Genetic diversity of Dengue virus serotypes circulating among Aedes mosquitoes in selected regions of Nigeria

The DENV mosquito vector is endemic to tropical and subtropical climates, placing ∼ 40% of the world’s population at direct risk of dengue infection. Currently, in Nigeria the status of DENV serotypes circulating among mosquitoes’ vectors is unknown. Our study was designed to identify and characterize the DENV serotypes circulating in Aedes mosquito populations collected in selected sites in Nigeria. The mosquitoes were collected, identied morphologically to species level using colored identication keys of Rueda. Generally, each species identied was tested in pools of 20 individuals of each Aedes species. RT-PCR and nested PCR were used to detected DENV serotypes in mosquitoes and characterized using Sanger sequencing methods. The results show that DENV serotypes were detected in 58.54% of the pools of mosquitoes screened. This corroborates with our unpublished report on the presence of DENV IgM antibodies in humans in the same study area. All DENV1-4 serotypes were detected in Ae. aegypti, Ae. albopictus and Ae. gaillosi with DENV4 serotype being reported for the rst time in Nigeria. DENV2 (37.8%) was the most detected serotypes, while double and triple co-infections of serotypes were detected in 24.4% of the pools. Phylogenetic analysis revealed a strong evolutionary relatedness of DENV serotypes in our study with that of South and Southeast Asia, North America, and other African countries. This is the rst reports on the co-infection of natural DENV in Aedes species pools in Nigeria. Thus, our study speculates a possible linkage between DENV serotypes and febrile u-like disease burden being experienced by host communities in northern Nigeria.


Introduction
Dengue is a mosquito-borne viral disease that has spread to almost all the regions of the world in recent years. Climate change and globalization has rmly established Dengue as threat to public health systems (1). Dengue virus (DENV) is transmitted by infected female mosquitoes mainly of the species Aedes aegypti and to a lesser extent, Aedes albopictus. These are also vectors of Chikungunya, Yellow fever and Zika viruses (1). The Ae. aegypti is recognized as the main vector of DENVs worldwide, largely attributable to its vector competency and strong host preference for humans compared to Ae. Albopictus (2). However, Ae. albopictus has been a driving force in the worldwide emergence of Chikungunya virus since 2004 (3).
The African Aedes mosquito breeds in both the domestic environment and ancestral sylvatic habitat. Whereas domestic Aedes mosquito larvae develop in arti cial containers (tires, cans, jars, owerpots, plastic/metal drums) within or proximity to human habitation, while larvae of the sylvatic ecotype are found in natural breeding sites such as rock pools, phytotelmata in forested areas (4). High populations of Aedes mosquitoes were reported in Adamawa state (5).
It has been estimated that 3.9 billion people in 129 countries are at risk of DENV infection and 390 million dengue infections occur per year of which 96 million manifests clinically. In addition, the largest number of dengue cases was recorded in 2019 that affected most regions of the globe, and dengue transmission was recorded in some countries for the rst time (1). Despite its global signi cance, little attention has been given to Aedes-borne arboviruses in Africa. These posed a greater public health threat for immunologically naive non-African human populations. The neglect of these viruses could be due to the high burden of malaria and other neglected tropical diseases in Africa, in addition to poor-resource settings for accurate identi cation of arboviral infections (6). In Nigeria, most cases of dengue are undiagnosed or misdiagnosed as malaria or referred to as fever of unknown cause (7). Nigeria's urban ecology, with limited sanitation infrastructure, multiple rainy seasons, pervasive household water storage, and virtually no public awareness of dengue transmission, presents a similarly prime environment for breeding of Aedes mosquitoes. The vectorial capacity of Aedes species in Nigeria is still poorly studied (8). A study conducted in Zaria, Kaduna Nigeria reported co-circulation of dengue and yellow fever viruses in the Aedes populations (9). A prevalence rate of 51.9% of IgM antibodies to dengue fever virus was reported in the northwest zone of Nigeria (10).
DENV is a positive sense, single-stranded RNA virus of the family Flaviviridae which comprised of more than 70 viruses. The viruses have four antigenically distinct serotypes, DENV 1-4 (11). These viruses have a broad geographic range, circulating on every continent except Antarctica (12). The virus circulates in two distinct transmission cycles; the sylvatic cycle between the mosquito vector and non-human primates and the human transmission cycle which includes domestic and peri domestic Aedes mosquitoes (13). Limited available evidence suggests only Ae. aegypti is involved in transmission of dengue in Nigeria, as several isolates of DENV were obtained from this mosquito species and none from Ae. albopictus and Ae gaillosi (14). Also, despite the growing effect of dengue infections amongst human populations, there was virtually no information on the natural presence of DENV serotypes circulating in adult eld Aedes mosquitoes' populations in Adamawa region of Nigeria. Therefore, this study was designed to evaluate and characterized DENV serotypes circulating in wild caught Aedes mosquitoes in this area.

Study Area
The study was carried out in three purposively selected Local Government Area (LGAs) of Adamawa State, Nigeria that is located between latitude 9 0 14 ′ N and longitude 12 0 28 ′ E (Fig. 1). It has an estimated area of 39,742,12sq. Km which accounts for 4.4% of the total landmass of Nigeria. The projected population as per 2006 population gure stands at 3,737,223 and it is traversed by mountainous landforms like the Mandara Mountains, Cameroon Mountains and Adamawa hills and large rivers: Benue, Gongola and Yadzarem. The vegetation in Yola and environs is secondary type due to human activities through construction, farming, wood gathering for fuel and grazing which have altered the natural vegetation. Adamawa state is prone to ood water and have swampy terrain especially during the rainy season. The mean annual rainfall ranges from 700 mm in the northeastern and of the 1600 mm in the southern part (15). The study areas are also characterized with high population movement due upheavals in the northeast with many people storing water around their homes and dumping of solid waste in open gutter and close to water bodies, which provide suitable environment for breeding of Aedes species and interaction with humans (16).

Aedes mosquitos' collection and preparation for detection of DENV serotypes
Aedes species were collected both indoors and outdoors from Mubi, Yola and Numan LGAs. By volunteers for four months (July to October). Collection was in the morning between 6:00-10 am); while the evening collection was from 3:00 to 7:00pm. Collection consisted of a combination of aspiration of indoor resting adults, human-landing techniques that attracted mosquitoes on exposed parts of human-bait and sweep nets (17,18). All mosquitoes collected were sorted and Aedes mosquitoes were identi ed using colored identi cation /taxonomic keys by Rueda (19). The Aedes species identi ed morphologically were segregated generally, in pools of 20 females labeled according to species, LGAs and communities. A total of 41 pools were used. Sixteen (16) pools from 3 communities in Numan, 17 pools from 3 communities in Yola and 8 samples pools from 3 communities in Mubi were analyzed (Fig. 1). Samples were stored in − 80 0 C for RNA extraction.
RNA extraction and cDNA synthesis from pools of Aedes species DENV RNA was extracted with All Prep DNA/RNA Mini extraction kit (Qiagen, Germany, Cat. No. 80204) in a Level 3 (L3) Safety Laboratories for Viral Research, University of Bremen, Germany, following manufacturer's instruction with little modi cation (www.giagen.com).Each pool of Aedes species was thoroughly homogenized with a mini handheld homogenizer and single-use microfuge pestle (Sigma Aldrich, Munich, Germany) in 350 µL buffer RLT (lysis buffer). The homogenate was centrifuged for 3 mins at 13000 rpm. The homogenized lysate was transferred to an AllPrep RNA spin column in a 2 mL collection tube and one volume (350 µL) of 70% ethanol was added to the ow-through and mixed thoroughly by pipetting repeatedly. Seven hundred microliter (700 µL) of the lysate was transferred to an RNAeasy spin column placed in a 2 mL collection tube and was centrifuged for 15 secs at 13000 rpm. The ltrate from the centrifuge was discarded and 700 µL of buffer RW1 (washing membrane-bound RNA) was added to the RNeasy spin column, centrifuged, and discarded. Five hundred microliter (500 µL) of buffer RPE was added twice to the column, centrifuged for 2 mins at 13000 rpm and the ow-through was discarded. The column was dried centrifuging for 1 min at 13000 rpm after which 30 µL of RNA free water was used to elute the RNA into a 1.5 mL collection tube by centrifuging for 1 min at 13000 rpm. The RNA was quanti ed using Nanodrop 1000 apparatus (Thermo Scienti c, Dreieich, Germany) at a wavelength of 260 nm. The RNA extracted was stored at − 80°C until used. RNA quality and integrity were determined using cytochrome oxidase 1 (COI) gene forward and reverse primers (Table 1). RNA was ampli ed in a 25 µL reaction mixture containing 2.5 µL (10x dreamTaq buffer containing 20 MgCl 2 ), 0.5 µL (dNTPMix,10 mM),1.0 mL (25mM of MgCl ), 0.5 µL (500U, 5U/µL green dream Taq polymerase) and 5µL of RNA template. The temperature cycling conditions are as follows: initial denaturation at (95°C, 5 min), 35 cycles of denaturation (94°C, 1 sec), primers annealing (55°C, 1 min), primer extension (72°C, 2 mins) and nal extension at 72°C for 10 mins. For cDNA syntheses, 80ng of RNA extract were mixed with 1µL (0.2µg/µL) random hexamer primer and 1µL (10 mM) dNTP mix in a 0.2 mL PCR tube. The mixture was made up to 15 µL with nuclease-free water on ice. The following component was then added to the PCR tube: 4 µL of 5X reaction buffer, 0.5 µL of Ribolock RNase inhibitor (40U/µL) and 0.5µL of Maxima H minus Reverse Transcriptase (10000U, 100 U/µL) making up to a total volume mixture of 20 µL. The reaction was vortexed, centrifuged and incubated in a thermocycler with the following conditions: 25 0 C for 5 mins, 50 0 C for 30 mins. The reaction was then terminated by heating at 85 0 C for 5 mins (Thermo scienti c).

Serotyping Of Denv Serotypes
The PCR product obtained from the RT-PCR ampli cation using D1 and D2 as primers was further used as the template for DENV1-4 different serotypes. Based on the above temperature conditions and reagent volumes, 0.5 µL and 1µL of the Nested 1 RT-PCR product (1:10 and 1:1000 in ddH20 for DENV serotypes 2 and DENV serotypes 1, 3 and 4 respectively) was used as template in the subsequent nested PCR reaction. A 25 µL reaction volume containing forward primer D1 and type-speci c (TS) reverse primers: TS1, TS3 and TS4 (reverse primers for serotype 1, 3 and 4), and forward primer DV1 and DSP2 reverse primers for serotype 2 (Table 1) was further ampli ed by nested PCR step II. The entire PCR product was resolved on a 3% agarose gel containing 0.5 µg/mL ethidium bromide (SERVA, Heidelberg, Germany). The expected sizes of the ampli ed products were 482 for DENV1, 362 for DENV2, 290 for DENV3 and 392 for DENV4 (20).

Sequencing And Phylogenetic Analysis
The PCR amplicons were puri ed using GeneJet DNA puri cation kit (Thermo Scienti c, Dreieich, Germany) following manufacturer's protocols and sequenced at Microsynth Sequencing Laboratories Göttingen, Germany. Forward and reverse strands were sequenced, and ampli ed target bands of Sanger sequence chromatograms were assembled and visualized using Geneious Pro Version 5.5.9. The sequences were compared with available sequences using Basic Local Alignment Search Tool (BLAST) for Virus Pathogen Resources (ViPR) (https://www.viprbrc.org/brc/blast.spg) and the Genbank data base to validate the identity of the virus isolate. Phylogenetic analysis was inferred using MEGAX (43).
Consensus sequences were aligned using Muscles alignment tool. The phylogenetic tree was reconstructed using a Maximum Likelihood (ML) method and estimated using best-t General Time Reversible (GTR) model with gamma-distributed rate variation among sites. Other sequences like the study sequences in GenBank were obtained using the BLAST algorithm were also included in the analysis. Bootstrap replicates of 1000 were employed to assess the robustness of individual nodes of phylogeny. Complete gap deletion was employed for all the E gene sequences.

Data Analyses
Descriptive statistics was used to analyse the distribution of DENV serotypes in Aedes species in Adamawa state. The proportion of DENV positive individuals True Infection Rate per 100 mosquitoes was calculated and the minimum infection rate (MIR) was also estimated per 1000 Aedes mosquitoes.

Detection of DENV serotypes in Aedes species
A total of 706 Aedes mosquitoes were collected from the study locations. Samples were identi ed morphologically and con rmed molecularly targeting CO1 gene. Overall, 58.5 % (24/41) pools were positive for DENV in the study locations ( Table 2). Mubi had the highest 100% (08/08) pools positivity rate while Numan 43.75% (07/16) with the lowest DENV pools positivity rate. The proportion of DENV positive individual's true infection rate (TIR: estimated number of positive mosquitoes per 100 mosquitoes tested) determined to 5.24% (37/706 x100), corresponding to presumed infected Aedes species (9,12 and 16 detected serotypes from Numan, Yola, Mubi respectively) per total of 706 Aedes mosquitoes processed in all the pools as some of the pools could not reach 20 mosquitoes. The Minimum Infection Rate (MIR) was 33.99% (24 positive pools/706 total processed individual mosquitoes × 1,000). There was a strong indication from the estimation that several individual mosquitoes were infected with the same serotypes thereby increasing the proportion of infection in individual's mosquitoes. However, since our study design did not permit us to separate the mosquitoes from each other, we decided to estimate the proportion as simply as possible.

Phylogenetic analysis of DENV serotypes distribution in Aedes species
Out of the 27 samples puri ed and sequenced, only 11 samples had good quality reads, which was used for the phylogenetic analysis for DENV1-4. The genetic relationship of DENV strain was analyzed targeting the E gene sequences of the virus. The tree was generated using E gene sequences of only one (1) DENV1 serotype and 7 reference strains (Fig. 5A). Notably, DENV1 from the study location form clade with that of South East Asia, West Africa, and North America. This study indicated diverse geographical distributions of DENV1 serotype in different continents. Also, 7 DENV2 E gene sequences were aligned with 7 reference strains from Genbank (Fig. 5B). The DENV2 from Mubi in Adamawa, Nigeria clustered together with that of Cameroon, Ghana, Senegal, and Kenya (Africa continent) and Philippines in Asia continents. While four of the DENV2 from Yola, Numan, and Mubi formed a separate clade.
Furthermore, one (1) DENV3 serotype from Yola, Adamawa Nigeria was determined with that of 10 reference serotypes from Genbank (Fig. 5C). DENV3 serotype from Adamawa, Nigeria formed a separate cluster which is distantly related with the referenced DENV3 serotype. However, DENV3 serotype from Kenya (East Africa) is closely related with DENV3 from India (South Asia). Only 2 DENV4 serotype were used from the study location with 9 referenced DENV4 (Fig. 5D). The result of the Phylogenetic analysis revealed that DENV4 from Yola appears to be more closely related to viruses from Asia continents. While DENV4 from Mubi is distantly related to the reference viruses.

Discussion
Dengue virus (DENV) is considered endemic in Africa (21), but there has been limited information documenting active DENV transmission in Nigeria with few of the available information relying hugely on serological surveys and has not fully described the circulating serotypes in eld caught Aedes mosquitoes. Our study indicated that Aedes aegypti, Ae. galloisi, Ae. albopictus, Ochlerotatus vigilax and subspecies of Ae. aegypti formosus occurred with variable abundance in Adamawa State, Nigeria with the four DENV serotypes circulating in the study area ( Table 3).
The circulating DENV serotypes were detected in 58.54% of the Aedes mosquito pools (Table 2). This rate is lower than 62% detected in Ae. aegypti pools collected at rural areas in Colombia (22). This is high in comparison to 33% and 11% detected in adult mosquito pools in Armero and Brazil respectively (23,24). The high pools positivity rate obtained in our study could be due to an increased interaction between Aedes species and humans, because the Aedes mosquitoes were collected at the peak period of mosquitoes' activities during the raining season, July-October. This agrees with Halstead, (44) who stated that DENV peak transmission is usually associated with periods of higher rainfall in most dengueendemic countries. Also, it might have been exacerbated due to changes in human behavior that favored the Aedes mosquitoes in the study locations, which may not be unconnected to the protracted humanitarian crisis in the northeast that lasted for over 10 years displacing millions of people from the neighboring state who settles in the study areas thereby overstretching the infrastructure and increasing interaction between Aedes species and humans. Moreover, it could also be speculated due to increase in the cross-border socioeconomic activities between the state and Cameroon since they share borders.
Thus, highlighting the possible cross-border importation of the vectors and DENV. This is a latent potential reservoir which can spill over to other parts of the country if left unchecked.
Virtually, all the Aedes mosquitoes' pools in Mubi were positive 100 % (08/08) for DENV (  (25). However, the high infection rate reported in our study may not be due to carry-over contamination but rather an attestation to a possible and continual transmission of DENV in the study area.
Our study detected DENV2 in Ae. galloisi and DENV1-4 serotypes circulating in Ae. albopictus (Fig. 3). This is the rst time we are reporting this occurrence in Nigeria. However, DENV2 has been proven experimentally in Ae. galloisi and Ae. albopictus in Japan (26). Similarly, Paupy (27), reported Ae. albopictus as being susceptible to DENV2 and highly competent vector for CHIKV in Cameroon. Thus, our study here con rmed the susceptibility of Ae. albopictus infection to all the four DENV serotypes. Ae. albopictus (Asian Tiger mosquito) is an invasive species in Nigeria and can easily spread through international trade especially of used tires and transportation. It is known to exhibits strong physiological and ecological adaptability. There are several reports of epidemics of CHIKV and DENV infection coinciding with Ae. albopictus spread in Cameroon (28).
Also, Ae. albopictus was reported as the main vector of dual outbreaks of CHIKV and DENV in Gabon (29). Changes usually occur in the epidemiology of arboviruses after an introduction of invasive species (30). Ae. albopictus is competent enough to propagate zoonotic pathogens across humans and animal species. In 2014, part of the study areas was captured by the insurgents who led to the displacement of thousands of residents who took refuge in Cameroon. Thus, our results further con rm the possibility of cross border transmission since there are reports of outbreaks associated with Ae. albopictus in Cameroon. Therefore, we have an emergency in our hands to curtail before it gets out of control as the risk of infection with DENV serotypes could increase in the area.
We detected all the four DENV serotypes circulating in the pools of Aedes mosquitoes from Mubi and Yola, while DENV3 was not detected in Numan (Fig. 3). This is not surprising because Mubi and Yola share similar characteristics as the main entry points into the state for travelers and visitors from Cameroon and different parts of the world because of international airport in Yola. In addition, many IDPs camps and humanitarian partners are situated in Yola and Mubi, accommodating thousands of refugees from Borno and different parts of Adamawa state, affected by the insurgency. Besides, dengue control activities are not practiced in the study areas. The sites are characterized with large river Yedzaram and river Benue, the people engaged in shing and irrigation farming around the rivers, household water storage, digging of earth for clay, damaged septic tanks, indiscriminate discarded used food containers such as cans and old tire and presence of solid waste blocking water channels and around water bodies (32,5), such practices provides good and abundant breeding sites for Aedes species and increase DENV serotypes susceptibility in the study areas.
All the four serotypes of DENV have been detected in Africa. However, in Nigeria, Ayolabi (33) reported DENV1 and DENV3 as the active circulating serotypes in Lagos, Nigeria from serum samples taken from febrile patients. Similarly, DENV1 and DENV2 serotypes were found circulating in serum samples collected in Cross River, Nigeria (34). In addition, DENV3 was ampli ed in Mansonia africana in Bayelsa, Nigeria (25).
Interestingly, our study detected DENV4 serotype, which has not been previously reported in Nigeria. This may be due to vector evolutionary competence or from the invasive arthropods' species found in the area since we detected DENV4 also in Ae. albopictus. This is a serious public health concern since mosquitoes are usually infected for life and infections in humans sometimes can be asymptomatic (35). Therefore, the circulation of DENV in humans and mosquito populations could be maintained for an exceptionally long periods and portend a public health hazard. Also, being infected with one serotype does not make someone immune from being infected with the other serotypes. Circulation of multiple serotypes could also lead to increased cases of severe form of dengue such as Dengue Shock Syndrome (DSS) and Dengue Hemorrhagic Fever (DHF) (36).
Furthermore, DENV2 was the most frequent serotypes detected in our study followed by DENV1 (Fig. 4A).
This agrees with previous research by Amarasinghe (21) that reported DENV2 as the cosmopolitan serotypes and cause most epidemics in the Africa continent. However, in contrast Konongoi (31) who reported DENV1 as the most common serotypes circulating in serum samples collected in selected regions of Kenya.
The study showed concurrent detection of more than one DENV serotypes in single pools of Aedes species (Fig. 4B). Co-infection of individuals' mosquitoes is not common; however, Thavara (37)  Infection of larvae with up to three DENV serotypes is possible experimentally (39). Although there is paucity of information on natural infections of four serotypes in an individual Aedes or humans, that narrative also could be possible during intense transmission in hyperendemic areas, due to transovarial transmission. However, the concurrent infections detected in this study may likely re ect single infections per Aedes mosquito. The serotypes can remain in the vector for transmission which could cause future severe dengue related cases with sequential secondary infection in the same individuals (22).
Our study investigated, variation in DENV serotypes using sequence analysis. DENV 1, 2 and 4 serotypes (Fig. 5A, B and C) respectively from the study locations formed clades together with that of South and Southeast Asia, North America and other African countries indicating strong evolutionary relatedness of the virus, signifying a common ancestor. Our study corroborates with that of Konongo (31) who reported relatedness of DENV serotypes in Kenya (East Africa) with that of South and Southeast Asia. Therefore, these countries portend high-risk areas for DENV because of the possibility of importation and exportation of DENV serotypes from any of these continents. Thus, our study reveals continued spread and wide geographic range of DENV serotypes.
However, DENV2 (Fig. 5A) in our study showed different level of relatedness. Some of the DENV 2 serotypes from the study location showed close relationship while others are distantly related with the referenced DENV sequences used. The differences observed could be due to possible genetic distances since evolution is a continuous process and geographic variations are inevitable. Point mutation could occur in lineages over time which could lead to divergence (40). The study also indicated DENV3 as distantly related with serotypes from Asia and African countries suggesting possible variation from the other serotypes.
Our study further con rms the reports that the major epidemics thought to be dengue, emanated from three continents: Asia, Africa, and North America (42). There is a sudden increase in the number of imported DENV cases worldwide due to globalization (41). Adamawa is a state in Nigeria, a country located in West Africa region and shares a porous border with many African countries and accelerated in global trade and bilateral exchange program with many Asian countries such as China, India, Philippines, Malaysia, and Dubai. It is imperative to note that travelers may serve as vehicles for DENV transmission, since 50% of the world's populations are living in dengue endemic countries (31).
In conclusion, our study detected all the DENV1-4 serotypes circulating in Aedes species in Adamawa state and reporting for the rst time the occurrence of DENV4 serotype, as new serotype, which has not been previously described in Nigeria. Also, the isolation of DENV in Ae. albopictus and Ae. galloisi present a new threat of increase in vector competency in the transmission of DENV across animal species and to humans. And may likely be capable of transmitting other arboviruses. There was strong evolutionary relatedness of DENV serotypes in our study with that of South and Southeast Asia, North America, and other African countries. Our study also speculates possible linkage between the humanitarian crisis in the northeast and sudden surge of DENV serotypes infection in Aedes species. Therefore, it identi ed potential virus reservoir which may likely spread in humans during populations' movement. This is a wakeup call for prevention of future epidemics.

Declarations Acknowledgements
We wish to appreciate the support of all the trained eld volunteers who assisted in the mosquitos'

Availability of data and materials
Data supporting the conclusions of this article are included within the article.

Declarations
Ethics approval and consent to participate  Occurrence of DENV serotypes and Fig 4B, co-infections of DENV serotypes in pools of Aedes mosquitoes in Adamawa state.