Genomic Surveillance and Intervention on Dengue Virus in an Urban Setting in the Philippines

Abstract

This is part of the ReMoVE Dengue Program (i.e., research on mosquito, virus, and eco-socioeconomics of dengue) initiated under the auspices of the National Research Council of the Philippines, which started in 2012 aimed to develop locally adapted technologies, products, and systems, which would control the spread of dengue virus and reduce the eco-socioeconomic impact of dengue. Here, will be reported the results of the genomic surveillance of community-collected mosquitoes from a dengue hotspot community of Barangay Old Balara in Quezon City, Philippines using serotype-specific dengue PCR, and the developed antisense RNA product platform for dengue virus control based on surveillance results. Implications and recommendations for this work are outlined.

1. Introduction

Dengue remains to be a major problem in several Metro Manila cities and in the entire country. Since 2011, dengue cases in the Philippines continue to rise at an average rate of 3900 cases per year, with recorded cases of 34,940 in 2022 [1]. Among the regions, Central Luzon is with 6641 or 13%; Central Visayas, 6361 or 12%; and Zamboanga Peninsula, 4767 or 9% were the top contributors [2]. The increasing cases reflect a number of underlying scenarios and causes, which are difficult to pinpoint, although, one thing is clear, the current strategies for control and mitigation may not be as successful in containing the growing problem of dengue. The Philippines has stood as first in dengue cases globally, like the first recorded dengue epidemic in Southeast Asia that occurred in Manila in 1954, and the highest dengue case contribution ever recorded globally in 2019 of 437,563 cases [3]. It is amazing that despite government efforts and programs, this mosquito-borne disease continues to successfully become endemic and ravage the population [4]. Perhaps, a series of unfortunate events contributed to this unsuccessful mitigation, like the “lack of empowerment among the stakeholders in taking responsibility for dengue prevention” despite the Philippine government’s established National Dengue Prevention and Control Program in 1993 as well as the failed Dengvaxia vaccine program launched in 2016 [3], although other reasons may underlie this scenario. Nevertheless, optimistic perspectives remain as new research shed light on better strategies for control and mitigation [5]. Among these strategies, the dynamics of the virus-vector interaction and the phenomenon-based targeting may hold the key to dengue’s long-term prevention and control (Figure 1).

Figure 1.

The conceptual framework for genomic-based surveillance and intervention. A. Dengue as a threat to all with breaches in vector control and host protection. B. Genomic surveillance and intervention as a cornerstone in the fight to control dengue.

Aedes aegypti is the primary vector of the dengue virus, although Aedes albopictus has also been identified as a minor vector [6, 7]. Apparently, as more urban communities expand (i.e., which is a common trend among cities with increasing population) to cover semi-urban, semi-rural, and forested areas, the Aedes mosquitos have adapted to survive and breed in water pools and deposit in these areas. The Philippine Department of Science and Technology’s Ovicidal and Larvicidal (DOST’s-OL) trap technology was adopted in 2011 as a widespread strategy for controlling mosquitos [8]. The OL-trap technology involved the use of agents that can kill mosquito eggs and larvae in stagnant freshwater containers that serve as traps. Meanwhile, another kind of trap, the Orbi-traps has been validated as means to monitor mosquitos in different localities [9]. In particular, the Orbi-trap procedure has been utilized in monitoring A. aegypti mosquitoes and correlated with dengue cases in Manila [9]. Following a simple mosquito trap design [10] with modifications, adult mosquitos may be caught and morphologically identified. Further propagation of this scheme through a DOST invention [11] could be augmented by monitoring the virus present in the collected mosquito through a PCR analysis, thus may give early advice on the type of dengue virus circulating among the mosquito population in the community.

The PCR platforms for dengue virus detection are widely available in various institutions, the academe among others, and may be utilized for community-based surveillance, especially when appropriate community-academic institution link-up is established, especially now with several molecular laboratories with PCR machines being underutilized as COVID-19 testing declines [12, 13, 14]. Furthermore, what is needed would be a system for sample collection, the reagents for the genomic surveillance work, and an online-based reporting system accessible to the community, a model considered and espoused in this project [15]. PCR protocols for routine MDRTB, H1N1 virus, and leptospirosis detection have been established in the Department of Biochemistry and Molecular Biology—University of the Philippines College of Medicine (DBMB-UPCM). The laboratory had an extensive experience with MIRU-VNTR analysis and sequence analysis, including the use of appropriate bioinformatic software. Application of similar strategies to the dengue virus would be convenient.

Genomic-based interventions could be installed in the Orbi-traps, such as a mosquito-feeding device, where anti-dengue interventions can be incorporated into the blood formula. Several studies have demonstrated the positive response of mosquitoes in feeding warm blood [16, 17, 18, 19]. The feeding device, however needs to be designed, although “blood-filled condoms” have been reported to work in luring mosquitos [19, 20]. Anti-dengue DNA vaccines have also been reported to elicit an immune response in humans [21]. Meanwhile, mosquitos can be ideal vaccine targets, capitalizing on their endogenous defense system to block the spread of the dengue virus [22].

A set-up where a suitable container, such as a condom, with a chemical-based heating system [20] and incorporated with lactic acid and carbon dioxide would be ideal to attract mosquitos to the feeding device and insure consumption of an anti-dengue vaccine. The DBMB-UPCM has reasonable experience in recombinant DNA work, including the design and production of plasmids for various uses, including sequencing, expression, and DNA vaccines. Some constructs reported in the literature can be tested in the process.

DNA vaccines have been demonstrated in the control of West Nile virus via vaccination of American robins—the intermediate host involved in viral amplification that is feeding on Culex mosquitos [23]. Meanwhile, the mosquito defenses against the dengue virus have been studied to involve the JAK–STAT pathway, where a specific RNAi-based inhibition of PAIS or protein inhibitor of activated STAT results in increased survival of mosquitos from bacterial or viral challenge [24]. Similarly, an oral administration of DNA nanoparticles synthesized by complexing plasmid DNA with chitosan, a natural biocompatible polysaccharide, was shown to result in transduced gene expression in the intestinal epithelium [25]. Furthermore, the Wolbachia wMelPop strain, an endosymbiotic bacterial pathogen was found to be transferrable from D. melanogaster to the mosquito A. albopictus with the consequential effect of reduced longevity and fecundity, and high embryonic mortality [26]. While in A. aegypti, increased locomotor activity and metabolism were reported [27]. Thus, a protocol involving orally delivered DNA construct that would modulate the mosquito immune response combined with bacterial coinfection would manage dengue viral and bacterial residency in the vector, thus presenting an avenue for combined antiviral and bacteria-based control. This concept was applied in a study, where a cationic liposome was utilized to deliver an expression construct with the gene for Ae. aegypti thioester-containing proteins (AeTEPs), (i.e., involved in the control of flavivirus infection), resulting in reduced dengue virus infection [22].

This project was proposed to add value to a program of wide-scale mosquito monitoring by surveillance of the virus present in the collected mosquito by PCR analysis of its DNA/RNA extract and provide advice on which dengue type is circulating in the mosquito population of a given site. Furthermore, in the Orbi-trap, a mosquito-feeding device could be installed, where an anti-dengue DNA vaccine could be introduced.

In the first year of the project, a study to evaluate a holistic vector control program was embarked, which involved strategies, such as genomic surveillance and intervention, herbal-based larval destruction, irradiation-induced sterility of mosquito, and biocontrol-organism based larval control among others (ReMoVE Dengue Program—research on mosquito, virus, and eco-socioeconomics of dengue). Mosquito traps were installed in three sites in Barangay Old Balara, Quezon City, at 15 houses per site and 6 traps (i.e., 3 inside and 3 outside) per household, with GPS coordinates determined. The captured mosquitos were counted bi-weekly from May 2012 to January 2013 and serotype-specific dengue PCR was used for monitoring viral presence. In years 2 and 3, monitoring work was continued in the three sentinel sites (i.e., three houses per sentinel site). Also started the development of anti-dengue dsRNA as well as the validation of trans-ovarian dengue transmission and virulence testing in a mouse model. This report outlined the findings of this community-based study.

2. Materials and methods

2.1 Sampling and nucleic acid extraction

The experimental site has been identified and mosquito traps have been set up in a total of 45 houses (15 houses per site with 3 traps indoors and 3 traps outdoors). The sites were Area 1- Luzon, Area 1- Old Balara, and Area 4- Sitio Payong, with prior consultation and approval of the Quezon City Health Department (Figure 2G). Two Orbi-traps per site and six OL-traps per household (3 indoors and 3 outdoors) were installed.

Figure 2.

Mosquito collection summary and dengue PCR test results of the three sampling sites A. Luzon, B. Old Balara, and C. Sitio Payong in year 1. The red highlights indicate the households are positive for dengue, and the yellow is negative for dengue. Entries with zero values and yet show red highlights indicate the late emergence of mosquitoes from the collected sample, which comes from the eggs. Note the increasing number of houses with positive for dengue more than 3–4 houses, starting from the sixth to the eighth collection. There were houses that consistently show dengue positive mosquitos. Mosquito counts of the three sampling sites: D. Luzon, E. Old Balara, and F. Sitio Payong. Most of the houses yielded in their vicinity (outside) a count of 0–50 mosquito individuals (i.e., larvae, pupa, and adult), a moderate number of houses with 50–100 counts, and a few houses with 100–150 counts across the different collection times. In terms of counts greater than 150, Old Balara had more instances across the different collection times followed by Sitio Payong and lastly by Luzon. G. Project experimental site in Quezon City, Philippines [28].

The Barangay Health Workers (BHW) together with the project science research assistant conducted the biweekly sample collection. The Orbi-traps were utilized to monitor the adult mosquitos, while the OL-traps were utilized to monitor the egg and larval stages. The collected samples were sent to the DBMB-UPCM, where all field samples were stored, counted, identified, and processed, for RNA extraction and dengue detection by PCR. A small area (i.e., a mosquito insectarium, Figure 3B) has been set up for growing larvae collected from the field prior to molecular analysis. Preserved samples per collection receptacle (Orbi-trap or OL-trap at 1–50 mosquitoes and in cases exceeding 50, random sampling was done) were pooled and processed for RNA extraction using Qiagen RNEasy Kit following the manufacturer’s protocol.

Figure 3.

Antisense RNA trial testing on mosquitoes in the community. A. Preparation of antisense RNA formula with honey solution. B. Simulated mosquito setup in the community for the trial testing, collection of specimens from ovi-traps, and the insectarium used in the lab for handling and hatching the specimens. C. Electropherogram of dengue-PCR results of antisense RNA fed vs. non-fed specimens. Representing TS1-serotype 1, TS2-serotype 2, TS3- serotype 3, and TS4-serotype 4 conducted in different locations, Area 1—Old Balara and Area 4—Sitio Payong. Notable is the specific disintegration of signals in fed vs. non-fed and brown sugar only. D. Dengue-PCR of larva specimens from fed samples with the persistence of serotypes 2 and 4.

Reagents, materials, and samples were procured for the project, including an electronic air temperature and wind velocity meter, mosquito traps, primers, and laboratory and office supplies.

2.2 Detection of dengue virus by reverse-transcriptase-PCR

2.2.1 First strand synthesis

Following the protocol of Lanciotti et al. [29], target viral RNA was converted to a DNA copy (cDNA) using reverse transcriptase (RT) and the dengue virus downstream consensus primer (D2). The first strand synthesis was done using the Omniscript or Promega (Qiagen, Macare Philippines, Golden Bat (Far East) Inc., respectively) following the manufacturer’s protocol.

2.2.2 Polymerase chain reaction

Serotype-specific amplification was done from the cDNA template with the upstream dengue virus consensus primer (D1) and the downstream serotype-specific primers (i.e., TS1, TS2, TS3, and TS4). Target cDNA was amplified in 10-μl volumes containing the following components: 10 mM Tris (pH 8.5), 1.5 mM MgCl2, 10 uM each of the four deoxynucleotide triphosphates, 10 pmol each of primers 1 (D2) and 2 (i.e., either of TS1, TS2, TS3, and TS4) and 0.5 U of TopTaq (Qiagen, Macare Philippines) or GoTaq (Promega). The PCR reaction profile consists of the following: initial denaturation (94°C, 1 minute), and then to proceed with 35 cycles of denaturation (94°C, 30 s), primer annealing (55°C, 1 min), primer extension (72°C, 2 min), and followed by a final extension step of 72°C for 10 min.

2.2.3 Agarose gel electrophoresis

The PCR products were analyzed by gel electrophoresis on a 2.5% agarose gel (Vivantis) containing Gel Red (0.5 ug/ml), with the settings of 75 volts, for 40–45 minutes. A band on the agarose gel of the correct size was interpreted as a positive result. A faint band of the correct size was considered an equivocal result.

2.3 Anti-dengue dsRNA study

Primers targeting conserved regions in the UTR-Core gene were designed. The primers are:

UTR36 5′-GCTTAACGTAGT(T/G)CTAACAGTTT-3′ 62 deg

CAP521rc 5′-AACATGTGCACCCTTATAGCGA-3′ 64 deg

T7UTR36 5′-GAAATTAATACGACTCACTATAGGGGCTTAACGTAGTKCTAACAGTTT-3′

T7CAP521 5′-GAAATTAATACGACTCACTATAGGGTCGCTATAAGGGTGCACWTGTT-3′

The translation product of the target region is shown in Figure 4 panel A, and the region targeted in the viral genome is shown in panel B. The primers are used in the subsequent cDNA and dsRNA synthesis of RNA extracted from female A. aegypti mosquitos infected with DENV obtained from Barangay Old Balara, Quezon City. The synthesized dsRNA is administered to a group of mosquitos alongside a parallel treatment of control RNA with mosquitos coming from the same population. The presence of the dengue virus serotypes, after 2 days posttreatment, is detected through nested PCR.

Figure 4.

A region in dengue viral genome targeted for dsRNA.

3. Results

3.1 Sampling and nucleic acid extraction

In the initial year of community genomic surveillance, 12 sampling events were conducted. Mosquitos from the first to the twelfth sampling were tested for dengue-PCR using standard protocols as described. An increasing trend of more than 3–4 houses per site was found positive for dengue (Panel A–C, Figure 2). Twelve sampling data points for Area 2- Luzon, Area 1- Old Balara, and Area 4- Sitio Payong were uploaded to the project website [15]. The goal was to provide online access to the Barangay Health workers and use the information in their search and destroy program for the breeding ground of the mosquitos. This way, their campaign will be focused on the critical spot in the community. The other objective also was to guide the community as to which areas to avoid as possible exposure sites for the dengue-infected mosquitos.

The mosquito counts from the first to the twelfth collection were plotted as shown in Figure 2 (Panel D–E). It can be noted that variable counts were obtained for each household. Most of the houses yielded mosquitos in their vicinity (outside) with counts of 0–50 mosquito individuals (i.e., larvae, pupa, and adult). A moderate number of houses yielded counts of 50–100 mosquito individuals and a few houses with 100–150 counts across the different collection times. There were collection times where these counts were exceeded; and in Area 1- Old Balara, in particular, there were two instances, where it had counts exceeding 150 specimens per collection, followed by Area 4- Sitio Payong and lastly by Area 2- Luzon. This result was correlated with the cleanup program of the community and the prevalence of dengue cases.

Mosquito samples were submitted to the Research Institute for Tropical Medicine (RITM), Dept of Medical Entomology, for taxonomic identification. The results showed a 100% match for the preliminary identification in the lab and those identified in RITM, where most of the samples are A. aegypti and a few are A. albopictus and Culex sp. (See Figure A1).

A coordination meeting with RITM was conducted toward organizing a dengue study group. The RITM Virology Lab shared a protocol to detect dengue by RT-PCR. This procedure was optimized to detect the dengue virus in mosquitoes and was used in the analysis of the specimens collected from the different communities, including those submitted by the Philippine Nuclear Research Institute (PNRI, which were reared in the PNRI Mosquito laboratory for several generations) (Figure 5). Rearing of Ae. aegypti larvae were done using deionized water and commercial fish meal (Tetramin, Tetra GmbH) at 0.02 mg/larva/day. Pupae are collected as soon as they develop. Adults were confined in a rearing cage (1 ft³) and fed with a 10% sugar solution. Adult females were blood-fed using immobilized live mice. Egg collection was done using an egging cup (40 mL cap.), containing about 10 mL deionized water and lined with white filter paper for oviposition. Ae. aegypti was reared in laboratory conditions with a mean temperature of 27°C, relative humidity of 70%, and photoperiod of 12:12 (light: dark).

Figure 5.

Results of selected mosquito samples analyzed for the 4 serotypes. Lanes 21 and 31 indicate the presence of serotype 2 and 3 in the same male Aedes sample and only serotype 2 (lane 22) for the female sample. Note the multiple bands, indicating multiple genotypes for each serotype.

Initial results indicated that 8 out of 10 specimens are dengue positive, with both male and female mosquitos showing positive results for the dengue virus infection. These infections could be vertically transmitted (maternal to offspring) as the samples were obtained from hatched larvae that were reared to the adult stage (Figure 6). The same pattern was observed in the samples collected from the traps set up in Barangay Old Balara where larvae were made to hatch, and the emerging adult mosquitos were analyzed for the presence of the dengue virus. Results showed that multiple dengue serotypes could infect one mosquito and multiple genotypes within a serotype can also occur (Figure 5). Vertical transmission has been reported to occur among dengue viruses in the Aedes mosquito [30, 31, 32]. The findings of this study are consistent with these previous reports.

Figure 6.

Distribution summary of mosquitoes collected from the three sentinel sites in Bgy Old Balara. Area 1 (Old Balara), Area 2 (Luzon), and Area 4 (Sitio Payong) in year 2. Firstto eleventhcollection (A). Table 5-1 summarize the counts of specimens collected (B). Dengue-PCR results of the specimens collected and scoring best on the degree of dengue positivity (C-D) with red meaning strongly positive and yellow as weakly positive.

In the second year of genomic surveillance in the community, eleven sampling events were established and monitored. The mosquito collection from the first to the eleventh collection was processed, recorded, and summarized as shown in Figure 7. It was seen from the trend that there was a persistent prevalence of mosquitos in Area 2 followed by Area 4 and least by Area 1. As expected, most collections were larvae and pupae with most samples coming from Area 2, followed by Area 4, and lastly by Area 1 (Figure 6).

Figure 7.

Antisense RNA synthesis and trial testing on mosquitoes. A. Preliminary antisense RNA synthesis results. B. Results on antisense RNA representing 4 days and 7 days feeding with RNAi. TS1-TS4 represents the four dengue serotypes. Notable is the disintegration of signals in 4 days (i.e., shearing of DNA) as compared to brown sugar and 7 days (i.e., faint shearing of DNA). The antisense produced also affords protection of different serotypes. C. Trial 3 results in antisense RNA representing 4 days feeding with RNAi and comparison of different RNAi types. Notable is the specific disintegration of signals in TS2 but not TS1 as compared to brown sugar only. The lowest bands are primer dimers.

Areas 1 and 4 are more forested communities as compared to Area 2, which contrasts with the expected pattern that A. aegypti tends to prefer forested areas. The low trapping yield in Area 1 could also reflect on the anti-dengue mosquito program of the community as the community health center is located in Area 1.

It can also be noted that the pattern where the majority of the strongly dengue-positive mosquitos were those collected in the second to the sixth collection, which were from August to November and began to decline in the seventh to the eleventh collection, which was from December to February. Most of the strongly dengue-positive specimens were found in Areas 2 and 4. These findings on the dengue PCR pattern tend to support the collection data in Figure 6. This means that vector surveillance with accompanying RT-PCR detection of the dengue virus serotypes can provide an additional layer of information that would reflect the seasonal variations of dengue infestation of the mosquito vector as well as the dengue management program of the community.

3.2 Development of antisense RNA for dengue virus in mosquito

The RNAi protocol developed was implemented, and was able to yield antisense RNA products for testing on mosquitos. Two trials were conducted on RNAi-based inhibition of the dengue virus in the Aedes mosquito derived from Barangay. Old Balara is shown in Figure 7. Evaluation of a mosquito feeding device for RNAi construct delivery system was not pursued as direct brown sugar feeding was found to be a simple strategy considering that RNA is stable in a sugar solution. Apparently, the simpler the delivery system is, the better it will be for community-based intervention. In the result shown in Figure 7, we can see that the feeding of antisense RNA in 4–7 days resulted in the shearing of dengue viral RNA. The mosquitos mostly die on the fourth day, indicating the lethality of the antisense RNA. It was also observed that those mosquitos that survived on the 7th day showed clearance from all serotypes.

The test was repeated to evaluate which of the following antisense RNAs will best work: dsRNA non-hybridized, dsRNA hybridized, ssRNA+ strand, or ssRNA-strand. The results of this experiment are shown in Figures 7 and 3. Only the dsRNA hybridized showed activity in the specific inhibition of TS2. There is an endogenous reduction in TS2 dengue signal with sugar alone, while intensification of signal with either ssRNA+ strand or ssRNA-strand was observed. In this sample, no serotype 3 or 4 was present. It can be noted that TS1 is not inhibited when the designed RNAi is generic.

3.3 Vertical transmission of dengue in in-house propagated mosquito stocks

Specimens from F1 to F15 generation obtained from the PNRI were analyzed with dengue PCR results shown in Figure 8. It can be observed that a persistent occurrence of the dengue virus from F1 to F10 samples (variable pattern could arise from a random sampling of 3 M/F/E samples). We observed the presence of different serotypes in one mosquito in females, males, and egg samples. We also observed the presence of different genotypes in one serotype (multiple bands were verified by sequencing, which will be reported in another paper). A yield of dengue-free eggs was observed in F10. Thus, the transovarial transmission of the dengue virus in local A. aegypti mosquitos has been verified. We also analyzed a batch sample of mosquito eggs consisting of 20 eggs. The dengue PCR results show only four eggs positive with serotype 1 and none for all the other serotypes (See Figure A2). This shows that the infection rate for vertically transmitted dengue virus template from parent to egg is approximately 20%.

Figure 8.

Summary dengue PCR results of F1 to F15 generation of mosquito samples in-house bred in PNRI. Dengue-free eggs emerged in F10 as shown in green. Dengue-positive mosquitoes are shown in yellow (i.e., weakly positive) and red (i.e., strongly positive).

Whether the amplicons detected through this dengue-specific PCR represent authentically, and live viruses may require definitive proof by DNA sequencing. Such will be presented in another paper.

A study on transovarial transmission of dengue in correlation with virulence in mice was conducted by our graduate student (i.e., Mr. Ralph Bawalan—MS Trop Med) in collaboration with Dr. Nelia Salazar—RITM Entomologist. A mouse model for dengue testing was developed. This model was able to show histological similarities with humans as well as pathological symptoms of thrombocytopenia and fever (See Figure A3).

3.4 Genomic surveillance in year 3 and strategy for mitigation

The genomic surveillance was continued for Area 1-Old Balara, Area 2-Luzon, and Area 4-Sitio Payong through their sentinel sites. The mosquito/larvae/pupae collected from the sites from first to the nineth collection and their respective dengue PCR results are summarized as shown in Figure 9. It can be observed that at all sites, there was a strong positivity for the dengue virus, indicating a worsening dengue infection from the community mosquito population after 3 years of surveillance. Apparently, the 4S strategy of the community does not seem to work sustainably in containing the spread of the virus. How this correlates with the dengue disease burden of the community may have to be closely evaluated.

Figure 9.

Distribution summary of the genomic surveillance in year 3. A. Table 8-1 contains the consolidated mosquito specimens collected from the different sites. B. Dengue PCR amplicon electropherograms of community specimens. C. Dengue-PCR results of the specimens collected and scoring based on the degree of dengue positivity with red meaning strongly positive.

Figure A1.

Entomological report certifying the taxonomic identity of the collected mosquito from the community.

Figure A2.

Electropherogram of the PCR products of 20 mosquito eggs from F10 generation showing approximately 20% infection rate.

Figure A3.

Viral expansion through neonate intra-cranial injection of RNA extracts from community-collected mosquito specimens and virulence assay in mice of the subsequent generation.

The remaining mosquito larvae’ homogenates and extracts were evaluated for their potential use as templates for the antisense experiments. The plan was to amplify the dengue virions that can be recovered from intra-cranial injections in suckling mice, a procedure that was previously optimized by our team (i.e., based on Figure A3).

A community trial of the antisense RNA formulation was installed in Area 1 (Old Balara) and Area 4 (Sitio Payong). The fourth and sixth batches of collected samples that are positive for dengue virus (serotypes 1 to 4) were utilized in the experiment (Figure 3).

After 2 weeks of exposure, specimens were collected, extracted, and tested for dengue-PCR. The results showed low to the absence of signals in antisense RNAi-fed.

specimens as compared to those that were not fed. The results are consistent with the previous findings. Even in the community environment, the antisense RNA preparation was able to inhibit the dengue virus transmission in the mosquito (Figure 3). It was also noted that in an ovi-larvae trap model, there was inhibition of the dengue virus, which is carried over to the emerging mosquito. This is a new indication of the developed antisense RNA.

4. Discussion

Since 2012, three studies have been conducted in the Philippines that verified and validated the natural vertical transmission of the dengue virus in a community population [12, 13, 14]. Similarly, there were seasonal variations in dengue positivity in the mosquito, mostly occurring in the rainy seasons of August to November, which is aggravated by possibilities of multiple serotypes and multiple genotype patterns, indicating that the next generation of eggs laid by a dengue-infected mosquito become the melting pot for possible dengue virus recombination. Apparently, the data also indicate that the A. aegypti mosquito has been successfully adapting, breeding, and successfully thriving as a coexisting dengue vector in urban communities and the absence of forested areas no longer limits its geographical spread. Through the years, while the Aedes mosquito has continued to infiltrate the urban communities, the mitigation strategy implemented by the Department of Health to all the Barangay communities has not taken major leaps and still follows the 4S Program (DOH website). Thus, updated and leveled-up interventions may have to be implemented and integrated into community-based strategies to be able to see concrete progress in dengue intervention. As outlined in the conceptual framework in Figure 1B, the genomic surveillance strategy for the dengue virus harbored by the natural stocks of mosquitos captured from ovi-traps in sentinel sites in the community may have to be set up in coordination with the barangay health center and neighboring academic or coordinating research institutions with existing PCR facilities for routine molecular detection followed by online reporting of results to allow quick action of the community to implement various interventions. Furthermore, given the knowledge of the circulating dengue serotypes and genotypes in the mosquito community population and their potential to persist in the next generation of mosquitos, various genomic-based interventions may be designed and implemented, among them are the sterile insect technique, which introduces noninfected, sterile male mosquitos that will breed in natural stocks and control the egg-laying potential, thus gradually controlling the natural mosquito population. The other approach is through this Wolbachia sp. infection, a natural bacterium selectively growing in Aedes mosquitoes and would result in the eventual death of the infected mosquitos, thereby reducing the natural population. Another approach, which is done in this study is by antisense RNA, which can be designed based on the circulating variant of the virus, the templates of which are derived from the genomic surveillance DNA/RNA extracts, and are introduced or actively fed to mosquitos in the communities to block the vertical transmission of the circulating dengue viruses. Since it is not entirely possible to eliminate the mosquito population, the antisense RNA can be designed to not only block the dengue vertical transmission but also provide gene-targeting strategies that would reduce reproductive capacities, including among others egg-laying or hatching potentials. The feeding platform may involve simple technologies readily adaptable to communities such as the brown sugar solution used for feeding insects, such as butterflies, which may be enhanced with lactic acid or blood meal to promote mosquito consumption of the antisense RNA formula. The challenge; however, in this approach is assessing the long-term safety and efficacy of double-stranded RNAs and their effect in reshaping the patterns and demographic structure of the dengue virus in the natural mosquito population. The health benefits to humans though may outweigh the ecological impact of this type of mitigation.

5. Conclusions and recommendations

The genomic surveillance from year 1 of the three areas in Barangay Old Balara, an urbanized area showed an increasing trend of more than 3–4 houses per site (20–27%) that are found to be positive for dengue. A website established to report results of the genomic surveillance found utility for the online access of the Barangay health workers and provided support for their search and destroy program against the breeding ground of mosquitos. Species identification of the community-collected specimens indicated the majority to be A. aegypti and a few A. albopictus and Culex sp. The RT-PCR surveillance revealed the presence of multiple dengue serotypes in one mosquito specimen and multiple genotypes within a serotype. There was a persistent prevalence of mosquitos in Area 2 followed by Areas 4 and 1, considering that Area 2 was less forested, which contrasts with the expected pattern for A. aegypti, which tends to prefer forested areas. The low trapping yield in Area 1 reflects the anti-dengue mosquito program of the community as the community health center is in Area 1. Strong dengue positivity was found in mosquitos collected in the second to the sixth collection, which was from August to November and declined on the seventh to the eleventh collection, which was from December to February. This indicates that the vector surveillance with accompanying RT-PCR detection of the dengue virus serotypes can provide an additional layer of information that would reflect the seasonal variations of dengue infestation of the mosquito vector as well as the possible congruence of the dengue management program of the community.

The antisense RNA preparation that was developed based on the dengue amplicons obtained from the genomic surveillance was able to inhibit the dengue virus transmission in the mosquito from one generation to the other in a simulated community setting. Further studies can be done to evaluate the potential utility of a genomic surveillance-based antisense RNA platform in real-life community scenarios. While vertical transmission of dengue has been established as a known mechanism for the persistent presence of dengue in Aedes mosquito populations found in the communities, the current 4S strategy implemented locally may not be adequate to control the rising dengue cases and an active genomic-based intervention to block this vertical transmission must be done.

Acknowledgments

Special thanks to the National Research Council of the Philippines (NRCP) for the research grant of FMH (NRCP Grant # N-001). Also, thanks to the assistance and support of the Quezon City Health Department, particularly Dr. Antonieta Inumerable and Dr. Rolly Cruz; and the officials of Barangay Old Balara, Quezon City, particularly Hon. Beda Torrecampo and Dra. Karen Alcid-See. Special thanks to Dr. Cecilia Reyes, Entomologist and former Director of NRCP who engaged our team in this project, Dr., Lourdes J. Cruz, National Scientist and former President of NRCP, all the NRCP management and staff who help us through the years, especially Ms. Renia Corocoto and Mr. Caezar Arceo, and the other members of the ReM0VE Dengue Program, Dr. Grace Yu, Dr. Nelia Salazar, Dr. Judylin Solidum, Dr., Pio Javier, and Dr. Erlinda Torres.

Special thanks also to Ralph Bawalan, former MS Trop Medicine student who worked with us and now pursuing his Ph.D., and his adviser, Dr. Nelia Salazar.

Conflict of interest

The authors declare no conflict of interest.