Culex species, such as Culex quinquefasciatus and Culex nigripalpus display a range of feeding habits and act as vectors for pathogens that can cause diseases in both humans and animals. Understanding their feeding habits is pivotal for enhancing disease prevention strategies. The present study introduces the application of two multiplex real-time PCR (qPCR) followed by High-Resolution Melting (HRM) as a cost-effective and time-efficient alternative. This investigation involved the development of two multiplex qPCR-HRM: assay 1 detects human, dog, and chicken, while assay 2 detects cat, cattle, and horse in Culex sp. engorged female mosquitoes. The qPCR-HRM reactions showed a detection limit of one copy of genomic DNA when performed as single and multiplex qPCR-HRM. The reaction efficiencies were 97.96% for human, 100.60% for dog, 99.03% for chicken, 99.92% for feline, 99.18% for cattle, and 97.68% for horse. The qPCR-HRM method, employing multiplex 1 and 2, was applied to field-collected mosquitoes and demonstrated the ability to detect DNA from multiple blood sources within a single sample. By analyzing both multiplexes, it was possible to identify up to five distinct blood sources in Cx. quinquefasciatus and Cx. nigripalpus, and up to two sources in Culex coronatus. Sequencing corroborated the qPCR-HRM results, confirming the presence of DNA from one to four different blood sources with 100% accuracy. The development of these molecular methods may contribute for identification of blood-feeding patterns in mosquitoes. It contributes to studies on the dissemination and transmission of pathogens among various animals and humans, thereby bridging the gap between epidemiology and vector monitoring.

Study design

This study comprised three main sections. The first section involved qPCR-HRM design, which included primer design followed by in silico specificity testing. The second part focused on the standardization of two multiplex qPCR-HRM, including the determination of analytical sensitivity and in vitro specificity testing. The third section involved the validation of the technique, where 310 field-collected mosquitoes identified as Culex sp. were categorized into engorged females (females with abdomens completely full of blood), non-engorged females (evaluated in stereomicroscopic), and males (Figure 1). DNA was extracted from these specimens, followed by species identification using a multiplex PCR assay. Subsequently, the samples were analyzed using the two multiplex qPCR-HRM assays developed for this study: multiplex 1 for the detection of DNA from humans, dogs, and chickens, and multiplex 2 for the detection of DNA from cats, cattle, and horses. All samples analyzed by the two multiplex qPCR-HRM assays were also subjected to two multiplex conventional PCR to validate the results. In addition, 25 engorged female samples that tested positive for multiple vertebrate food sources were selected for Sanger sequencing to confirm the qPCR-HRM results.

Controls

The positive controls were obtained from whole blood samples representing five domestic animal species that serve as potential blood-feeding sources for Culex sp. mosquitoes: cattle (Bos taurus), horse (Equus caballus), chicken (Gallus gallus), cat (Felis catus), and dog (Canis lupus familiaris). Additionally, a human blood sample (Homo sapiens) was included as the sixth positive control. Extraction of DNA from these blood samples was carried out using the commercial kit PureLink™ Genomic DNA Mini (Invitrogen™, Waltham, Massachusetts, USA), followed by quantification through spectrophotometry using Nanodrop® (Thermo Fisher Scientific, Wilmington, DE, USA). Total DNA samples were standardized to a concentration of 10 ng/µL and stored in a freezer at -20ºC until molecular analysis.

Additionally, a positive control was prepared by combining DNA from human, dog and chicken for multiplex 1 and cattle, horse and cat for multiplex 2. Specifically, 10 ng/µL of DNA from each host was mixed in a 1.5 mL Eppendorf tube. This mixture was used as the positive control for the experiments.

Primer design

Six primer pairs were designed to standardize two multiplex qPCR-HRM assays for detecting each of the six animal species that may serve as food sources for the dipterans (Table 1). Primers were designed employing Primer Express® 3.0 (Thermo Fisher Scientific Inc., Waltham, MA, USA) based on sequences of mitochondrial DNA from: Equus caballus (NC_001640); Canis lupus familiares (NC_002008.4); Felis catus (NC_001700.1); Gallus gallus (NC_012920.1); Bos taurus (NC_006853.1) and Homo sapiens (NC_012920.1), all available from GenBank.

Primer’s characteristics were evaluated using Oligo Analyzer (Gene Link^TM, Hawthorne, NY, USA) and specificity was tested in silico using primer-BLAST (NCBI, Bethesda DM, USA) and in vitro was assessed through cross-tests between the targets.

The qPCR-HRM reactions were organized into two multiplexes based on the combination of primers, determined by analyzing the dissociation temperatures of each target species using uMELT^SM software (University of Utah, USA). Melting temperature differences for each species were determined based on GC content and amplicon length.

Multiplex qPCR-HRM standardization

Primer optimization involved concentration and annealing temperature tests. Initially, the last dilution points of each amplified control obtained in the detection limit assay were subjected to a new qPCR with annealing temperatures ranging from 55°C to 60°C. Additionally, each primer pair underwent testing at three different concentrations (200nM, 400nM, and 600nM), resulting in 27 unique concentration combinations for each species in each multiplex.

Multiplex 1 qPCR-HRM (dog, human and chicken)

The first multiplex reaction was standardized to a final volume of 12 µL containing: 1x MeltDoctor™ HRM Master Mix (Thermo Fisher Scientific, Wilmington, DE, USA), 400nM primers for dog DNA detection, 400nM for human DNA detection, 600nM primer pair for chicken DNA detection and 10ng/µL of DNA from each species. The thermocycling conditions were 95°C for 10 minutes, 40 cycles of 95°C for 15 seconds, 57°C for 30 seconds, and 72°C for 30 seconds. At the end of the amplification cycles, a melting curve was added from 60 to 95°C with fluorescence capture every 0.1°C.

For the molecular analysis, positive controls used were DNA samples of each host species as described in the “Control” subsection, and two negative controls were employed: UltraPure™ DNase/RNase-Free Distilled Water (Thermo Fisher Scientific, Wilmington, DE, USA).

Multiplex 2 qPCR-HMR (feline, cattle and horse)

The second multiplex was standardized to a final volume of 12 µL containing: 1x MeltDoctor™ HRM Master Mix (Thermo Fisher Scientific, Wilmington, DE, USA), 600nM primers to detect cat DNA, 600nM to detect cattle DNA, 600nM to detect horse DNA and 10ng/µL of DNA from each species. The thermocycling conditions were 95°C for 10 minutes, followed by 40 cycles of 95°C for 15 seconds, 58°C for 30 seconds, and 72°C for 30 seconds. A melting curve was then added with the same parameters as the previous reaction. The dissociation curves of the amplified products were analyzed in High Resolution Melt Software v3.2 (Thermo Fisher Scientific, Wilmington, DE, USA).

For the molecular analysis, positive controls used were DNA samples of each host species as described in the “Control” subsection, and two negative controls were employed: UltraPure™ DNase/RNase-Free Distilled Water (Thermo Fisher Scientific, Wilmington, DE, USA).

Analytical sensitivity

DNA samples extracted from each host species were quantified by fluorimetry using the Qubit™ dsDNA Quantification Kit (Thermo Fisher Scientific, Wilmington, DE, USA). Based on the genome size and DNA concentration of each food source host, the number of copies of each spot was determined.  The equation used to calculate the number of copies was as follows N = [(X g/µL DNA/genome size) x 649] x 6.022 10²³. The analytical sensitivity for each qPCR-high resolution melting assay (qPCR-HRM) was assessed by generating dilution curves starting from 10⁴ genomic copies, followed by five serial dilutions (1:10) down to 100 copies. Each dilution curve underwent individual testing in qPCR-HRM, with triplicates for each data point and a final volume of 12 µL, utilizing the same concentrations of reagents as outlined in the previous section. The six reactions were subjected to the following thermocycling conditions: 95 ºC for 10 minutes, and 40 cycles to 95 °C for 15 seconds, 60 °C for 30 seconds, and 72 °C for 30 seconds.

Analytical specificity

Analytical specificity was assessed through an in vitro cross-test using controls to verify the specificity of each primer in the presence of the target host species.  Subsequently, analytical specificity was evaluated by testing DNA from other vertebrate animals that may be potential host feeding sources such as pig (Sus scrofa domesticus), sheep (Ovis aries), goat (Capra aegagrus hircus), quail (Coturnix coturnix), pigeon (Columbia livia), turkey (Meleagris gallopavo domesticus), rabbit (Oryctolagus cuniculus), guinea pig (Cavia porcellus), rat (Rattus norvegicus), marmoset (Callithrix sp.), capuchin monkey (Sapujus sp.), duck (Anas platyrhynchos domesticus), and capybara (Hydrochoerus hydrochaeris). The analytical specificity of each multiplex qPCR-HRM assay was evaluated using DNA samples from a single non-target vertebrate host species. Thus, each multiplex qPCR-HRM assay was tested separately with the DNA of different non-target host species.

For the molecular analysis, positive controls used were DNA samples of each host species as described in the “Control” subsection, and two negative controls were employed: UltraPure™ DNase/RNase-Free Distilled Water (Thermo Fisher Scientific, Wilmington, DE, USA).

Validation of the multiplex qPCR-HRM designed in field samples

Sampling

The specimens used in this study were collected from mosquito populations from both rural and urban areas of Seropedica, Rio de Janeiro, Brazil, spanning from July 2016 to August 2017. The mosquitoes were captured using CDC light traps placed in properties with animal breeding activities, operating continuously from 6:00 PM to 6:00 AM over three consecutive days.

Upon capture, mosquitoes were sorted by sex, feeding status, and genus. Blood-fed females, identified visually by their swollen red abdomens, non-engorged females and males were individually preserved in RNA later® solution at -20°C until molecular analyses.

Following identification according to the taxonomic key (Forattini's criteria, 2002), a total of 310 mosquitoes belonging to the Culex genus were selected. Prior to DNA extraction, the mosquitoes were submitted to removal of RNA later® solution. Each sample was washed three times with 1000 μL of phosphate-buffered saline (PBS) buffer 1×, and centrifuged at 19,000g for 15 min to ensure complete removal of RNAlater™. Subsequently, mosquitoes head were removed as previous research indicates that corneal pigment cells (ommatidia) can inhibit PCR reactions (Boncristiani et al., 2011; Senne et al.,2022).  To control for contamination during the DNA extraction, males and non-engorged females were included in each batch of DNA extractions from engorged females.

The subsequent DNA extraction process focused on the specimens’ abdomen, which were thoroughly crushed using a sterilized glass rod. This process unfolded in 2.0 mL microtubes containing 500 μL of a cell lysis buffer (comprising 400 mM NaCl, 20 mM Tris-HCl at pH 8, 10 mM EDTA at pH 8.0, and 1% SDS) along with 15 μL of proteinase K (20 mg/ml). The Salting Out protocol, as outlined by Ayres et al. (2002), was then employed for further purification.

Quantification of the extracted DNA was accomplished through spectrophotometry using Nanodrop® (Thermo Fisher Scientific, Wilmington, DE, USA). The concentration was standardized to 10 ng/µL, and the samples were stored at -20°C, ensuring the integrity and stability of the genetic material for subsequent analyses. The DNA samples were subjected to a conventional PCR (cPCR) assay targeting the cytochrome c oxidase I (cox1) gene (Folmer et al. 1994). The assay was used to verify the presence of amplifiable DNA and to exclude the presence of PCR inhibitors. Only positive samples were used in the validation process.

Identification of Culex species

Culex quinquefasciatus, Culex nigripalpus, and Culex coronator were selected for identification in field-captured mosquitoes due to their significance in epidemiological and ecological studies, as well as being among the most frequent mosquito species in the studied region (Forattini, 1996). The multiplex PCR developed by Kent et al. (2010) is an effective tool for identifying these species in a single reaction. A multiplex PCR employing the primers forward CXFOR 5’-ATCGCTGAAGTTGACCGAAC-3’; reverse NIGR4 5’-TACACGGTGCGCTAAGAGATACAC-3’ CORO3 5’-AGTACGCGCATTCCGACAG-3’; QUIN2 5’-ACTGTTTTATCGGTGGGTCGT-3’ targeting polymorphic regions of ITS1 was used to identify the sampled mosquitoes at species level: Cx. quinquefasciatus (580bp) Cx. nigripalpus (207bp) Cx. coronatus (244bp) (Kent et al., 2010). The molecular assay was performed using a final volume of 25 µL containing: 1X Buffer (Invitrogen®, Thermo Fisher Scientific, Wilmington, DE, USA), 3mM MgCl₂, dNTP 0.4 µM, 0.4 µM of each primer, 1U of Platinum Taq DNA polymerase, and 2 µL of DNA. The thermocycling conditions were initial denaturation at 95 ºC for 10 minutes, 35 cycles at 95 ºC for 1 minute, 60 ºC for 1 minute, and 72 ºC for 1 minute, and a final extension step at 72 ºC for 7 minutes. Samples that tested positive using the NIGR4 primer were subjected to digestion with the NcoI restriction endonuclease for the specific differentiation between Cx. nigripalpus and Cx. thriambus, following the protocol described by Kent et al. (2010).

Multiplex conventional PCR analysis

To validate the qPCR-HRM results, multiplex conventional PCR (cPCR) was performed using the same set of primers. The first multiplex cPCR was performed using a final volume of 25 µL compromising 1X AmpliTaq Gold ™ 360 Master Mix (Thermo Fisher Scientific, Wilmington, DE, USA), 400nM primers for dog, 400nM for human, 600nM for chicken and 10ng/µL of DNA extracted from engorged females of Culex sp. The thermocycling conditions were 95°C for 10 minutes, 40 cycles of 95°C for 15 seconds, 57°C for 30 seconds, and 72°C for 30 seconds. The second multiplex cPCR used the same PCR master mix and mosquito DNA concentration. Primer concentrations were 600 nM to cat, 600 nM to bovine, and 600 nM to horse. The thermocycling conditions remained the same except for the annealing temperature, which was set at 58°C.

For the molecular analysis, positive controls used were DNA samples of each host species as described in the “Control” subsection, and two negative controls were employed: UltraPure™ DNase/RNase-Free Distilled Water (Thermo Fisher Scientific, Wilmington, DE, USA).

Electrophoresis was performed in 3.5% agarose gels in 75V for 60min, subsequently stained with ethidium bromide (0.4 mg/mL) to detect the presence of amplification products in UV light.

Sequencing

The positive samples from the multiplex qPCR-HRM were further analyzed using singleplex cPCR, with a specific primer pair for each vertebrate host species serving as a feeding source. Amplicons were purified using ExoSAP-IT (Thermo Fisher Scientific, Wilmington, DE, USA) according to the manufacturer's recommendations. Sequencing was performed applying Sanger method with ABI 3730 DNA analyzer equipment (Applied Biosystems, Perkin Elmer, CA, USA) (Sanger et al., 1977). Sequencing results were analyzed in CLC Main Workbench (Qiagen, Valencia, CA, USA) and deposited to GenBank database.

Statistical analysis

The Kappa coefficient test at a 5% significance level was used to measure the proportions of agreement between qPCR-HRM and Multiplex-PCR to evaluate the performance of qPCR- HRM for molecular detection of blood feeding sources (Medronho, 2009). The level of agreement was calculated from a 2 x 2 contingency table comparing qPCR-HRM and Multiplex-PCR for the detection of one host species at a time. All statistical analyses were performed using BioEstat 5.0 software (Ayres et al., 2000).