Detection and quantification of natural Wolbachia in Aedes aegypti in Metropolitan Manila, Philippines using locally designed primers

Background The Philippines bears health and economic burden caused by high dengue cases annually. Presently, the Philippines still lack an effective and sustainable vector management. The use of Wolbachia, a maternally transmitted bacterium, that mitigate arbovirus transmission has been recommended. Cytoplasmic incompatibility and viral blocking, two characteristics that make Wolbachia suitable for vector control, depend on infection prevalence and density. There are no current Wolbachia release programs in the Philippines, and studies regarding the safety of this intervention. Here, we screened for Wolbachia in Aedes aegypti collected from Metropolitan Manila, Philippines. We designed location-specific primers for qPCR to test whether this improved Wolbachia detection in Ae. aegypti. We explored if host sex and Wolbachia strain could be potential factors affecting Wolbachia density. Methods Ae. aegypti mosquitoes (n=429) were screened for natural Wolbachia by taqman qPCR using location-specific Wolbachia surface protein primers (wspAAML) and known 16S rRNA primers. Samples positive for wspAAML (n=267) were processed for Sanger sequencing. We constructed a phylogenetic tree using IQ-TREE 2 to further characterize Wolbachia present in the Philippine Ae. aegypti. We then compared Wolbachia densities between Wolbachia groups and host sex. Statistical analyses were done using GraphPad Prism 9.0. Results Wolbachia prevalence for 16S rRNA (40%) and wspAAML (62%) markers were high. Wolbachia relative densities for 16S rRNA ranged from −3.84 to 2.71 and wspAAML from −4.02 to 1.81. Densities were higher in male than female mosquitoes. Wolbachia strains detected in Ae. aegypti clustered into supergroup B. Some 54% (123/226) of these sequences clustered under a group referred to here as “wAegML,” that belongs to the supergroup B, which had a significantly lower density than wAegB/wAlbB, and wAlbA strains. Conclusion Location-specific primers improved detection of natural Wolbachia in Ae. aegypti and allowed for relative quantification. Wolbachia density is relatively low, and differed between host sexes and Wolbachia strains. An economical way of confirming sporadic or transient Wolbachia in Ae. aegypti is necessary while considering host sex and bacterial strain.


Introduction
The Philippines continues to experience the health burden caused by high dengue cases, ranking number one in Asia with 17,630 deaths last March 2021 (Edillo et al., 2015(Edillo et al., , 2022;;Ong et al., 2022).The persistent high cases annually prompted the Philippine government to establish the National Dengue Prevention and Control Program with vector surveillance and management as one of the key targets (Dengue Prevention and Control Program | Department of Health website, n.d.).Aedes aegypti mosquito is the primary vector of Dengue in the Philippines owing to its adaptability in changing environments (Edillo et al., 2022).To mitigate infections, recommendations for vector management including the reduction of breeding sites, and improvement of water systems have been proposed.More importantly, Ong et al. (2022) highlighted the need for a more sustainable approach involving the use of the bacterium Wolbachia (Ong et al., 2022).Currently, the Philippines still lacks local studies that could provide a baseline information for assisting future mass release implementations (Ong et al., 2022).
Unlike other host species, studies on Ae. aegypti have mostly revealed the absence of natural Wolbachia (Gloria-Soria et al., 2018;Goindin et al., 2018).Novel Wolbachia transinfection in mosquitoes facilitates population suppression and replacement (Ross et al., 2020).However, the occasional presence of Wolbachia in Ae. aegypti populations without any preceding Wolbachia have been reported.These were found in the USA (Coon et al., 2016), Mexico (Kulkarni et al., 2019), Panama (Bennett et al., 2019), India (Balaji et al., 2019), Malaysia (Teo et al., 2017), Thailand (Thongsripong et al., 2018), China (Zhang et al., 2022), Taiwan (Chao and Shih, 2023), and Philippines (Carvajal et al., 2019;Regilme et al., 2021;Muharromah et al., 2023).Thus, understanding biological factors that may influence the variable presence of natural Wolbachia in Ae. aegypti is necessary.More so, finding an economical way to improve detection of Wolbachia could help initial surveillance in countries where mass release programs could be implemented.This is particularly important in low-income countries like the Philippines, gravely affected by arboviral diseases.
Natural Wolbachia hosts are characterized by consistently high Wolbachia prevalence (Inaćio da Silva et al., 2021).Culex pipiens, Culex quinquefasciatus, Ae. albopictus, Anopheles moucheti, and Anopheles demeilloni can exhibit 90% to 100% Wolbachia prevalence (Bergman and Hesson, 2021;Inaćio da Silva et al., 2021).In Anopheles gambiae and Ae.aegypti, prevalence is relatively low and can vary from 8-24% and 4.3-58%, respectively (Inaćio da Silva et al., 2021).The detection of natural Wolbachia in Ae. aegypti remains inconclusive.The reported presence of Wolbachia in the said mosquito host seems to be sporadic, transient, and low in prevalence making it difficult to quantify.Thus, more studies regarding prevalence, biological factors, and mechanisms are needed to understand the nature of natural Wolbachia sporadically found in different populations of Ae. aegypti.
Wolbachia density is host sex-dependent.Density is reported higher in adult female than male hosts, as observed in D. simulans as well as the planthoppers Laodelphax striatellus and Sogatella furcifera (Noda et al., 2001;Correa and Ballard, 2012).Wolbachia densities are also more stable in female than male insect hosts and decline with age in males (Noda et al., 2001;Unckless et al., 2009;Correa and Ballard, 2012).However, males exhibit higher Wolbachia density than females in some insects such as Diaphorina citri (Hoffmann et al., 2014).Wolbachia density and prevalence in host sex also vary with strain in Ae. albopictus (Tortosa et al., 2010;Yang et al., 2022).wAlbB density was higher in males than females, whereas wAlbA density was higher in females than males.Wolbachia wAlbA density in male Ae.albopictus also decreased with age (Tortosa et al., 2010).Further, a higher prevalence of wAlbA and wAlbB coinfection was observed in females than males, in which infection prevalence varied according to wAlb strain (Tortosa et al., 2010;Yang et al., 2022).
Currently, information on the presence of natural Wolbachia in Ae. aegypti varies and knowledge on density in relation to biological factors is limited (Teo et al., 2017;Bennett et al., 2019;Carvajal et al., 2019).Detection of Wolbachia also differs among states and cities within a country (Chen et al., 2016;Schuler et al., 2018;Zhang et al., 2022).Previous studies on prevalence have used general primers based on Wolbachia sequences from established natural hosts (Teo et al., 2017;Bennett et al., 2019;Carvajal et al., 2019;Kulkarni et al., 2019); thus primers currently used may not detect rare and low-density strains (Marcon et al., 2011;Simões et al., 2011).These limitations suggest that validation methods are necessary to perform initial surveillance targeting specific local Wolbachia populations.Lastly, Wolbachia density in naturally infected hosts is regulated by multiple factors; e.g., host genotype (Mouton et al., 2006), host sex (Tortosa et al., 2010;Mejia et al., 2022), and Wolbachia strain (Dutton and Sinkins, 2004;Hu et al., 2020).However, no study has clarified the biological factors that affect Wolbachia density in Ae. aegypti.
The present study aims to investigate if locally designed primers for qPCR could help validate the presence and density of natural Wolbachia in local mosquito populations.We also determined if density could be influenced by host sex and Wolbachia strain.We did this by utilizing locally designed Wolbachia surface protein (wsp) primers suitable for Ae.aegypti collected from Metro Manila, Philippines.We then used these primers to quantify Wolbachia density in 429 individual mosquitoes and compared against a known general primer used for conventional PCR.Next, we examined Wolbachia density between Wolbachia host sex and strains in Ae. aegypti populations.We hypothesized that the presence of sporadic, low density natural Wolbachia in Ae. aegypti can be detected and quantified by using primers designed for local mosquito populations and that Wolbachia density differs depending on Wolbachia host sex and strain.An economical way to conduct initial surveillance of Wolbachia in Ae. aegypti is necessary especially in low-income countries like the Philippines, where an effective vector control strategy is needed.

Mosquito sample collection
We used DNA samples extracted from Ae. aegypti adult mosquitoes collected in the National Capital Region (Manila) of the Philippines; these samples were previously used for Wolbachia detection via conventional PCR (Carvajal et al., 2019) and ddRAD-Seq (Muharromah et al., 2023).Wolbachia was detected in 11.9% (80/672) of these samples by PCR (Carvajal et al., 2019) and sequence reads of 26,299 (wAlbA) and 43,778 (wAlbB) were mapped across the entire Wolbachia genome out of 146,239,637 filtered reads obtained by ddRAD-Seq (Muharromah et al., 2023).Thus, the samples were considered suitable for validating primer incompatibility.Each individual mosquito was previously screened and processed for DNA extraction as described by Carvajal et al. (2019).The total DNA of individual mosquitoes was extracted using a Blood and Tissue DNEasy Kit (Qiagen, Hilden, Germany) according to the manufacturer's protocol with slight modifications (Crane, 2013).All samples were stored as DNA at −80°C for long term preservation.Of 672 individual mosquitoes used by Carvajal et al. (2019), we selected 429 samples based on sufficient volume and DNA concentration for downstream assays.

Wolbachia detection via conventional PCR
Data on conventional PCR results were obtained from the study of Carvajal et al. (2019) (Supplementary Data) and used as a baseline reference in the present study in relation to Wolbachia prevalence and density detected with the newly designed primers.Briefly, Carvajal et al. (2019) used two known markers targeting the 16S rRNA gene, which has a slow evolutionary rate, and another marker targeting the highly variable Wolbachia surface protein (wsp), which is suitable for strain identification (O'Neill et al., 1992;Zhou et al., 1998).The sequences of 16S rRNA and wsp Wolbachia-specific primers were as follows: Wspecf (5′-GAA GAT AAT GAC GGT ACT CAC-3′) and Wspecr (5′-AGC TTC GAG TGA AAC CAA TTC-3′) (O'Neill et al., 1992); wsp 81F (5′-TGG TCC AAT AAG TGA TGA AGA AAC-3′) and wsp 691R (5′-AAA AAT TAA ACG CTA CTC CA-3′) (Zhou et al., 1998).PCR thermocycling conditions were conducted according to the published protocol (Carvajal et al., 2019).

Primer design for wsp based on local wsp sequences
Most wsp primers designed for Wolbachia detection are strainspecific (Zhou et al., 1998;Dutton and Sinkins, 2004).Thus, we developed new primers specific for our local samples to consider for primer incompatibility and to quantify Wolbachia density.First, we obtained 118 wsp sequences from the Ae.aegypti samples of Carvajal et al. (2019) (GenBank popset 1712729902).Next, a multiple sequence alignment was performed using MUSCLE, and the results were visualized in Codon Code Aligner version 1.2.4 (https:// www.codoncode.com/aligner/).The consensus sequence produced from the alignment was then inputted into Primer-BLAST (Ye et al., 2012) to design wsp primers targeting the Ae.aegypti samples.Primer-BLAST generated five primer pairs (Supplementary Table 1), which were first validated via gradient conventional PCR using a known Cx. quinquefasciatus positive sample.Among the primer pairs, primers wspAAML 01 and wspAAML 05 were selected for further optimization because they exhibited the correct band size of target markers without nonspecific binding in the sample (Supplementary Figure 1).To select the most suitable wspAAML primer pair for downstream analysis, we determined the optimized annealing temperature and primer concentration for both pairs (Supplementary Figure 2).We selected wspAAML 05, given that its PCR efficiency (Supplementary Figure 3) fell within the standard MIQE guideline of ≥90% (Bustin et al., 2009).

Natural Wolbachia infection validation using TaqMan qPCR and strain identification via sequencing
To quantify Wolbachia density, Taqman qPCR targeting both 16S rRNA and wspAAML was conducted.Wolbachia quantification was performed using a well-established primer set targeting the 16S rRNA marker (16SF 5′-AGT GAA GA A GGC CTT TGG G-3′; 16SR 5′-CAC GGA GTT AGC CAG GAC TTC-3′) but with a modification of the fluorescent dye of the probe (Fraser et al., 2020).Instead of LC640, we used TET as the reporter dye and BHQ1 as its quencher (5′TET-CTG TGA GTA CCG TCA TTA TCT TCC TCA CT-BHQ13′).Wolbachia was also confirmed using the newly designed Wolbachia surface protein (wsp) primers (wspAAML F 5′-AGC ATC TTT TAT GGC TGGT GG-3′; wspAAML R 5′-AAT GCT GCC ACA CTG TTT GC-3′; wspAAML probe 5′FAM-ACG ACG TTG GTG GTG CAA CAT TTG C-TAMRA3′) with the Ae.aegypti ribosomal protein S17 (RPS17) gene as a reference gene (17SF 5′-TCC GTG GTA TCT CCA TCA AGC T-3′; 17SR 5′-CAC TTC CGG CAC GTA GTT GTC-3′; 17S probe 5′HEX-CAG GAG GAG GAA CGT GAG CGC AG-BHQ13′) (Frentiu et al., 2010).In total, 429 individual mosquitoes were screened for the presence of Wolbachia using qPCR with a cut-off Cq value of 35.The cut-off value was set from a qPCR experiment of Cx. quinquefasciatus samples representing true natural Wolbachia infection and three replicates of the no-template control for each target gene; negative detections of no template controls were confirmed using gel electrophoresis showing no bands, hence no true amplification (Supplemental Figure 4).
All singleplex PCR reactions were performed in a final volume of 10 µl containing 5 µl of 2x iTaq Universal Probes Supermix (Bio Rad) with 0.3 µM RPS17 primers, 0.2 µM 16S rRNA primers, or 0.5 µM wsp primers and 0.2 µM, 0.15 µM, or 0.3 µM of their corresponding TaqMan probes, respectively, with nuclease-free water added to reach the final volume.The following thermal profile was used for both RPS17 and 16S rRNA with a CFX96 touch deep well real-time PCR detection system (Bio-Rad Tokyo, Japan): an initial polymerase activation at 95°C for 30 s followed by 40 cycles of denaturation at 95°C for 5 s, and combined annealing/ extension at 60°C for 10 s.The wsp thermal profile included an initial polymerase activation at 95°C for 2 min followed by 40 cycles of denaturation at 95°C for 5 s, and combined annealing/extension at 58.8°C for 30 s.Each PCR amplification included a Wolbachiainfected Cx. quinquefasciatus positive control and a no-template control.Wolbachia density was derived from the Cq values expressed as the relative density of wspAAML normalized to single-copy RPS17 (Livak and Schmittgen, 2001).This was used to compare density between host sex and Wolbachia strains.Relative densities reported in this study were log-transformed (log10).Following detection, qPCR-confirmed wspAAML-positive products were cleaned using a mixture of alkaline phosphatase (TaKaRa) and exonuclease I (TaKaRa).The cleaned samples were subjected to Sanger sequencing for strain identification.

Wolbachia phylogeny
Wolbachia phylogeny was inferred using the maximumlikelihood criterion.For this analysis, we used Ae.aegypti samples in which the wsp gene was detected via qPCR (n = 226).We also obtained additional wsp sequences from other host species, e.g., Aedes spp., Anopheles spp., Culex spp., and others indicated as reference sequences (n = 511), from NCBI GenBank (Table 1).In total, 737 wsp sequences were aligned using MUSCLE in CodonCode Aligner version 1.2.4 (https://www.codoncode.com/aligner/) and then trimmed to a final length of 103 nucleotide bases.The sequences containing 103 nucleotide bases correspond to the homologous region across all sequences.Using DNASp version 6.12.03 (Rozas et al., 2017), we obtained 102 haplotypes, and we subjected the representative sequences of each haplotype to phylogenetic analysis.Tree reconstruction was conducted using only wsp given the high evolutionary rate of the gene; i.e., its suitability for strain identification (Zhou et al., 1998;Ren et al., 2020).IQ-TREE 2 (http://iqtree.org)(Minh et al., 2020) was used where the appropriate substitution model was first identified through ModelFinder implemented as a function of the software (Kalyaanamoorthy et al., 2017), from which TPM2+G4 was selected as the best-fit model.We set the ultrafast bootstrap approximation (UFBoot) in IQ-TREE to 1,000 iterations, the minimum correlation coefficient to 0.99, and the other parameters to their default settings (Hoang et al., 2018).For visualization and annotation, we used iTOL (https://itol.embl.de/)(Letunic and Bork, 2021).

Statistical analysis
Mann Whitney tests were used to determine statistical differences between Wolbachia densities of male and female mosquitoes as well as those between the densities of samples detected as positive or negative using conventional PCR.To compare densities between Wolbachia strains, Dunn's multiple comparison test was performed.Statistical calculations were conducted in GraphPad Prism version 9.2.0 for Windows (www.graphpad.com),and p -values of ≤0.01 or ≤0.0001 were considered statistically significant.

Validation of natural Wolbachia infection in Ae. aegypti
Screening of Wolbachia in Ae. aegypti using 16S rRNA and wsp qPCR revealed an overall prevalence of 40% (172/429) and 62% (267/429) in the mosquito population, respectively (Table 2).Wolbachia density expressed in logarithmic scale refers to the relative abundance of the target gene normalized to the reference gene, RPS17 (Livak and Schmittgen, 2001).Thus, the median relative Wolbachia density was −1.99 and −2.09 for 16S rRNA and wspAAML qPCR assays, respectively.The relative Wolbachia densities of 16S rRNA-and wspAAML-positive samples ranged from −3.84 to 2.17 and −4.02 to 1.81, respectively.
Comparing our qPCR results with the conventional PCR results of Carvajal et al. (2019), we found that 91% (40/44) of the mosquitoes positive for 16S rRNA in conventional PCR showed the same result in our qPCR, whereas only 9% equivalent to 4 out of 44 of the mosquito samples previously detected as positive by standard PCR were detected as negative via qPCR.Of the 385 samples confirmed as negative for 16S rRNA via conventional PCR, 34% (132/385) were positive according to qPCR, whereas 66% (253/385) were consistent with negative conventional PCR detection.Regarding wsp, 100% (57/ 57) of mosquito samples that were positive according to conventional PCR were also positive in our qPCR.Conventional PCR wsp-negative samples exhibited an infection prevalence of 57% positive (213/372) and 43% negative (159/372) detection according to qPCR.
In an attempt to explain the contrasting negative conventional PCR and positive qPCR Wolbachia detection results, we compared the relative Wolbachia densities expressed in logarithmic scales, between samples found to be either Wolbachia-negative or positive via conventional PCR by Carvajal et al. (2019) (Figure 1).We found an approximately 30-fold higher relative Wolbachia density in 16S rRNA (median = −0.8250)and wspAAML (median = −0.8550)Wolbachia-positive mosquitoes compared with the conventional PCR-negative samples for 16S rRNA (p < 0.001; median = −2.325)and wsp (p < 0.001; median = −2.250).Finally, it is worth mentioning that in qPCR alone, only 21% (91/429) of samples were positive for both 16S rRNA and wspAAML markers.

Natural Wolbachia density differs according to host sex
For both 16S rRNA and wspAAML qPCR assays, Wolbachia densities were ≥10-fold higher in male than female Ae.aegypti (p ≤ 0.01 for both; Figure 2).In the 16S rRNA marker, male mosquitoes (n = 69) exhibited relative Wolbachia densities between −3.57 and 2.17 with a median value of −1.67, whereas females (n = 103) exhibited a relative Wolbachia density range from −3.84 to 1.82 and

Brugia pahangi AY527207
The table provides the different reference sequences used for phylogenetic analysis.Each sequence represents a specific host and supergroup.

Phylogenetic analysis of Wolbachia strains in Ae. aegypti
For phylogenetic analysis, we used 226 wspAAML sequences obtained from 267 wsp-positive Ae. aegypti samples collected from Metro Manila (AAML).We excluded 41 samples due to low sequencing quality and an inability to repeat sequencing owing to an inadequate volume of DNA.According to the maximumlikelihood phylogenetic tree, 83% (187/226) of wspAAML sequences in the AAML samples clustered into supergroup B, whereas only 15% (34/226) of wsp sequences clustered into supergroup A (Figure 3).The bootstrap values (≥75%; indicated on the external tree branches) supported the divergence of three clusters between supergroups A and B.
Under supergroup A, a clear cluster similar to the strain wAlbA was observed.Two wsp sequences of Ae. aegypti did not exhibit a clear delineation with any Wolbachia reference sequence.Numerous weak bootstrap support values (≤74%) were observed in deep nodes within supergroup A, indicating high genetic diversity.Under supergroup B, the wspAAML sequences of Ae. aegypti samples were nested within Wolbachia cluster wAegB/wAlbB together with other wsp sequences derived from Ae. albopictus or Culex spp.We found 12 wspAAML sequences of Ae. aegypti samples that fall under the wPip strain.However, another cluster was solely composed of wspAAML sequences found in the AAML samples (n = 123).For clarity, we hereafter refer to this AAML group (shaded in red in Figure 3) as the "wAegML" cluster.The two Wolbachia strains under supergroup B, i.e., wAegB/wAlbB and wAegML, share the same branch as wPip, which was consistent with the grouping previously established by Zhou et al (Zhou et al., 1998).The 12 sequences clustering with wPip share similarity with wAegML group and  The table shows a comparison of Wolbachia-positive samples detected using conventional PCR and qPCR.
wAegB/wAlbB strain.We also noted that five wsp sequences from the Ae.aegypti samples formed a distinct cluster that did not fall under any of the supergroups considered; thus, these samples possibly belong to supergroups other than A and B. Lastly, Wolbachia density of the wAegML group was less than either wAegB/wAlbB or wAlbA in Ae. aegypti (Dunn's test, p <0.0001; Figure 4).However, Wolbachia density did not differ between wAegB/wAlbB and wAlbA strains in Ae. aegypti (Dunn's test, p >0.0001).

Discussion
The presence of natural Wolbachia in Ae. aegypti has been investigated in different countries.Based on these reports, Ae. aegypti is either absent or when present, exists sporadically and in low prevalence (Coon et al., 2016;Teo et al., 2017;Gloria-Soria et al., 2018;Goindin et al., 2018;Thongsripong et al., 2018;Balaji et al., 2019;Bennett et al., 2019;Carvajal et al., 2019;Zhang et al., 2022;Muharromah et al., 2023).Particularly in the Philippines, Wolbachia has been detected in cities within Metropolitan Manila finding a relatively low prevalence for both 16S rRNA (13.2%) and wsp markers (16.8%) by conventional PCR (Carvajal et al., 2019).More recently, natural Wolbachia detected in Ae. aegypti from the same mosquito population has been validated by ddRADsequencing.Sequence reads of 26,299 (wAlbA) and 43,778 (wAlbB) were mapped across the entire Wolbachia genome out of 146,239,637 filtered reads obtained (Muharromah et al., 2023).There are currently no Wolbachia mass release programs being implemented in the Philippines but its potential as a vector control method has been proposed (Ong et al., 2022).Hence, an economical way of conducting an initial surveillance to characterize not just the presence but also the density of natural Wolbachia in Ae. aegypti is warranted.In the present study, we used locally designed primers (wspAAML) to validate the presence of natural Wolbachia in Ae. aegypti that resulted to a higher prevalence of Wolbachia infection in wsp compared to the conventional PCR method used by Carvajal et al.We also found a high detection rate of Wolbachia, although to a lesser extent, using an established 16S rRNA primers.We found that relative Wolbachia density varied between host sex and Wolbachia phylogenetic groups in natural Ae.aegypti populations.The Wolbachia strains found present in Ae. aegypti were closely related to strains found in Ae. albopictus.We identified one cluster in the phylogenetic tree referred to as wAegML that was present in lower density but in higher prevalence in Ae. aegypti.
Wolbachia is ubiquitous in numerous host species (Noda et al., 2001;Kyei-Poku et al., 2005;Ros et al., 2009;Hughes et al., 2011) but its presence in mosquitoes vary.Presence of natural Wolbachia is usually high among Culex spp.and Ae.albopictus which are hosts regarded as naturally infected.On the other hand, Anopheles gambiae (8%-24%) and Ae.aegypti (4.3%-58%) exhibit variable prevalence results that differs according to geographical location (Inaćio da Silva et al., 2021).In the present study, Ae. aegypti collected from Metro Manila, Philippines demonstrated natural Wolbachia prevalence of 40.0% and 62.2% when targeting 16S rRNA and wsp, respectively.This finding is consistent with other studies that detected natural Wolbachia in Ae. aegypti (Coon et al., 2016;Teo et al., 2017;Thongsripong et al., 2018;Balaji et al., 2019;Bennett et al., 2019;Carvajal et al., 2019;Kulkarni et al., 2019;Muharromah et al., 2023).Given the high prevalence rate of natural Wolbachia detected here, we suggest that the use of location-specific primers for qPCR increased the sensitivity of our detection method.
In a previous study, our laboratory performed conventional PCR-based detection of Wolbachia infection in Ae. aegypti collected from the current study location, finding a relatively low prevalence considering both 16S rRNA (13.2%) and wsp markers (16.8%).Contrastingly, a study conducted by Gloria-Soria et al. (2018) revealed the absence of Wolbachia in 117 Ae. aegypti mosquitoes collected from Cebu province, Philippines (Gloria-Soria et al., 2018).In both studies, conventional PCR was used.Although the primers used can amplify multiple Wolbachia strains, they were not designed using natural Wolbachia populations from Ae. aegypti mosquitoes in the local regions.Additionally, the difference in the prevalence of natural Wolbachia in Metropolitan Manila (Carvajal et al., 2019;Muharromah et al., 2023) and Cebu, Philippines (Gloria-Soria et al., 2018) further supports the fact that the presence of natural Wolbachia is variable and sporadic.The same situation occurs in China where infection prevalence between prefectures differed from 0 to 41.7% (Zhang et al., 2022).Different Wolbachia primers used for PCR assays vary in terms of efficiency and coverage.The performance of 13 Wolbachia primer pairs was previously assessed using samples from a wide range of hosts representing supergroups A-F.The results varied, even among primers targeting the same gene, and only two primer sets yielded identical results, with other primers resulting in incorrectly sized amplifications (Simões et al., 2011).
In the present study, to address the potential issue of primer incompatibility, we designed location-specific primers for wsp based on published sequences of Wolbachia populations in the same regions (Carvajal et al., 2019).In order to avoid bias, we included 16S rRNA marker which we did not design and has already been established (Fraser et al., 2020).Using either marker demonstrated a high detection rate with wspAAML being 22% higher than 16S rRNA.This suggests that the qPCR method led to an improved natural Wolbachia detection rate in Ae. aegypti owing to primer compatibility.
Notably, the qPCR method requires the use of primers that yield short amplicons (150 bp), which may have contributed to an increase in sensitivity.Additionally, qPCR has a lower limit of detection and higher sensitivity relative to conventional PCR, which could also explain the higher detection rate in the current study (Mee et al., 2015;Xia et al., 2018).When we compared Wolbachia density between samples found to be Wolbachia-negative or positive using conventional PCR, we found a 30-fold higher median Wolbachia density in the Wolbachia-positive samples by qPCR.We also found low Wolbachia density.Therefore, it is likely that both the newly designed primers and qPCR helped increase the natural Wolbachia detection rate relative to the detection performance of conventional PCR.These results suggest that location-specific primers may help validate the presence and determine Wolbachia density in Ae. aegypti samples with transient, sporadic, and/or low density natural Wolbachia infection.
The use of Ae. aegypti mosquitoes previously subjected to Wolbachia detection (Carvajal et al., 2019), allowed us to define potential reasons (e.g., primer incompatibility) for the sporadic detection of natural Wolbachia in Ae. aegypti.However, the possibility of contamination of Wolbachia from other mosquito host species in the larval stage cannot be completely avoided and therefore should be carefully considered (Ross et al., 2020).It is recommended that a comprehensive approach including imaging-based technique (FISH, IFA), demonstration of maternal transmission, reproductive manipulation e.g., CI, and antibiotic treatment of Wolbachia can be used to confirm natural infection in Ae. aegypti.Nevertheless, an economical way to improve detection for initial surveillance is also needed considering the sporadic and low-density presence of Wolbachia in Philippine Ae. aegypti.
Meanwhile, the present study adds to the existing evidence that Wolbachia (wsp) sequences found in Ae. aegypti belong to either supergroup A or B. Most wsp sequences found in Ae. aegypti collected from Metro Manila, Philippines were categorized under supergroup B consistent with the previous studies that utilized known Wolbachia primers and ddRAD-Seq, respectively (Carvajal et al., 2019;Muharromah et al., 2023).Both supergroups are widespread among arthropods and belong to a single monophyletic lineage (Gerth et al., 2014;Zug and Hammerstein, 2015).Supergroup B is likely to be the dominant supergroup found in naturally infected Ae. aegypti, as reported in most previous studies (Coon et al., 2016;Balaji et al., 2019;Carvajal et al., 2019;Kulkarni et al., 2019).Wolbachia strains under supergroup B usually reside in their hosts at high density (Hu et al., 2020;Walker et al., 2021) and exhibit resilience to cyclical heat stress, allowing them to persist in host populations (Ross et al., 2017).In our study, Wolbachia density is found to be relatively low.Balaji et al. (2019) reported 1.01 and 1.76 wsp/Rps17 of Wolbachia in male and female, respectively whereas we found an overall range of -4.02 to 1.81 regardless of gender.Natural Wolbachia density in Ae. aegypti differs throughout the developmental stage where significantly low density commonly manifests from the adult stage (Balaji et al., 2019).Our study only used random fieldcollected adult samples without considering age of adult mosquitoes that could account for the difference.
Interestingly, we detected three clusters representing the Wolbachia strains wAlbA, wAegB/wAlbB, and wAegML.Due to the inability to conduct further experiments to validate strain, we refer to another cluster as merely wAegML group.The bacterial density of wAegML was significantly lower than that of wAlbA, wAegB/wAlbB, and 62.8% of the wsp sequences in our Ae.aegypti samples belong to the wAegML.It is important to recognize that our study only used wsp for phylogenetic analysis which somehow limit our capacity to establish relatedness.Therefore, it is important that future studies utilize other markers and incorporate MLST data for a more accurate strain differentiation (Wang et al., 2020).Further experiments are needed to provide conclusive evidence on what impact sporadic and low-density natural Wolbachia in Ae. aegypti have.
Lastly, previous studies have revealed sex-specific Wolbachia density differences in natural populations of planthoppers S. furcifera and L. striatellus (Noda et al., 2001), as well as D. citri (Hoffmann et al., 1986), Drosophila spp (Correa andBallard, 2012), Cx. Pipiens (Echaubard et al., 2010), Ae. albopictus (Tortosa et al., 2010;Calvitti et al., 2015), and Ae.aegypti (Mejia et al., 2022).We found that natural Wolbachia density in Ae. aegypti males was higher than in females, which is consistent with observations in D. citri (Hoffmann et al., 1986) and Ae.albopictus (Tortosa et al., 2010).Such sex-specific variation was only previously observed in Ae. albopictus mosquitoes carrying the wAlbB strain (Tortosa et al., 2010).A recent study on Wolbachia density in Ae. aegypti also reported the consistent densities in females throughout their lifetime whereas males demonstrate relatively higher and variable Wolbachia densities, in testes (Mejia et al., 2022).Although the biology underlying sex-specific differences in Wolbachia density is unknown, biological differences between the host sexes could explain our finding.For instance, female mosquitoes tend to have an expanded microbial composition relative to that of males, resulting in more bacterial competition (Zouache et al., 2011).Female mosquitoes are hematophagous, and the composition of bacterial microbiota in mosquitoes largely depends on nutrient intake (Gaio et al., 2011;Minard et al., 2013).Thus, the digestion process of female mosquitoes may act as a barrier to the survival of some symbionts, including Wolbachia.
Other studies have found that females exhibit higher Wolbachia densities than males in S. furcifera, L. striatellus (Noda et al., 2001), Drosophila spp (Correa andBallard, 2012), Cx. pipiens (Echaubard et al., 2010), and Ae.albopictus (Tortosa et al., 2010;Calvitti et al., 2015).Thus, further investigation on gender-specific effects considering other coexisting factors is warranted.Our study only explored differences of relative Wolbachia density between male and female adult mosquitoes.It is important to further characterize how Wolbachia density is affected by host sex in different stages of the mosquito life cycle.Determining an absolute Wolbachia density rather than a relative one will provide more accurate information.Additionally, supplementing this with microscopic evidence will demonstrate the changes in Wolbachia density occurring between male and female mosquitoes during the insects' life cycle.

Conclusion
Host species in which Wolbachia is naturally found usually exhibit stable high prevalence, high density, and heritable infection.Nevertheless, this is not the case for Ae.aegypti.Our study proposes that presence of natural Wolbachia in Ae. aegypti does not mean absolute presence, rather natural Wolbachia in this host is sporadic, and low-density.In the present study, we focused on investigating the factors that affect such occurrence.The use of Ae. aegypti samples previously subjected to Wolbachia detection by conventional PCR and ddRAD-Seq (Carvajal et al., 2019;Muharromah et al., 2023) allowed us to explore the biological (Wolbachia strain and host sex) and methodological (method of detection, primer incompatibility) factors affecting Wolbachia prevalence rate and density.This potentially explains the varying presence of natural Wolbachia in this host.Our findings suggest that (i) location-specific primers can improve Wolbachia density detection; (ii) relative Wolbachia densities are influenced by host sex and bacterial strain; (iii) majority of Wolbachia sequences clustered into a group referred to here wAegML that was present at a low density in Ae. aegypti.Overall, designing location-specific primers is economical for initial validation considering limited resources in other countries.These factors must be accounted for when conducting initial surveillance in places where no mass release programs have been conducted (e.g., Philippines).Moving forward, a comprehensive detection of natural Wolbachia infection in Ae. aegypti using multiple methods is still warranted.
FIGURE 1 Wolbachia density of qPCR-positive Ae. aegypti grouped according to conventional PCR results.Individual mosquitoes detected as either positive for 16S rRNA (n =172) or wspAAML (n = 267) via qPCR were grouped based on Wolbachia detection results according to conventional PCR (cPCR).Relative Wolbachia density is expressed as the ratio of the target gene to single-copy RPS17 in logarithmic scale.(A) Using 16S rRNA primers, cPCR positive (n = 40) and negative (n = 132) samples had a median relative Wolbachia density of −0.8250 and −2.325, respectively.(B) Using wspAAML primers, cPCR positive (n = 54) and negative (n = 213) samples had a median relative Wolbachia density of −0.8550 and −2.250, respectively.**** indicates a significant difference between cPCR-positive and cPCR-negative Ae. aegypti at p ≤ 0.0001..
FIGURE 2Presence of natural Wolbachia between male and female Ae.aegypti.Mosquitoes that were Wolbachia-positive according to TaqMan qPCR were classified as either male (16S rRNA = 69, wspAAML = 109) or female (16S rRNA = 103, wspAAML = 158).Relative Wolbachia density is shown as the ratio of the target gene to RPS17 in logarithmic scale.The 95% confidence interval of the median is indicated by the blue and violet lines.Relative densities are shown for (A) 16S rRNA and (B) wspAAML markers.In both (A, B), Wolbachia density differed significantly according to host sex.**p ≤ 0.01.
FIGURE 3 Phylogenetic analysis of Wolbachia according to wsp.Maximum-likelihood tree showing wspAAML sequences from Ae. aegypti samples collected in Metro Manila, Philippines (AAML in black bold text) and other wsp sequences obtained from other mosquito hosts (data from NCBI Blast; shown in black italicized text).Reference sequences are shown in red with an indication of the corresponding Wolbachia strains.Numbers on branches reflect the bootstrap support estimated with 1000 replications.The gray triangles are collapsed clades, each representing a node class.Ninety-eight percent of the wspAAML sequences clustered into either supergroup A or B. The tree shows two Wolbachia strains, namely wAlbA (violet) under supergroup A, wAlbB/wAegB (blue), and a group AAML under supergroup B (red).Scale bar indicates a phylogenetic distance of 0.1 nucleotide substitutions per site.

FIGURE 4
FIGURE 4 Wolbachia density in Ae. aegypti according to three identified Wolbachia clusters.Relative Wolbachia density of strains wAlbB/ wAegB, wAegML, and wAlbA in Ae. aegypti.Relative density was considered the ratio of the target gene to RPS17.Values are shown on a logarithmic scale.Median Wolbachia relative density is indicated by a dark grey line.**** indicates p ≤0.001.ns indicates no significance between two groups.

TABLE 1
Reference sequences for Wolbachia supergroups.

TABLE 2
Infection prevalence of natural Wolbachia in Ae. aegypti based on conventional PCR (cPCR) and qPCR.