens pallens [84], Anopheles albimanus (PRJNA657052), Anopheles sinensis [85], An. stephensi (PRJNA661063), An. gambiae [86], Anopheles merus (PRJNA719141), Anopheles arabiensis and Anopheles coluzzii [87], it has only been found in the SG transcriptome of two mosquito species, Cx. quinquefasciatus and Ps. albipes [13, 37]. In Sa. cyaneus female SGs, the upregulated ScSigP-388 gene is homologous to the Ae. albopictus hyaluronidase A isoform X1 gene (E-value = 0.0, % identity = 78, % coverage = 98) and is closest to other Aedes hyaluronidases than any other mosquito species with available sequences (Fig. 8). To date, only the Ae. aegypti, An. stephensi, and Cx. quinquefasciatus salivary hyaluronidases have been tested for hyaluronidase activity in mosquitoes, and of these, Culex is the only one found to have enzyme activity [88]. We found that similarly to the Aedes hyaluronidases, the SGE of female Sa. cyaneus had no hyaluronidase activity. The percent remaining hyaluronic acid of samples incubated with Sa. cyaneus SGE were nearly 100% and no different than the PBS negative control, while the sand fly Lu. longipalpis positive control showed significant reduction in hyaluronic acid (Fig. 9).

Figure 8:

Phylogenetic tree of the hyaluronidase protein sequences found in other mosquito species available in NCBI. Color blocks highlight the different genus and each number represent a species. 1. Culex quinquefasciatus, 2. Cx. pipiens, 3. Aedes aegypti, 4. Ae. albopictus, 5. Anopheles sinensis, 6. An. albimanus, 7. An. stephensi, 8. An. gambiae, 9. An. merus, 10. An. arabiensis, 11. An. coluzzii.

Figure 9:

Sa. cyaneus salivary gland extract had no effect on hyaluronidase activity. Numbers above Sabethes SGE represent the different dilutions used. Phosphate buffered saline (PBS) was used as a negative control and the sand fly Lutzomyia longipalpis as the positive control (C+).

Vertebrate innate and acquired immune responses to blood feeding is yet another barrier hematophagous arthropods must combat. Peptides and proteins injected into the mammalian host through the arthropod saliva to evade hemostatic responses also elicit immune responses in the host [89, 90]. To overcome this barrier, anti-inflammatory and immunomodulatory components are secreted into the host during blood feeding [29, 89, 90]. One way to slow down the host immune response is by suppressing the proliferations of immune cells that produce Immunomodulating agents such as cytokines. While this immunomodulation activity has been well characterized in ticks and described in other blood feeding insects [89, 90], so far, the suppression of concanavalin A induced murine spleen lymphocyte cell proliferation has only been observed in Ae. aegypti mosquitoes [53, 91, 92].

To investigate the effect of Sa. cyaneus SGE on the proliferation of T cells, we incubated lymphocytes isolated from the spleen of BALB/c mice. Sa. cyaneus SGE did not inhibit the polyclonal activation of T-cells at any concentration (Fig 10).

Figure 10:

Sa. cyaneus SG does not inhibit lymphocyte proliferation. Negative control = cells incubated with medium only. Test control = cells incubated with Con A, a polyclonalactivator of T cells. All SGE samples were added to wells containing cells stimulated with Con A. Ae. aegypti SGE was used as a positive control. ANOVA test was used to determine statistical significance. For multiple comparison, test control (cells with medium + ConA) was used as control sample.

Male mosquitoes are not blood feeders and therefore do not require the complex and abundant cocktail of proteins needed to overcome a host’s hemostatic and immune system. Male Sa. cyaneus SGs have 18 overexpressed transcripts. A study looking at the SG gene expression in Ae. aegypti found 65 transcripts overexpressed in male SGs compared to females [55]. Of the 18 transcripts overexpressed in male Sa. cyaneus, 5 were unknown conserved proteins (Log FC between −3.6 and −10), 2 were unknown secreted products (Log FC of −10.4 and −11.3), 2 unknown proteins (Log FC of −3.7 and −11.5), 2 were nuclear proteins (LogF C ~ −7), 2 were structural (Log FC of −10.5 and −11.7), 2 were cytoskeletal proteins (Log FC of −11.3 and −12), and there were one each of signal transduction (Log FC = −3.56), G protein-coupled receptor (Log FC = −4.49), and antimicrobial product (Log FC ~ −12) (Table 2). The most downregulated transcript codes for an antimicrobial product and on the list of the top 7 are 2 cytoskeletal proteins and 2 structural proteins (all with Log FC > −10). A signal transduction gene was found upregulated in both Ae. aegypti and Sa. cyaneus male SGs, however, there were no other matches for male SG overexpression genes between the two species. Further validation might be necessary to confirm differentially expressed genes.

Table 2:

Significantly down-regulated transcripts in female Sa. cyaneus salivary glands.

Gene ID	Total Reads Males	Total Reads Females	Log FC	P-Value	Protein Classification
Sc-13561	473	0	−12.08	6.35E-10	Antimicrobial products
Sc-4319	444	0	−11.99	8.68E-10	Cytoskeletal proteins
Sc-17120	360	0	−11.69	2.43E-09	Structure protein
ScSigP-2669_FR4_147-349	309	0	−11.47	5.14E-09	Unknown protein
Sc-25608	279	0	−11.32	8.47E-09	Cytoskeletal proteins
ScSigP-3149_FR3_89-207	279	0	−11.32	8.47E-09	Unknown secreted product
Sc-2669	155	0	−10.47	1.47E-07	Structure protein
ScSigP-1398_FR2_1-113	149	0	−10.42	1.79E-07	Unknown secreted product
Sc-25623	1667	2	−9.97	2.54E-11	Unknown conserved protein
ScSigP-21474_FR5_1-128	6036	42	−7.53	1.61E-09	Nuclear protein
ScSigP-24761_FR3_1-132	6279	67	−6.92	1.16E-08	Nuclear protein
Sc-24247	4131	46	−6.85	1.50E-08	Unknown conserved protein
Sc-940	9363	538	−4.49	3.14E-05	G protein-coupled receptor
ScSigP-16912_FR2_1-173	1240	123	−3.70	4.06E-04	Unknown protein
ScSigP-11559_FR1_6-403	410	42	−3.65	5.86E-04	Unknown conserved protein
Sc-10672	145	15	−3.63	9.59E-04	Unknown conserved protein
Sc-16178	225	24	−3.59	8.46E-04	Unknown conserved protein
Sc-3767	931	102	−3.56	6.54E-04	Signal transduction

Similar to other sialotranscriptomes, most of the transcripts overexpressed in Sa. cyaneus SGs belong to either known families of unknown function or are entirely of unknown function and family. The data provided here will be helpful for defining candidate proteins and for future functional characterization of the SG proteins found in Sabethes mosquitoes.

Supplementary Material

1

Supplemental Figure 1: Sequencing output from Sa. cyaneus salivary glands. (A) Library sizes after normalization. (B) distribution of the read counts on the log2 scale.

2

Supplemental Figure 2: Mean Difference (MD) plot shows no composition bias between male and female samples.

Sabethes cyaneus, a sylvatic mosquito species found in Central and South America can carry arboviruses

We generated the first male and female salivary glands transcriptome for this mosquito genus.

Salivary gland secretions are known to facilitate blood feeding and pathogen transmission

Transcriptomic and proteomic studies are instrumental in the understanding of biology of salivary components