Systematic Review of Wolbachia Symbiont Detection in Mosquitoes: An Entangled Topic about Methodological Power and True Symbiosis

Wolbachia is an endosymbiotic bacterium that naturally infects several arthropods and nematode species. Wolbachia gained particular attention due to its impact on their host fitness and the capacity of specific Wolbachia strains in reducing pathogen vector and agricultural pest populations and pathogens transmission. Despite the success of mosquito/pathogen control programs using Wolbachia-infected mosquito release, little is known about the abundance and distribution of Wolbachia in most mosquito species, a crucial knowledge for planning and deployment of mosquito control programs and that can further improve our basic biology understanding of Wolbachia and host relationships. In this systematic review, Wolbachia was detected in only 30% of the mosquito species investigated. Fourteen percent of the species were considered positive by some studies and negative by others in different geographical regions, suggesting a variable infection rate and/or limitations of the Wolbachia detection methods employed. Eighty-three percent of the studies screened Wolbachia with only one technique. Our findings highlight that the assessment of Wolbachia using a single approach limited the inference of true Wolbachia infection in most of the studied species and that researchers should carefully choose complementary methodologies and consider different Wolbachia-mosquito population dynamics that may be a source of bias to ascertain the correct infectious status of the host species.


Introduction
Wolbachia pipientis is an endosymbiotic bacteria from the Rickettsiales order, identified for the first time in 1923 in Culex pipiens ovaries [1]. Since then, Wolbachia has been widely studied from basic to applied biology. It is estimated that strains of the genus Wolbachia are naturally present in 66% of all insect species, showing a wide array of ecological interactions, varying from parasitism, commensalism and mutualism, with their eukaryotic host cells [2][3][4][5]. In past years, several studies have been published focusing on Wolbachia's ability to manipulate their host reproductive system due to the applicability of different derived phenotypes in new strategies to control arthropod species populations [6,7]. Different Wolbachia strains can generate parthenogenesis, feminization and cytoplasmic incompatibility (CI) on their hosts [8,9]. Such phenotypes increase the frequency of host infected individuals consequently enhancing the Wolbachia transmission to their progeny [10]. Moreover, Wolbachia can also be transmitted and infect new host species through horizontal transfer, This systematic review evaluated the presence of Wolbachia in culicids, analyzed the methods employed to detect the bacterium and provide guidelines and perspectives for researchers in this area.

Articles
The 59 selected articles were published between the years of 2000 and 2020, with 2018 being the year with the highest number of publications (12). The average number of mosquito species evaluated for each article was 9.06, with 16.22 as the standard deviation. The largest number of species analyzed by a single article was 87, however, 37% of the studies evaluated only one species [49].

Methods Used to Detect Wolbachia in Culicids
Wolbachia has been screened using different methodologies that can be subdivided into two larger groups based on the molecule/cellular structure investigated: Amplificationbased strategies: PCR, real-time PCR (qPCR), restriction fragment length polymorphism (RFLP), multilocus sequence typing (MLST), metabarcoding and Loop Mediated Isothermal Amplification (LAMP); and cell/structure visualization strategies: electron transmission microscopy (MET) and cell culture. Amplicon-based strategies have used several different target genes for Wolbachia detection, such as wsp, fstz, 16s rRNA, orf7 (from an integrated bacteriophage WO found in the Wolbachia genome), Tr1, pk1, ank2, groE, 18s rRNA, GP15, ISWpi1 transposable element, besides specific targets for the different strains of Wolbachia. While cell/structure visualization strategies relied on specific Wolbachia protein staining and/or the staining of the entire Wolbachia cell.
Of the total articles selected for this systematic review, 83% of them employed only one technique to detect Wolbachia in mosquitoes (Supplementary Material). The conventional PCR technique for more than one Wolbachia target gene was used in 51% of those with 4.16 targets per article on average, while 17 target genes was the highest number of target genes that was used in only a single article [49]. Of the articles that used only conventional PCR for amplification (20) of a single Wolbachia target gene, most of them chose the wsp gene, in 45% of the studies. Twelve studies used more than one PCR target genes as complementary methodologies and only two articles (16%) used more than one technique (Amplicon-based and Cell/Structure visualization) to detect Wolbachia. Wolbachia cell culture, metabarcoding, LAMP, and MET, were employed in only one article each [45,[50][51][52] (Supplementary Material). It is important to note such highly heterogeneous and non-standardized use of different molecular biology techniques to detect Wolbachia in mosquitoes. Moreover, many of the methodologies employed alone are not able to differentiate between a true Wolbachia infection and the detection of Wolbachia molecule traces irrespective of the infection status (see Discussion section). Therefore, from now on we described the results collected in this review using a general term "Wolbachia detection" and the results should be taken cautiously regarding the Wolbachia infection status unless we stated that multiple methodologies have been employed corroborating a true Wolbachia infection.

Distribution of Wolbachia in Culicidae Species
Two hundred and seventeen Culicidae species belonging to 22 different genera were screened for Wolbachia so far, which corresponds to only 6% of all mosquitoes recorded [53]. Anopheles, Aedes, and Culex have the largest number of screened species, which correspond to 75% of all recovered species, with species of the Anopheles genus being the most abundant (76 species) ( Figure 1B,C). Some genera such as Ficalbia, Malaya, Haemagogus, Hodgesia, and Limatus were represented by only one species but the species investigated from the last three genera were classified taxonomically only at genus level. Ae. albopictus and Ae. aegypti species were screened by Wolbachia in several studies, 26 and 21 different articles, respectively. There is a strong linear correlation between the number of mosquito species per genera and the number of species per genera in-  Figure 1A). An interactive map with all species screened for Wolbachia detection so far is available on the microreact platform ( https://microreact.org/project/rxDRQdWjzg86eXCTF8oNn4). gogus, Hodgesia, and Limatus were represented by only one species but the species investigated from the last three genera were classified taxonomically only at genus level. Ae. albopictus and Ae. aegypti species were screened by Wolbachia in several studies, 26 and 21 different articles, respectively. There is a strong linear correlation between the number of mosquito species per genera and the number of species per genera investigated for Wolbachia infection (R = 0.883) ( Figure 1A). An interactive map with all species screened for Wolbachia detection so far is available on the microreact platform (https://microreact.org/project/rxDRQdWjzg86eXCTF8oNn4). Aedes and Culex genera were the most studied ones, the other genera were grouped in the "Others" category. The sum of separated values does not totalize 217 species, since Ur. Spp. is considered positive by one study and negative by other, but cannot be classified in "Both" categories, once we cannot guarantee if it corresponds to the same species. Aedes and Culex genera were the most studied ones, the other genera were grouped in the "Others" category. The sum of separated values does not totalize 217 species, since Ur. Spp. is considered positive by one study and negative by other, but cannot be classified in "Both" categories, once we cannot guarantee if it corresponds to the same species.
Seventeen out of 22 genera had at least one species positive for Wolbachia, whereas the remaining five (Culiseta, Haemagogus, Lutzia, Heizmannia and Mimomyia) were negative. All species investigated so far from the Coquillettidia, Limatus, and Psorophora genera were Wolbachia-positive ( Figure 1C).
Of the total species screened for Wolbachia, 122 were considered negative, and 66 were positive, which corresponds to approximately 56% and 30% of the total, respectively ( Figure 1C). In addition, a different group of species (30 species-14%) were both considered positive and negative by different studies ( Figure 1C, Supplementary Material). However, due to the high variability of the methods used in these studies (mainly amplicon-based), the infection status of most species needs further assessment.

Wolbachia Diversity and Infection Rate
Forty-four distinct strains of Wolbachia were characterized; 36% of those belong to supergroup A and 45% belong to supergroup B. The remaining 19% were not genotyped at the supergroup level ( Figure 2). The most frequent strain found was wPip detected in 16 different mosquito species, followed by the wCon strain, detected in 9 species. In addition, some studies recorded the presence of Wolbachia from other supergroups (in addition to A and B), such as C in Ae. aegypti, D in An. baimai, and F in An. minimus and An. maculatus. Seventeen out of 22 genera had at least one species positive for Wolbachia, whereas the remaining five (Culiseta, Haemagogus, Lutzia, Heizmannia and Mimomyia) were negative. All species investigated so far from the Coquillettidia, Limatus, and Psorophora genera were Wolbachia-positive ( Figure 1C).
Of the total species screened for Wolbachia, 122 were considered negative, and 66 were positive, which corresponds to approximately 56% and 30% of the total, respectively ( Figure 1C). In addition, a different group of species (30 species-14%) were both considered positive and negative by different studies ( Figure 1C, Supplementary Material). However, due to the high variability of the methods used in these studies (mainly amplicon-based), the infection status of most species needs further assessment.

Wolbachia Diversity and Infection Rate
Forty-four distinct strains of Wolbachia were characterized; 36% of those belong to supergroup A and 45% belong to supergroup B. The remaining 19% were not genotyped at the supergroup level ( Figure 2). The most frequent strain found was wPip detected in 16 different mosquito species, followed by the wCon strain, detected in 9 species. In addition, some studies recorded the presence of Wolbachia from other supergroups (in addition to A and B), such as C in Ae. aegypti, D in An. baimai, and F in An. minimus and An. maculatus. The infection rate, the percentage of Wolbachia-positive mosquitoes, was calculated for 32 out of 65 species mostly using amplification-based strategies (Supplementary Material). A large variation was found such as Ae. albopictus and Ma. uniformis, which showed infection rates varying from 15% to 100% (Table 1). The infection rate, the percentage of Wolbachia-positive mosquitoes, was calculated for 32 out of 65 species mostly using amplification-based strategies (Supplementary Material). A large variation was found such as Ae. albopictus and Ma. uniformis, which showed infection rates varying from 15% to 100% (Table 1). NA: not annotated at supergroup or strain level.

Infection Rate Variability between Studies Considering Widely Distributed and Studied Species
We observed that the detection of Wolbachia in culicids was carried out in mosquitoes from 52 countries, in Africa, Oceania, Europe, Asia, and the South, Central, and North America.
Thirty-one species were considered positive by some studies and negative by others regarding the presence of Wolbachia absence of Wolbachia in this species from Russia. The Wolbachia infection rate for Ae. aegypti and An. gambiae varied by site between 4.3% to 58% and 8% to 24%, respectively. It is possible that Wolbachia infection is limited to certain geographical regions and populations, but it is important to take into consideration the methodological approach employed by each study in order to confidently access the infection status of species and populations.
Ae. aegypti populations for the presence of Wolbachia, 31 populations were Wolbachia free. While four populations sampled in the USA, Malaysia, Thailand, India, Panama, and the Philippines were positive for Wolbachia from supergroups A, B, C, and wAlbB and wAegB strains. A similar pattern was seen for An. gambiae, evaluated in eight countries, in which populations from Burkina Faso and the Democratic Republic of Congo were Wolbachia free while two other studies reported infection, with the presence of wAnga strain in one of them. The same scenario occurred with Cx. pipiens, in which only one article registered the absence of Wolbachia in this species from Russia. The Wolbachia infection rate for Ae. aegypti and An. gambiae varied by site between 4.3% to 58% and 8% to 24%, respectively. It is possible that Wolbachia infection is limited to certain geographical regions and populations, but it is important to take into consideration the methodological approach employed by each study in order to confidently access the infection status of species and populations.

Discussion
Detailed knowledge of Wolbachia diversity and their ability to manipulate the host's reproductive system and affect the replication of viruses have established Wolbachia-based vector and arbovirus control as one of the most promising strategies to mitigate the impact of mosquito-transmitted pathogens. Currently, there are two main approaches that use Wolbachia with this aim [37,102]: I-the replacement strategy that releases males and females of Ae. aegypti transfected with Wolbachia to replace the natural population to one that is refractory to several arboviruses [23]; II-the second approach is conducted by releasing male mosquitoes infected with Wolbachia seeking to induce CI and reduce the mosquito population [38]. To effectively deploy either strategy, re-

Discussion
Detailed knowledge of Wolbachia diversity and their ability to manipulate the host's reproductive system and affect the replication of viruses have established Wolbachia-based vector and arbovirus control as one of the most promising strategies to mitigate the impact of mosquito-transmitted pathogens. Currently, there are two main approaches that use Wolbachia with this aim [37,102]: I-the replacement strategy that releases males and females of Ae. aegypti transfected with Wolbachia to replace the natural population to one that is refractory to several arboviruses [23]; II-the second approach is conducted by releasing male mosquitoes infected with Wolbachia seeking to induce CI and reduce the mosquito population [38]. To effectively deploy either strategy, researchers require precise information about any potential target mosquito species infected naturally by Wolbachia before and during any intervention, since it might have unexpected effects on the strategy employed [103].
This systematic review summarizes all available knowledge about Wolbachia in mosquito species focusing on the Culicidae diverse genera, the methodological approaches employed for Wolbachia detection and discuss the different sources of biases that can emerge due to methodological limitations and/or biological symbiotic factors.

Distribution of Wolbachia in Culicidae
The Culicidae family is one of the largest families of insects comprising more than 3574 known species [53]. Only 217 of these were evaluated for the presence of Wolbachia so far, which represents 6% of Culicidae species ( Figure 1B). Thus, the distribution and frequency of Wolbachia in more than 90% of Culicidae species remains unknown, as well as the possible existence of different strains and induced phenotypes in the host mosquito species. As expected, there is a strong correlation between the abundance of mosquito species per genera and the number of species per genera screened for Wolbachia (Figure 2). Anopheles, Aedes, and Culex, were the three most studied genera screened for Wolbachia comprising 75% of the total. The most studied species was Ae. albopictus (44% of articles). Such higher representativeness of these three genera was expected since they are also the most abundant and have enormous epidemiological importance transmitting several pathogens to humans [34,104,105]. However, it is important to note that very few mosquito species from other genera, such as Haemagogus and Sabethes that also transmit highly pathogenic pathogens, were barely screened for Wolbachia [106]. For instance, no species from the Sabethes genus was screened for Wolbachia so far and only one species from the Haemagogus genus was investigated. Other examples of mosquito species with epidemiological importance, such as Ae. furcifer, Ae. taylori, Ae. luteocephalus, and Ae. simpsoni that maintain the peridomestic cycle of YFV in different countries on the African continent, have not been screened for Wolbachia so far [107].
On the other hand, mosquitoes that belong to the genera Toxorhynchites, Malaya, and Topomania were not investigated at all, maybe because they are not considered of medical importance due to their non-hematophagous feeding habits [106]. However, knowing the Wolbachia diversity in these mosquitoes is important to understand the evolutionary history of Wolbachia since species from these genera are basal in the phylogeny of culicids. Moreover, in-depth knowledge of the interaction with their hosts may provide new Wolbachia strains that can be exploited for biological control. Thus, there are still major gaps in the knowledge about Wolbachia diversity in mosquitoes that should be addressed in the coming years.

Detection Methods and What They Can Tell Us
Molecular biology techniques continue to improve as the years pass, many developments have been incorporated to detect different Wolbachia molecules and cells in insects. However, each methodology has its strengths and weaknesses that can directly influence the results of Wolbachia diagnosis. Some factors must be considered before establishing the method to detect Wolbachia in mosquitoes including the sensitivity of the technique; the specificity of the chosen target; the biological characteristics of the bacterium and hosts in different scenarios, such as the possibility of the Wolbachia genome being integrated into the mosquito's genome, tissue tropism of Wolbachia infection and variable infection rate (see Section 3.3).

Amplification-Based Strategies
According to the results found in this review, 33% of the studies performed detection of Wolbachia in culicids using conventional PCR with only one Wolbachia target gene. Of these, the wsp gene that encodes the main protein on the bacterium's membrane surface, was the main target choice [108]. Although PCR is a sensitive technique that allows the detection of bacterium DNA even at low infection titer, it has some limitations, such as the inability to differentiate if Wolbachia DNA was derived from Wolbachia cells, a true Wolbachia infection, or if Wolbachia DNA is integrated into the host genome [109]. Several studies reported cases of lateral gene transfer (LGT) between Wolbachia and its hosts, such as the presence of fragments of the wBruAus genomic DNA in the X chromosome of the species Callosobruchus chinensis and evidence of transfer between the wMel strain and C. chinensis genome [110,111]. Conventional PCR that amplifies only one target is unable to differentiate a bacterium gene integrated in the mosquito's genome from an active and true infection by Wolbachia (Figure 4). face, was the main target choice [108]. Although PCR is a sensitive technique that allows the detection of bacterium DNA even at low infection titer, it has some limitations, such as the inability to differentiate if Wolbachia DNA was derived from Wolbachia cells, a true Wolbachia infection, or if Wolbachia DNA is integrated into the host genome [109]. Several studies reported cases of lateral gene transfer (LGT) between Wolbachia and its hosts, such as the presence of fragments of the wBruAus genomic DNA in the X chromosome of the species Callosobruchus chinensis and evidence of transfer between the wMel strain and C. chinensis genome [110,111]. Conventional PCR that amplifies only one target is unable to differentiate a bacterium gene integrated in the mosquito's genome from an active and true infection by Wolbachia (Figure 4). An alternative to circumvent this problem could be using a large set of Wolbachia gene targets (in a multiplex PCR or MLST) since the integration of several bacteria genes in the host's genome is less likely to occur [110]. Only three of the articles evaluated by this review applied the MLST technique, which consists in amplifying five conserved Wolbachia genes widely distributed in the genome, namely: gatB, coxA, hcpA, fbpA, and ftsZ [112]. Next-generation sequencing (NGS) employing long reads of mosquito and Wolbachia DNA may offer additional data that can help to distinguish between a true Wolbachia infection and integrated bacterium genomic fragments. Long DNA reads allow the detection of Wolbachia DNA integration sites into the mosquito genome or the confirmation of circular Wolbachia genomic DNA that further supports a true Wolbachia infection hypothesis [46]. However, many genome assembly parameters must be considered when analyzing these genomes, including genome coverage and sequencing depth. Triangles and squares represent DNA-based and visualization methods, respectively. Each circle represents one specific scenario and which of the following methods could be used to detect Wolbachia molecules. * methods that can suggest the presence of a bacteria establishing a true infection but cannot confirm if this infection is caused by Wolbachia.
An alternative to circumvent this problem could be using a large set of Wolbachia gene targets (in a multiplex PCR or MLST) since the integration of several bacteria genes in the host's genome is less likely to occur [110]. Only three of the articles evaluated by this review applied the MLST technique, which consists in amplifying five conserved Wolbachia genes widely distributed in the genome, namely: gatB, coxA, hcpA, fbpA, and ftsZ [112]. Next-generation sequencing (NGS) employing long reads of mosquito and Wolbachia DNA may offer additional data that can help to distinguish between a true Wolbachia infection and integrated bacterium genomic fragments. Long DNA reads allow the detection of Wolbachia DNA integration sites into the mosquito genome or the confirmation of circular Wolbachia genomic DNA that further supports a true Wolbachia infection hypothesis [46]. However, many genome assembly parameters must be considered when analyzing these genomes, including genome coverage and sequencing depth.
Another alternative is to use qPCR for Wolbachia genes, which may indicate a true infection. The bacterium titer variability is most likely explained as a result of a true infection and not a variable amount of Wolbachia genes integrated into the genome of different specimens, although Wolbachia titer variation may emerge from superficial mosquito contamination. Compared to conventional PCR, the other techniques cited are more laborious and expensive; however, they offer a more precise result for the origin of the detected Wolbachia gene.

Cell/Structure Visualization Strategies
Of the 59 articles analyzed, only one used a method for visualizing the bacterium, which was performed using MET. Although molecular techniques to visualize antigens or cell structures, such as FISH and MET are laborious, require expensive equipment and trained personnel, they can discern the first scenario (integration of Wolbachia gene into the mosquito genome) from a true Wolbachia infection, since the visualization of the bacteria and not just a single structure or gene trace confirm a true infection [46] (Figure 4). An alternative to the high cost required by these cited techniques is to perform a squash of the mosquito's ovaries followed by staining with May-Grunwald-Giemsa method (GIEMSA) or Gimenez staining to visualize pleomorphic structures suggestive of Wolbachia cells, through an optical microscope [61,113,114]. This simpler and cheaper technique was performed by a single study among all investigated in this review. Thus, despite the low amount of resources, there are alternatives for the correct detection of Wolbachia in culicids. For example, the use of conventional PCR associated with the visualization of pleomorphic structures in the ovaries of mosquitoes stained by GIEMSA, is able not only to confirm the Wolbachia infection but also to detect the specific presence of Wolbachia and possibly its lineage or strain.

Laboratory Colony Establishment of Field-Collected Population
The establishment of a colony of field-caught mosquitoes is one of the most complex methodologies to study Wolbachia infection. An adequate minimum insectarium structure for colony establishment is required with adjusted temperature, humidity, and specific light/dark cycles adjustments for each species [115,116]. In addition, many species reproduce in particular conditions in the field, such as An. gambiae whose males gather in swarms at specific mating sites or species that feed on specific hosts and plants, such as Uranotaenia macfarlanei whose preferred source of blood are amphibians [117,118]. Thus, the use of this approach could be required if there was no possibility to confirm an active Wolbachia infection by other methodologies, however, given the possibility to use simpler associated molecular methods, the establishment of laboratory colonies becomes a last resource. The establishment of field colonies is so labor-intensive that it was performed by a single study with Ae. aegypti, a well-known species regarding basic conditions required to raise and keep a laboratory colony.

Contamination Sources
The contamination of the mosquito samples by Wolbachia remains from environmental sources is an important source of bias that should be ruled out before Wolbachia infection is determined. Several potential contamination sources have been proposed including the environment, ecto and/or endoparasites [46]. Filarial nematodes of the Onchocercidae family have a mutualistic relationship with Wolbachia, whose supergroups C, D, J are present exclusively in these worms [119,120]. Many of these nematodes can be found in mosquitoes since they are involved in their transmission. For instance, Wuchereria bancrofti is commonly found in Cx. quinquefasciatus populations [121]. Therefore, the detection of Wolbachia in mosquitoes could be a result of filarial worm infection instead of a true mosquito infected by Wolbachia (Figure 4). The detection of Wolbachia strains belonging to supergroups other than A and B (commonly found in mosquitoes) should be taken with caution and further experiments are needed to evaluate the mosquito species infection [20]. Two studies from our systematic review detected Wolbachia from supergroups C and D in Ae. aegypti and An. baimai respectively. In these cases, the two previously mentioned approaches (PCR plus visualization by microscopy methods) are insufficient to differentiate contamination from true Wolbachia infection, as these methods do not exclude the possibility of contamination by worms. One alternative experiment to circumvent this scenario would be to perform an additional PCR targeting worm species in the samples considered positive for Wolbachia, a negative PCR would add evidence that the Wolbachia detection is a result of a true mosquito infection.
Several studies reported Wolbachia in Ae. aegypti, but the majority used only diagnostic amplicon-based molecular tools [51,[54][55][56][57][58] (Supplementary Material). Two studies went further and analyzed the presence of Wolbachia through maternal transmission and electron microscopy [45,52], but several criticisms have been made due to the lack of methodological consistency or experimental reproducibility issues [103]. Thongsripong et al., 2018 andCarvajal et al., 2019 detected Wolbachia belonging to supergroups C and D in Ae. aegypti, but using only amplification-based methods [51,58]. With the lack of further validation with complementary methodologies such results are likely an indication of the presence of contamination by other sources (worms, exuviae, etc.) since these supergroups of Wolbachia were not previously found in Diptera [21]. Even if Wolbachia is infecting Ae. aegypti natural populations, it remains to be assessed if it would induce any phenotype that could interfere with the effectiveness of the Wolbachia-based strategies being currently employed. If we consider Ae. aegypti bearing worms with Wolbachia supergroup C and D, it would mean that Wolbachia did not establish an endosymbiotic association with the mosquito and, therefore, no consequences would be expected to control program strategies.
The contamination of Wolbachia in mosquito samples can also derive from the external environment in which the mosquitoes insects with plants, water, or niches shared with other infected arthropods [122,123]. Wolbachia debris could be acquired by the mosquito feeding and, therefore, two approaches can be taken to differentiate a stable infection from contamination, the first of which is to visualize the bacterium by microscopic methods in mosquito ovaries. The second would be to collect water from breeding sites of the mosquitoes investigated and perform PCR for Wolbachia to exclude the possibility of environmental contamination.
Given these examples, it is possible to see that the detection of a true Wolbachia infection is a challenging task that can be highly impacted by the choice of the appropriate technique. Thus, the best way to correctly infer Wolbachia infection is to choose the best set of complementary techniques that can discern the different possible scenarios regarding the presence and/or infection of Wolbachia in mosquitoes (Figure 4).

Wolbachia Detection in Different Mosquito Populations: The Symbiotic Population Dynamics Hypothesis
Several biological phenomena derived from the intricate symbiotic relationship between Wolbachia and its hosts must be taken into consideration to interpret the Wolbachia detection results using different methodologies. Some species investigated in this review were reported both negative and positive for Wolbachia infection. In some cases, contaminated samples, misidentified species or low sensitivity of the technique used to detect Wolbachia molecules or cells can lead to divergent results regarding the presence of the bacteria in a given species/population [103]. However, this discrepancy can also be a result of the host-parasite population dynamics itself. Wolbachia infection rate is not uniform throughout every population due to a series of environmental and biological factors. Since Wolbachia infection is dynamic and mosquito populations can be very large and widely distributed, the infection rate in mosquitoes in each population can be variable, then the sampling of a few individuals from single or few sampling sites might not represent the full dynamics of the Wolbachia infection at the population level as a whole. Due to methodological differences employed by studies that investigated populations of the same species, it is not straightforward to find reliable examples of such biases. Only one clear example of the species Ae. cantans, from populations from Italy and Russia was detected in this systematic review [85,109] To avoid sampling bias a minimum number of specimens that is sufficient to represent different populations of the species distributed along the host species niche is necessary to determine whether a species of Culicid does or does not harbor Wolbachia naturally and complementary methodologies should be employed to assess the real infectious status of the species.
The minimum number of specimens necessary to accurately assess the infection status of a species is difficult to determine once each host species has different population size and may have different symbiotic relationship with Wolbachia. Some mosquito species are present in large population sizes, such as cosmopolitan species Ae. aegypti and Cx. quinquefasciatus. In this case, a low number of diagnostic specimens will likely not reveal the true Wolbachia infection rate [124,125]. While other species are present in limited population size and inhabit very specific niches, such as species from the Sabethes and Haemagogus genera [126], where a limited number of specimens would be reasonable. On the other hand, the Wolbachia symbiotic relationship with the mosquito host can also impact the minimal number of specimens to be investigated since Wolbachia strains that can manipulate the host reproductive systems can reach a high infection rate. Hence, few host individuals would be enough to evaluate the infection status of a given host species. While strains that do not induce any phenotype may have a very low infection rate and a much higher number of host specimens would be required to ascertain its infection status. Although there is no general rules to define the minimum number of specimens needed to accurately evaluate the infection status of a giving mosquito species, there are some interesting guidelines on disease ecology that highlight several sources of bias that can be readily transferred to the Wolbachia-host relationship dynamic such as Colvin et al., 2015, andLachish andMurray, 2018 [127,128].
Ae. aegypti and An. gambiae are two epidemiological relevant species in which the presence of Wolbachia has been investigated by several studies [46,103]. So far, six Ae. aegypti natural populations from different countries have been Wolbachia positive, in addition to a characterization of a new strain (wAegB from supergroup B) [45], while several other studies analyzing several populations from 27 different countries, have not identified the presence of Wolbachia in this species [129]. Comparing these studies is difficult since none of them used the same methodological approach, with the same Wolbachia markers or with the same number of screened mosquitoes. These diverging results can be a result of methodological and/or experimental biases as well as something derived from the population dynamics between symbionts that may be influenced by the environment and interaction with other species. Some studies have demonstrated that high temperature influences the Wolbachia titer present in mosquitoes, as observed in An. stephensi in the laboratory and in field tests with Ae. aegypti [130][131][132]. Such a phenomenon may lead to the elimination of Wolbachia infection in a given population. Environmental characteristics are known to influence the establishment of the mosquito microbiota, where populations sampled from different sites show different midgut bacterial composition due to factors such as, breeding water composition, temperature, and anthropogenic activities in these regions [133]. Thus, these factors could also influence the presence of Wolbachia in certain species facilitating or preventing the symbiotic establishment between the host and the bacteria. As several species of epidemiological importance are present in regions with different geographical characteristics, the symbiotic population dynamics might explain the divergence for the results regarding the presence of Wolbachia in different mosquito populations. A better understanding of how these biological factors influence the presence of Wolbachia in mosquitoes is necessary to plan Wolbachia infection surveillance in different mosquito populations and guide intervention measures appropriately.

Materials and Methods
This systematic review follows the criteria established by PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) and the checklist for reporting systematic reviews and meta-analyses [134].

Search Strategy
A search in the PUBMED platform was performed between March and July 2020 with the following terms: (1) mosquito OR vector OR Culicidae AND Wolbachia AND infection, (2) mosquito OR vector OR Culicidae AND Wolbachia AND detection, (3) mosquito OR vector OR Culicidae AND Wolbachia AND surveillance, (4) mosquito OR vector OR Culicidae AND Wolbachia AND distribution. The last search was performed on 2 July 2020.

Eligibility Criteria
The inclusion criteria selected were: articles that detected Wolbachia strains in natural populations (directly sampled from the field) of insects of the Culicidae family, regardless of the methodology chosen for detection, study publication year, or region. The exclusion criteria were: (1) review articles, (2) detection of Wolbachia in insects other than the Culicidae family, (3) detection of the Wolbachia only in cell culture, (4) detection of the bacteria in mosquito colonies, (5) articles that validate a molecular detection technique, (6) articles that detected Wolbachia in Culicidae from places where mosquitoes infected with Wolbachia were released, (7) articles not published in English language.
These criteria were established to understand the natural distribution and diversity of Wolbachia pipientis in culicids of different genera and locations.

Study Selection
The initial search on National Centers for Biotechnological Information (NCBI) returned 1431 articles. After duplicates and screening using the inclusion and exclusion criteria, 59 articles remained and were used to extract the relevant data ( Figure 5).

Data Extraction
These were the data extracted from the selected articles were: (1) title of the article, (2) year of publication, (3) Culicidae species analyzed for Wolbachia detection (4) Infection status by Wolbachia: positive or negative; in case of positive infection (5) infection rate and (6) Wolbachia supergroup or strain, (7) detection method used, and (8) mosquito collection site (Table S1).
The analysis excluded which stages of the mosquito life cycle were used for detec-

Data Extraction
These were the data extracted from the selected articles were: (1) title of the article, (2) year of publication, (3) Culicidae species analyzed for Wolbachia detection (4) Infection status by Wolbachia: positive or negative; in case of positive infection (5) infection rate and (6) Wolbachia supergroup or strain, (7) detection method used, and (8) mosquito collection site (Table S1).
The analysis excluded which stages of the mosquito life cycle were used for detection, the different infection rates between males and females (only the general percentage of mosquito infection was retained), and taxonomic classifications lower than species level for culicids.
When the data were unclearly described by the authors, the annotation was performed as without information, for example, the absence of Wolbachia supergroups or strains description and the absence of infection rate information. Regarding the sites, only the countries in which the Culicidae collections were provided were recorded, and no other territorial designations, such as cities and/or districts.
Some studies used more than one detection method for Wolbachia infection detection, however, the species of culicid was considered positive if Wolbachia was detected in at least one of the techniques.
Finally, the nucleic acid sequencing method was not considered as a method to detect Wolbachia, but as a method to classify into supergroups or as a way to infer relationships between the bacterium and its hosts.

Conclusions
Wolbachia is an endosymbiotic bacterium with large biotechnological interest for vector/disease control. Despite the great advances and discoveries made on Wolbachia mediated host physiological changes, its presence in several species and genera of culicids, is not well described yet. The absence of the correct classification of the bacterium in supergroups as well as the lack of consensus on the establishment of a standard methodology capable of discerning between a true Wolbachia infection or other sources of Wolbachia molecules, hinder proper comparison between studies and obscure the evolution at species and population level of this bacterium as a whole. Thus, more precise investigations in a wide range of mosquito species must be performed to allow a better understanding of the natural Wolbachia infection in Culicids. Such information will be crucial to planning and implementing different strategies that use Wolbachia to reduce vector population and/or decrease the public health burden of different mosquito-borne pathogens.
Supplementary Materials: The following are available online at https://www.mdpi.com/2076-081 7/10/1/39/s1, Table S1: All data used to review the articles, including search terms, articles selected, and positive and negative species.
Funding: This research received funding for a master scholarship of LMIS from Fundação de Amparo a Ciência e Tecnologia do Estado de Pernambuco (FACEPE) and doctorate scholarship for FZD from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES). GLW is supported by the National Council for Scientific and Technological Development by the productivity research fellowship level 2 (303902/2019-1).

Data Availability Statement:
The data presented in this study are available in Supplementary Material.