Transcriptome Analysis of Deltamethrin Resistance Mechanisms in Aedes aegypti Populations from Metro Manila, Philippines Authors: Marc Ralph M. Solomon, Ma. Anita M. Bautista Published: 2026 ABSTRACT Background: Arboviral transmission by Aedes aegypti is exacerbated by insecticide resistance, which undermines effective vector control strategies. In the Philippines, deltamethrin resistance in several Ae. aegypti populations has been reported and prompts an investigation into the possible molecular mechanisms involved. Methods: Two Ae. aegypti populations were subjected to susceptibility assays to deltamethrin and were RNA-sequenced for transcriptome analysis. Results: The two (2) Ae. aegypti populations were resistant to deltamethrin and had LC50 values of 1.26 and 2.15 μg/mL at the zeroth generation (G0) that shifted to 2.21 and 2.08 μg/mL at the second generation (G2) after sublethal deltamethrin exposure, although not statistically significant. Relative to a control population unexposed to any insecticide, DGE analysis of the RNA-Seq data revealed significant upregulation of cytochrome P450, glutathione S-transferase, and carboxyl/cholinesterase genes in resistant populations. Moreover, GO and KEGG analyses revealed enrichment of cytochrome P450-related terms and xenobiotic metabolism pathways. Conclusions: RNA-Seq analysis results of selected genes show concordance with their respective gene expression measured via qPCR (R = 0.89; p = 0.00049). This study provides insights into the possible molecular mechanisms of deltamethrin resistance in Philippine Ae. aegypti populations, providing informed decision-making in vector management. 1. Introduction Aedes mosquitoes are the primary vectors of arboviruses such as Zika, Chikungunya, yellow fever, and dengue, which confer diseases that remain a global public health concern due to the increasing number of infections, morbidity, and mortality, owing to their vectors’ wide global distribution, high vector competence, frequent human bites, and survival in urban areas [1,2,3]. To date, there are still no reliable and efficient treatments, such as vaccines, for these arboviral diseases; hence, controlling the population of their insect vectors has been the primary strategy to prevent transmission and outbreaks [4]. The most common vector control strategy is the use of insecticides, particularly pyrethroids, which are extensively used among other insecticides owing to their low mammalian toxicity and high insecticidal effectiveness [5]. Pyrethroids are preferred active ingredients for indoor residual spraying (IRS) and long-lasting insecticidal nets, hence accounting for the majority of IRS coverage worldwide [6,7]. In the Philippines, where dengue is endemic, synthetic pyrethroids are commonly used for regular insecticidal fogging during outbreaks and after the rainy season [8,9]. Despite ongoing efforts to curb arboviral disease transmission using insecticides, more than 110,000 cases (437 deaths) were documented in the Philippines from January to May 2025, with 19,572 cases coming from Metro Manila (216% increase vs. 2024) [10]. Continuous incidences of infections are being linked by several studies to the overuse and improper application of pyrethroids, leading to increased selection pressure for the development of pyrethroid-resistant Aedes mosquitoes [7]. The World Health Organization pointed out that heavy reliance on pyrethroids has significantly increased resistance in Ae. aegypti, putting global efforts at risk of becoming ineffective [6]. Given that this dilemma promotes the transmission of mosquito-borne diseases, comprehensive surveillance and monitoring of insecticide resistance in the Philippines is imperative to reformulate the use and application of pyrethroids. However, studies tackling insecticide resistance in the Philippines are scarce [11,12]. There are two (2) known pyrethroid resistance mechanisms previously identified in Ae. aegypti, namely: target-site insensitivity from structural modifications in the insecticide’s target proteins, and increased expression of metabolic detoxification enzymes [13]. Non-synonymous mutations in the voltage-gated sodium channel (vgsc) gene result in target-site resistance, which confers reduced neuronal sensitivity of mosquitoes to insecticides [14]. Meanwhile, induction of detoxification genes enhances the expression levels of cytochrome P450 monooxygenases (P450s or CYP for genes), carboxy/cholinesterases, glutathione S-transferases (GSTs), and UDP-glycosyltransferases, which in turn promote insecticide metabolism [15]. Mutations resulting in this type of resistance, however, remain elusive because there are approximately 160 CYPs in Aedes mosquitoes, and the factors involved in CYP gene expression regulation are largely unidentified [2]. Among the pyrethroids, deltamethrin is widely used in Southeast Asian countries, including Malaysia, Singapore, Indonesia, and Thailand [4]. Due to its long history of use in the region, several studies have reported the development of resistance in Ae. aegypti to deltamethrin [16,17,18,19,20]. Underlying resistance mechanisms reported in these studies include target-site mutations, metabolic detoxification, reduced insecticide penetration through the mosquito cuticle, and behavioral changes [5,21]. In the Philippines, the information on insecticide resistance in Ae. aegypti is based on a susceptibility study in Manila [11], a pyrethroid susceptibility study in Ae. aegypti in Cebu City [12], and a report of insecticide resistance in Ae. aegypti in Metro Manila [22]. However, the molecular basis of resistance in these reports remains understudied; hence, this study sought to identify and characterize the underlying molecular mechanism of pyrethroid resistance in the Philippine populations of Ae. aegypti by sequencing and analyzing the transcriptomes of deltamethrin-resistant populations. 2. Materials and Methods 2.1. Mosquito Collection Mosquito eggs were sourced from the collections of Philippines’ Department of Health—Metro Manila Center for Health Development (DOH-MMCHD), obtained using ovitraps placed indoors and outdoors of selected households in Barangay Manggahan, Pasig City (PG13) (latitude from 14.6051° N to 14.59335° N and longitude 121.09304° E to 121.10635° N; 9–12 m above sea level), and Barangay Moonwalk, Parañaque City (PE07) (latitude from 14.49217° N to 14.48983° N and longitude from 121.01825° E to 121.02006° E; 12–13 m above sea level), both located in Metro Manila, during the month of June 2024. These sites were selected based on dengue incidence and deltamethrin usage in Metro Manila. Wooden paddles containing eggs were collected and transported to the insectary (ACL-4/BSL-2) of the National Institute of Molecular Biology and Biotechnology, University of the Philippines (UP) Diliman, for rearing and maintenance following the guidelines from the Joint Food and Agriculture Organization and International Atomic Energy Agency (FAO/IAEA) [23]. Larvae were fed with Brewer’s yeast, and rearing conditions were maintained at 27 ± 1 °C, 70–80% relative humidity, and a photoperiod of 12 h of light and 10 h of dark cycles, with one (1) hour each for dusk and dawn, inside a Fitotron® SGC2 insect growth chamber (Weiss Technik GmbH, Reiskirchen Germany). Emerged adult mosquitoes in rearing cages were maintained at 26 ± 2 °C. Adult Ae. aegypti mosquitoes were maintained with a supply of cotton soaked in 10% honey solution. 2.2. Deltamethrin Susceptibility Bioassay and Rearing of Deltamethrin-Resistant Mosquitoes Susceptibility to deltamethrin of Ae. aegypti field-collected mosquitoes from PG13 and PE07 was determined by performing the CONUS bottle bioassay protocol of the US Centers for Disease Control and Prevention (CDC) [24]. From this point, adult mosquitoes that emerged directly from eggs collected from the field (PG13 and PE07), or the zeroth generation (G0), are called Field-G0. For each population, six (6) Wheaton bottles were used, four (4) of which were insecticide-treated and two (2) were controls. Insecticide-treated bottles were coated internally with deltamethrin at a diagnostic concentration (0.75 μg/mL) prepared using absolute ethanol AR (Analytical Reagent Grade) (RCI Labscan Limited, Bangkok, Thailand), whereas the control bottles were coated only with absolute ethanol AR. Using a pooter, 20–25 non-blood-fed mosquitoes were introduced into each bottle. The number of alive and dead mosquitoes was counted upon release (time 0) and at the 30 min mark diagnostic time. Based on calculated percent mortalities, susceptibility of each population was determined following the phenotypic resistance frequency assessment guide from WHO presented in Table 1 [25]. To validate the susceptibility results from the CDC CONUS bioassay, Ae. aegypti eggs were collected from PG13 and PE07 in November 2024 and were reared to adulthood for WHO’s insecticide susceptibility tube testing, which was performed using 0.03% deltamethrin in silicone oil impregnated on filter papers [25]. The test was performed by introducing 20–25 non-blood-fed female mosquitoes into each tube with impregnated paper (5 replicates) and allowing exposure for one (1) hour. The number of dead and alive mosquitoes was counted after a 24 h recovery period. Two control tubes were used containing filter paper impregnated with silicone oil only. Resistance level was determined through median lethal concentration (LC50)-based modified bottle bioassay [26]. Seven concentrations with two (2) replicates each, including control, were prepared by 10-fold serial dilution of deltamethrin in absolute ethanol AR, producing the following concentrations: 0, 0.0075, 0.075, 0.75, 7.5, 75, and 750 μg/mL. The mosquitoes were exposed for one (1) hour and transferred afterward to untreated bottles for a 24 h recovery. Computed percent mortalities for each concentration after the recovery period were plotted for Probit analysis using the BioRssay package [27] in RStudio (version 2024.12.0+467) to obtain LC50 values. Resistant and deltamethrin-exposed Field-G0 adults that survived the assays were then allowed to lay eggs to perform further rearing with sublethal exposure to deltamethrin for two (2) generations (G2), generating Field-G2 adult mosquitoes. Exposures were performed to every generation of each population using their respective LC25 values calculated from the baseline (G0) LC50 assay results of PG13 and PE07. Surviving female and male mosquitoes (3:1) were allowed to mate for 24 h. Female mosquitoes were then fed with defibrinated pig’s blood from a slaughterhouse twice a week using an artificial feeder made of corrugated aluminum sheets covered with parafilm. Two days after blood feeding, black plastic cups lined with filter paper and filled 1/3 with seasoned (left to stand for 24 h) tap water were placed inside the cages and left for three (3) days to allow blood-fed female mosquitoes to oviposit. Filter papers with eggs were air-dried and stored in containers for subsequent rearing. 2.3. RNA Sequencing Total RNA was extracted from PG13 and PE07 Ae. aegypti mosquitoes, i.e., both from G0 and G2 subpopulations (reared from June 2024 collection) using TRIzol reagent (Invitrogen, Carlsbad, CA, USA). In order to perform differential gene expression analysis, RNA samples were also extracted from a population from the University of the Philippines Los Baños, Laguna, Philippines (UPLB-UNXP-G122) that has been reared in the laboratory for 122 generations and served as a control or reference population, based on the assumption that its gene expression profiles are at the basal level. Each RNA sample was a pool of five (5) non-blood-fed female mosquitoes [28]. Potential genomic DNA contamination was removed by TURBO DNase I (Invitrogen, Carlsbad, CA, USA). DNase-treated RNA extracts were checked for integrity, purity (A260/A280), and quantity using agarose gel electrophoresis (AGE), UV-Vis spectrophotometry using Jenway® Spectrophotometer 7415 (Cole-Parmer, Vernon Hills, IL, USA), and fluorometry using QubitTM RNA BR protocol in QubitTM fluorometer (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA), respectively. Prior to library preparation, we checked the RNA integrity number (RIN) of each sample using an Agilent 2200 TapeStation System (Agilent, Palo Alto, CA, USA). Sample concentrations were normalized to at least 1000 ng using nuclease-free water (NFW), creating final sample volumes of 50 μL. Using TruSeq RNA Library Prep Kit v2 (Illumina, San Diego, CA, USA), the cDNA libraries of the RNA samples were prepared following the manufacturer’s protocol. Libraries were quantified and checked for quality and then normalized to a 4 nM concentration using 10 mM Tris-HCl (pH 8.5) with 0.1% Tween 20. Libraries were then pooled, diluted to a final loading concentration of 400 pM, and loaded onto the NovaSeqTM 6000 Sequencing System (Illumina, CA, USA) with a NovaSeqTM SP flow cell, creating 100 bp paired-end reads. Sequencing was done at the Philippine Genome Center, University of the Philippines Diliman, Quezon City, Metro Manila, Philippines. The Illumina sequencing datasets were deposited in NCBI’s Gene Expression Omnibus (GEO) under Series accession number GSE305050, with raw reads archived in the Sequence Read Archive (SRA) under BioProject ID PRJNA1304288. 2.4. Transcriptome Analysis FASTQ sequencing reads were demultiplexed, trimmed to remove adapter sequences, and filtered based on a Phred quality score of 30 (Q30) through the sequencer-on-board NovaSeqTM Control Software v.1.8 (Illumina, CA, USA). Read qualities in each sample were assessed using FastQC v.0.12.1 [29], and those with scores less than 33 and lengths lower than 35 bp were trimmed using Trimmomatic v.0.40 [30]. Trimmed sequences were sorted and indexed by Samtools v.1.20 [31] and were mapped and aligned to the annotated reference genome of Ae. aegypti (GCA_002204515.1) AaegL5.3 [32] available in Vector Base database (https://vectorbase.org/vectorbase/app/, accessed on 7 April 2026) using STAR v.2.7.11b [33]. Features of the mapped reads to detect biases in sequencing and/or mapping were assessed using QualiMap v.2.3 [34]. Consensus transcriptome of all the samples was also assembled using Cufflinks v.2.2.1 [35] and StringTie v.3.0.0 [36]. Completeness of the assembly outputs from Cufflinks and StringTie was compared using Benchmarking Universal Single-Copy Orthologs (BUSCO) v.5.8.2 [37] in transcriptome mode using the diptera_odb10 (20 November 2019) lineage dataset with 3285 BUSCO markers. Prior to BUSCO assessment, output assemblies were filtered first using CD-HIT v.4.8.1 [38] to cluster similar sequences into representative transcripts to reduce redundancy and avoid artificial inflation in BUSCO completeness scores. To identify differentially expressed genes (DEGs), gene counts were generated using featureCounts v.2.0.8 [39]. Counts of each gene were size-normalized and analyzed for differential gene expression using DESeq2 v.1.46.0 [40] in RStudio (version 2024.12.0+467), with counts from the UPLB-UNXP-G122 population used as the reference for the relative expression of genes in the Field-G0 and Field-G2 populations. Using the lfcShrink() wrapper function of DESeq2, apeglm (Approximate Posterior Estimation for GLM) was called to perform Bayesian shrinkage estimation of logarithmic fold changes (LFCs) to minimize false positives from noisy low-count genes [41,42]. The p-values were also adjusted using the Benjamini–Hochberg (BH) method to account for false discovery rates (FDR = 0.1) (32,33). Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed on the DEGs with LFC greater than 1.0 and FDR-corrected p-values less than 0.10 using the g:Profiler2 package v.0.2.3 [43] in RStudio (version 2024.12.0+467). Using the Ae. aegypti genome assembly and annotation (AaegL5.3), upregulated DEGs were analyzed by categorizing them based on GO terms for biological processes (BP), molecular functions (MF), and cellular components (CC), as well as their association with KEGG pathways. 2.5. Quantitative PCR Validation Expressions of selected DEGs from the RNA-Seq results of the Field-G2 population were checked using quantitative PCR (qPCR) to confirm the validity of the DGE analysis. Primers were designed using NCBI’s Primer-BLAST (https://www.ncbi.nlm.nih.gov/tools/primer-blast/, accessed on 7 April 2026). Ribosomal protein L32 (rp49), with primer sequences: forward 5′-ACAAGCTTGCCCCCAACT-3′ and reverse 5′-CCGTAACCGATGTTTGGC-3′, was used as an internal control with an annealing temperature of 60 °C and a product size of 97 bp [44]. First-strand cDNA from the Field-G2 RNA samples was prepared using SuperScript™ III First-Strand Synthesis System (Invitrogen, MA, USA), following the manufacturer’s protocol. Using iTaqTM Universal SYBR® Green Supermix (Bio-Rad Laboratories, Inc., Hercules, CA, USA), qPCR reaction mix for each sample was prepared by creating 10-μL reactions composed of 2.5 μL NFW, 0.25 μL each of forward and reverse primers (10 μM), 5 μL iTaqTM Universal SYBR® Green Supermix (2X), and 2 μL cDNA Template (20 ng/μL). qPCR reactions were incubated in CFX Opus Real-time PCR System (Bio-Rad Laboratories, Inc., CA, USA), programmed with the following qPCR conditions using CFX Maestro Software (Bio-Rad Laboratories, Inc., CA, USA, version 2.3): initial denaturation at 95 °C for 5 min, 40 cycles of denaturation at 95 °C for 10 s and annealing and extension at 60 °C for 40 s, followed by final extension at 72 °C for 10 s, and finally melt curve analysis from 65 to 95 °C for 5 min. Standard curves were generated by amplifying each DEG in four (4) standard concentrations (0.02, 0.20, 2.00, and 20.0 ng/μL) of the samples in three (3) technical replicates. These standards were prepared by pooling the cDNA of each sample that was normalized to 20 ng/μL, and then performing a 4-point 10-fold serial dilution. Using CFX Maestro Software (Bio-Rad Laboratories, Inc., CA, USA), quantification cycle (Cq) values were analyzed to generate the standard curves with computed primer efficiency (E), goodness of fit (R2), and slope of the best-fit line. Measurement of the expression of the selected genes was performed using cDNA samples from the Field-G2 populations, PG13 and PE07, with three (3) biological replicates for each population and three (3) technical replicates for each biological replicate. Using the primer efficiencies, mean gene expression ratios of each DEG were then manually computed using Pfaffl’s method [45], log2-transformed and compared to the fold change values from the RNA-Seq analysis. Pearson correlation coefficient was also computed from the comparison of the two (2) methods to check for concordance in the expression of the selected genes between the methods. 3. Results and Discussion 3.1. Deltamethrin-Resistance in Ae. aegypti Populations from Metro Manila Based on the number of dead and alive mosquitoes after exposure, the following percent mortalities were computed in each assay (Table 2 and Table 3): PG13 and PE07 Ae. aegypti mosquito populations (Field-G0) exhibited deltamethrin resistance, indicated by low mortality rates of 1.92% and 3.45%, respectively, in the CDC CONUS bottle bioassay and confirmed by WHO tube assay with 2.94% and 6.25% mortalities recorded, respectively. Their corresponding LC50 (G0) values were 1.26 and 2.15 μg/mL, respectively (Table 4). These levels of resistance could be attributed to the prolonged and repeated use of deltamethrin. Since 1997, local government units (LGUs) in Metro Manila have consistently employed pyrethroids, including deltamethrin and permethrin, for space spraying as part of their vector control strategies [22]. This long history of pyrethroid use may have subjected these Ae. aegypti populations to selection pressure, favoring the survival of mosquitoes that developed a resistance phenotype to pyrethroids. The resistant PG13 and PE07 populations were reared up to the 2nd generation (Field-G2) with exposure to deltamethrin at LC25. The subsequent LC50 assay (Table 4) showed an increase in the LC50 of the PG13 population from 1.26 (G0) to 2.21 μg/mL (G2), suggesting a possible enhancement of resistance. However, this increase from G0 to G2 is not statistically significant due to the overlap in their corresponding confidence intervals (CIs). Similarly, the PE07 population exhibited a slight decrease in LC50 from 2.15 (G0) to 2.08 μg/mL (G2), which is also not significant. Although not statistically significant, it is important to acknowledge the possible biological implications of the observed trends [46]. Changes in LC50 following sublethal exposure may suggest a potential trend towards enhanced insecticide resistance in the field-collected Ae. aegypti populations. The increase could be due to molecular mechanisms activated upon exposure to insecticides like deltamethrin [47]. The wide range of CIs suggests low precision, which may be caused by the large gap between the deltamethrin concentrations used in the LC50 assay (10-fold increments) and biological variability among field-collected mosquitoes, possibly resulting in heterogeneous responses to deltamethrin exposure [46,48]. This limitation of the study may be addressed by using concentrations with smaller fold increments around the expected mortality range. 3.2. Ae. aegypti Transcriptome Assemblies with Duplicated Orthologs Total RNA extracts obtained from the Field-G0 and Field-G2 populations of PG13 and PE07 mosquitoes, including the control population UPLB-UNXP-G122 (Table S1), exhibited high quality (RIN 7.2 to 9.8) (Figure S1) and were used to produce cDNA libraries. Sequencing generated 136.89 Gb of paired-end sequencing data with approximately 1.28 billion reads passing the sequencer’s internal quality filtering (Table S2) and over 97% of the base calls having quality scores (Q-scores) greater than 30 (>Q30). Qualities were consistent at >Q35 before and after trimming, which retained approximately 99% of each sample’s paired-end reads. Alignment to the Ae. aegypti reference genome (GCA_002204515.1) [32] showed that over 99% of the reads successfully mapped, with very low mismatch rates ranging from 0.44% to 0.47%, confirming the overall accuracy and quality of the sequencing data (Table S3) [49]. The majority of the reads (73.9% to 93.1%) mapped to single, unambiguous locations in the reference genome, exceeding the commonly accepted threshold for successful RNA-Seq alignment (>70%) [50]. Using the annotated Ae. aegypti reference genome (AaegL5.3), the consensus transcriptome assemblies from all samples, as shown in Figure 1, demonstrated high completeness (~98%), with low percentages of fragmented (1.1–1.2%) and missing (0.7–0.9%) BUSCOs. Between the two (2) assemblers used, StringTie produced a slightly more complete assembly (98.2%) compared to Cufflinks (97.9%). However, Cufflinks yielded a higher proportion of complete single-copy BUSCOs (60.5%) and fewer duplicated BUSCOs (37.4%), in contrast to StringTie (44.5% single-copy; 53.6% duplicated BUSCOs). While a high proportion of duplicated BUSCOs is expected in transcriptome assemblies [51], it is notable that a substantial number of duplicates remained even after redundancy filtering. Given that the samples primarily consist of deltamethrin-resistant Ae. aegypti mosquitoes, the elevated level of duplicated orthologs observed may potentially reflect true biological duplications. It has been found that gene duplication events could be contributing to the overexpression of detoxification enzymes, as suggested by Bacot et al. [52], who reported that genomic duplications can play a role in enhancing the expression of genes involved in insecticide resistance in mosquitoes. Nevertheless, it is imperative to further investigate whether the observed duplicated orthologs are indeed genomic duplications contributing to overexpression of detoxification-related genes. 3.3. Overexpression of Detoxifying Enzymes Based on Differential Gene Expression Analysis Gene-level counts revealed that at least 50% of the reads per sample were successfully assigned to annotated gene features (Figure S2), except for two (2) samples (37.3% and 29.8%, respectively) due to observed bias in their read coverage towards the 3′ end of their transcripts (NCBI GSE305050) (Figure S3). Differential gene expression (DGE) analysis comparing the Field-G0 and Field-G2 populations to the control population, UPLB-UNXP-G122, identified 3563 DEGs in the Field-G0 and 1671 DEGs in the Field-G2 population. Based on an LFC cutoff of ±1, the Field-G0 population exhibited 406 significantly downregulated DEGs (118 annotated) and 2007 significantly upregulated DEGs (764 annotated) (Figure 2A). Meanwhile, in the Field-G2 population, 174 significantly downregulated DEGs (43 annotated) and 468 significantly upregulated DEGs (164 annotated) were identified (Figure 2B). In both populations, upregulated DEGs include several genes potentially linked to metabolic insecticide resistance, notably those belonging to cytochrome P450 (CYP), glutathione S-transferase (GST), and carboxy/cholinesterase (CCE) gene families (Tables S4 and S5). Most of the CYP genes identified belong to the CYP9, CYP6, and CYP325 subfamilies. These CYPs have been previously confirmed to be present in Ae. aegypti through multiple studies utilizing microarray, transgenic expression, and biochemical assays [53,54,55,56]. Notably, CYP9J28, CYP9J26, CYP9J24, CYP9M6, CYP6BB2, and CYP6N12 genes are consistently overexpressed in both Field-G0 and Field-G2 populations, each exhibiting LFC values greater than 2.2, corresponding to at least a 4.6-fold increase in expression. CYPs have been reported as the primary group of enzymes implicated in pyrethroid resistance in several mosquito species [57]. Generally, CYPs catalyze the oxidation of xenobiotics, producing oxidized metabolites that subsequently undergo conjugation reactions catalyzed by uridine diphosphate (UDP) glucosyltransferases (UGTs) or glucuronosyltransferases (UDPGTs) [58,59]. These conjugated products are then eliminated from mosquito cells via ATP-binding cassette (ABC) transporters [58]. Interestingly, this detoxification pathway may be supported by the current study, as UDPGT (AAEL008749) and ABC transporters (AAEL012700 and AAEL001938) were also observed to be upregulated (LFC > 1, adjusted p < 0.1) in both Field-G0 and Field-G2 populations (Tables S4–S7). Most of the overexpressed CYP genes in this study belong to the CYP6 and CYP9 families. These two (2) families are part of the CYP3 clan, which is widely recognized in insects for containing multiple-copy paralogs that are often arranged in tandem clusters on chromosomes [60]. These paralogs typically arise from gene duplication events, a mechanism increasingly implicated in metabolic resistance in mosquitoes. Recent studies have associated CYP gene duplications with insecticide resistance in Ae. aegypti [52]. Because CYP6 and CYP9 families are insect-specific, they may be particularly prone to evolutionary expansion through tandem duplications, leading to large genomic clusters [60,61]. Within these clusters, positive selection may act on a single gene that confers a strong detoxification advantage, like enhanced pyrethroid metabolism, and may trigger a selective sweep. This event allows neighboring CYP paralogs to hitchhike, increasing in frequency regardless of their individual functional contribution. If these clustered genes share cis-regulatory elements, co-regulation can occur, resulting in the simultaneous transcriptional activation of multiple CYPs. A similar phenomenon was observed in Anopheles funestus, where a selective sweep in a genomic region containing CYPs contributed to metabolic resistance against pyrethroids [62]. As the majority of upregulated CYPs belong to the CYP6 and CYP9 families, it is hypothesized that this pattern may suggest the presence of a duplication-rich genomic region under strong selective pressure from insecticide exposure. Gene duplication can increase gene dosage, resulting in elevated expression of detoxifying enzymes. This, in turn, enhances the mosquito’s capacity to metabolize and withstand insecticide exposure [63]. Taken together, this hypothesis about possible genomic duplications and selective sweeps concerning CYP6 and CYP9 gene families needs to be further investigated. Similar to CYPs, GSTs are also overexpressed in both populations, with LFCs ranging from 1.3 to 5.8 (2.5- to 55-fold). While most GSTs previously associated with insecticide resistance belong to the Epsilon class [64,65], the majority of overexpressed GSTs in this study belong to the Delta class, including GSTD4, GSTD3, and GSTD11. Notably, GSTD4 has been identified as overexpressed in microarray and mRNA-Seq studies of insecticide-resistant Ae. aegypti [66,67]. Additionally, several CCEs (e.g., CCEJHE4F, AAEL001517, and CCEAE1B) are also observed to be upregulated in the Field-G0 population. The observed upregulation of GSTs and CCEs, alongside CYPs, may reflect potential metabolic adaptations to the routine use of organophosphate- and carbamate-based insecticides in PG13 and PE07 [10]. The absence of upregulated CCEs in the Field-G2 population may also reflect the selective pressure exerted by two (2) generations of sublethal deltamethrin exposure, thereby favoring the expression of pyrethroid-specific resistance genes. Between the Field-G0 and Field-G2 populations, more upregulated detoxification enzymes were identified in the Field-G0 population, while no upregulated CCEs were observed in the Field-G2 population. This may suggest that pyrethroid resistance in PG13 and PE07 Ae. aegypti may be largely driven by constitutively high expression of detoxification enzymes, particularly CYPs, rather than by induction following insecticide exposure. As such, the performed sublethal exposure indeed has not significantly elevated their expression levels any further [60]. This interpretation is corroborated by the high LC50 values observed in PG13 and PE07 populations at G0, which were reared without prior insecticide exposure before being subjected to the LC50 bioassay (Table 4). The decrease in fold change and number of upregulated CYPs in the Field-G2 population may also be attributed to the manner of collecting the mosquito samples after exposure. The observed decrease in LFCs and fewer upregulated CYPs in the Field-G2 population could also be influenced by the timing and conditions of sample collection. Specifically, mosquitoes were collected 24 h after deltamethrin exposure during the LC50 assay, which may have coincided with a post-exposure recovery phase. A similar trend was reported in a time-series transcriptomic analysis by Mack and Attardo [68], who observed a global downregulation of detoxification genes 24 h post-permethrin exposure in Ae. aegypti. The authors hypothesized that this reflected a physiological return to basal gene expression levels, potentially as part of the mosquitoes’ recovery response after insecticide-induced stress. In addition, given the initial LC50 of PG13 and PE07 populations were already high (Table 4), subsequent sublethal exposures that further elevated their LC50 may have imposed fitness and energetic trade-offs. These trade-offs could have negatively impacted the expression of some genes, as cellular energy may have been diverted toward the expression of genes that are more efficient in conferring pyrethroid resistance or tolerance [47,69]. This hypothesis may be supported by the RNA-Seq analysis of the Field-G2 population, which showed downregulation of genes associated with energy metabolism, development, and reproduction, including glycerol kinase (AAEL008474), glucose dehydrogenase (AAEL004028), lozenge (AAEL007040), oviductin (AAEL007519), and sugar transporter (AAEL001257), among others. Lastly, the differential expression patterns observed may also be shaped by regulatory mechanisms that control the overexpression of detoxification enzymes. These mechanisms may have led to the selective activation of specific genes over others, contributing further to the mosquito populations’ adaptive responses to insecticide pressure [47]. Aside from detoxification enzymes, overexpression was also observed among genes associated with other resistance mechanisms. Notably, chitin synthase (AAEL002718) and several cuticle protein genes, which may be involved in cuticle thickening, a mechanism that reduces insecticide penetration, were upregulated in both populations [70]. Additionally, several odorant-binding proteins (OBPs) were also found to be upregulated. These OBPs are known to play roles in chemosensation, influencing excito-repellent behavior in mosquitoes, particularly by aiding in the detection and avoidance of xenobiotics such as residual insecticides [71,72]. Taking all these results together so far, with the UPLB-UNXP-G122 population used as the reference population for the DGE analysis, there is an important caveat that the observed overexpression of these genes may not be entirely because of deltamethrin resistance, but also influenced by population genetic background and long-term laboratory adaptation effects. Theoretically, this can be addressed using a susceptible Ae. aegypti population with a close genetic background to the field-collected populations, which was not available in the Philippines or in Metro Manila at the time of this writing. 3.4. Enrichment to GO Terms and KEGG Pathways Related to Metabolic Detoxification Further investigation of mechanisms underlying deltamethrin resistance by functional enrichment analysis of upregulated DEGs from both Field-G0 and Field-G2 populations showed significant enrichment of GO terms and KEGG pathways. In the Field-G0 population, GO:BP terms “transmembrane transport” and “monatomic ion transport” are enriched, suggesting alterations in ion channel activity or membrane permeability (Figure 3A). These may support the involvement of target-site insensitivity as a major resistance mechanism, where mutations in VGSCs result in conformational changes that impair insecticide binding and prevent prolonged depolarization [13,73]. Complementing this, enriched GO:CC terms including “membrane,” “plasma membrane,” and “postsynaptic membrane” (Figure 3B) may indicate that the upregulated DEGs are localized at or near the action site of deltamethrin, which targets membrane-bound VGSCs [74]. The enrichment of GO:MF terms such as “oxidoreductase activity,” “tetrapyrrole binding,” and other CYP-related functions may further indicate that metabolic detoxification is a primary resistance mechanism in Ae. aegypti (Figure 3C). Specifically, “tetrapyrrole binding”, “heme binding”, and “iron ion binding” are fundamental molecular features of CYP enzymes that enable their catalytic oxidation–reduction activity [75]. For the Field-G2 population, enrichment of GO:BP terms shifted toward “proteolysis”, “carbohydrate metabolic process”, and “microtubule-based process”, which may indicate a cellular stress response in Ae. aegypti following insecticide exposure (Figure 4A). Insecticide-induced stress can disrupt cellular proteostasis, leading to protein oxidation, misfolding, or damage. To mitigate this, cells activate protein degradation pathways, including upregulation of proteasomes via stress-inducible proteasome chaperones [76]. Furthermore, both stress responses and insecticide detoxification are metabolically demanding processes, necessitating increased carbohydrate metabolism to meet heightened energy requirements. These enriched terms suggest that the transcriptional focus in this population may have shifted from detoxification to stress and energy-related processes, possibly reflecting fitness and energetic trade-offs following repeated sublethal exposures. In addition to enrichment associated with target-site insensitivity, the GO:CC terms “endomembrane system” and “endoplasmic reticulum (ER) membrane” are also enriched in the Field-G2 population (Figure 4B), possibly further supporting the involvement of metabolic detoxification mechanisms in Ae. aegypti. CYPs responsible for the oxidative metabolism of pyrethroids are localized in the ER membrane, where xenobiotic detoxification occurs [77,78]. Beyond CYPs, other key detoxification enzymes, including GSTs and CCEs, as well as associated enzymes such as UDPGTs and ABC transporters, are also predominantly ER-resident [59]. This enrichment pattern may highlight the central role of the ER in coordinating metabolic responses to insecticide exposure. Similar to the Field-G0 population, enriched GO:MF terms in the Field-G2 population also include “oxidoreductase activity”, “tetrapyrrole binding”, “heme binding”, and “iron ion binding,” indicating possible consistent involvement of CYP-mediated mechanisms in deltamethrin resistance (Figure 4C). Additionally, both populations show enrichment in genes associated with “UDP-glucuronosyltransferase (UGT) activity” and GO:MF terms related to chitin and cuticle structure. As previously discussed, UGTs contribute to detoxification by facilitating the conjugation and excretion of oxidized insecticide metabolites, while overexpression of chitin and cuticle-associated genes suggests cuticle thickening. This structural modification enhances resistance by reducing insecticide penetration through the exoskeleton [58,70]. Enriched KEGG pathways, such as “xenobiotic/drug metabolism by cytochrome P450s”, “drug metabolism”, and “metabolism of xenobiotics”, are consistent between the Field-G0 and Field-G2 populations, potentially reinforcing the central role of CYP-mediated detoxification in the metabolic breakdown of deltamethrin (Figure 5). Overall, the results of the GO and KEGG enrichment analyses may supplement the in the presence of established resistance mechanisms in Ae. aegypti, particularly highlighting metabolic detoxification mediated by CYPs as a primary mechanism. This pattern is consistently observed in both the Field-G0 and Field-G2 populations. 3.5. RNA-Seq Results Validated by Quantitative PCR Primers were designed for 10 selected DEGs including the overall top three upregulated DEGs (AAEL012664, AAEL004275, and AAEL002138), the overall top three downregulated DEGs (AAEL003297, AAEL013198, and AAEL012565), the top upregulated GST (AAEL001054), and the top three upregulated CYPs (AAEL014617, AAEL014613, and AAEL014893) identified in the Field-G2 population (Table S8). Table S9 shows primers’ efficiencies ranging from 84.7% to 112.2%, which are between the acceptable range (80–120%) for reliable qPCR quantification [79,80]. Using these primers, normalized expression levels of each selected DEG were assessed in the Field-G2 population relative to the UPLB-UNXP-G122 control population (Figure S4), log2-transformed, and were then compared to their corresponding LFC values obtained from the DGE analysis of the RNA-Seq dataset (Figure 6). Comparison between the two (2) methods shows that qPCR confirms the technical validity of the observed direction of expression changes, both upregulation and downregulation, from the DGE analysis results. The top upregulated genes, including the selected GSTs and CYPs, are consistently validated as upregulated, and the same trend holds true for the downregulated genes. However, as shown by the trend lines in Figure 5, discrepancies in the magnitude of fold changes are observed between the two (2) methods, particularly among the upregulated DEGs. These minor variations are expected, as qPCR and RNA-Seq employ different quantification methods, sensitivity thresholds, and normalization strategies [81]. Moreover, biological variability from the use of field-collected mosquitoes, reflected by the wide error bars, can significantly influence RNA-Seq results [82]. Nevertheless, the overlap of error bars between qPCR and RNA-Seq data, along with consistency in the direction of gene expression changes, indicates concordance in the expression of the selected genes between the two (2) methods, supporting the findings of the RNA-Seq analysis. Furthermore, analysis of fold expression changes in each selected DEG using Pearson correlation revealed a strong (R > 0.80) and statistically significant (p < 0.0005) positive correlation (Pearson R = 0.89). This concordance further supports the reliability of the observed gene expression levels from the RNA-Seq analysis. 4. Conclusions Deeper understanding of the molecular mechanisms underlying insecticide resistance is critical for improving vector control strategies. Overall, this study may potentially provide evidence of deltamethrin resistance in Ae. aegypti populations from two (2) cities in the Philippines, demonstrated at both the phenotypic and molecular levels. This resistance may possibly be driven by the overexpression of CYP genes, as well as GST and CCE genes, which enhance the metabolic detoxification of pyrethroids, thereby reducing the effectiveness of these insecticides. However, as mentioned previously, the differential gene expressions may also be influenced by different genetic backgrounds and adaptation between the field populations used and the control population, UPLB-UNXP-G122. Nevertheless, these results may inform future resistance surveillance, monitoring programs, and the refinement of current vector control strategies in Pasig City and Parañaque City, Metro Manila, Philippines.