Risk of Arboviral Transmission and Insecticide Resistance Status of Aedes Mosquitoes during a Yellow Fever Outbreak in Ghana

Background In late 2021, Ghana was hit by a Yellow Fever outbreak that started in two (2) districts in the Savannah region and spread to several other Districts in (3) regions (Oti, Bono and Upper West). Yellow fever is endemic in Ghana. However, there is currently no structured vector control programme for the yellow vector, Aedes mosquitoes in Ghana. Knowledge of Aedes bionomics and insecticide susceptibility status is important to control the vectors. This study therefore sought todetermine Aedes vector bionomics and their insecticide resistance status during a yellow fever outbreak. Methods The study was performed in two yellow fever outbreak sites (Wenchi, Larabanga) and two non-outbreak sites (Kpalsogu, Pagaza) in Ghana. Immature Aedes mosquitoes were sampled from water-holding containers in and around human habitations. The risk of disease transmission was determined in each site using stegomyia indices. Adult Aedes mosquitoes were sampled using Biogents Sentinel (BG) traps, Human Landing Catch (HLC), and Prokopack (PPK) aspirators. Phenotypic resistance was determined with WHO susceptibility tests using Aedes mosquitoes collected as larvae and reared into adults. Knockdown resistance (kdr) mutations were detected using allele-specific multiplex PCR. Results Of the 2,664 immature Aedes sampled, more than 60% were found in car tyres. Larabanga, an outbreak site, was classified as a high-risk zone for the Yellow Fever outbreak (BI: 84%, CI: 26.4%). Out of 1,507 adult Aedes mosquitoes collected, Aedes aegypti was the predominant vector species (92%). A significantly high abundance of Aedes mosquitoes was observed during the dry season (61.2%) and outdoors (60.6%) (P < 0.001). Moderate to high resistance to deltamethrin was observed in all sites (33.75% to 70%). Moderate resistance to pirimiphos-methyl (65%) was observed in Kpalsogu. Aedesmosquitoes from Larabanga were susceptible (98%) to permethrin. The F1534C kdr, V1016I kdr and V410 kdr alleles were present in all the sites with frequencies between (0.05–0.92). The outbreak sites had significantly higher allele frequencies of F1534C and V1016I respectively compared to non-outbreak sites (P < 0.001). Conclusion This study indicates that Aedes mosquitoes in Ghana pose a significant risk to public health, and there is a need for continuous surveillance to inform effective vector control strategies.


Introduction
Aedes mosquitoes represent an ever-growing threat to public health worldwide due to their ability to transmit many infectious arboviral pathogens such as Dengue, Chikungunya, Zika and Yellow Fever (1).Yellow fever (YF), an acute viral disease affecting humans and non-human primates (NHP), is caused by the yellow fever virus (YFV) (2).The virus is transmitted by the bite of infected female Aedes mosquitoes (3).Aedes aegypti is one of the vectors for yellow fever in Africa alongside Ae. albopictus which is known to be more invasive and also a competent vector for Dengue fever and Chikungunya (4).The World Health Organisation (WHO) has reported that forty-seven (47) countries in Africa are either endemic or have regions that are endemic for YF and other arboviral infections (5).Furthermore, the WHO advises countries that have Aedes mosquitoes but no evidence of viral transmission to identify local regions with high mosquito densities and make proper preparations to deal with any possible arboviral outbreaks (6).The disease affects over 200,000 people and causes about 30,000 deaths annually (7).In Africa, annually an estimated 21 million people are at risk (8).
Yellow Fever is endemic in Ghana; this situation is exacerbated by the limited vaccine coverage, vector abundance and increasing insecticide resistance (9)(10)(11)(12), creating a risk for onward transmission and ampli cation of the virus among unvaccinated populations (10).Major arboviral disease outbreaks have been recorded in Ghana since 1969 (10).A recent outbreak of yellow fever was experienced in Ghana in 2021 within the Savannah, Oti, Bono and Upper West regions.Reports from the Ghana Health Service, as of 4th December 2021, 202 suspected cases had been reported with 85 con rmed cases and 46 mortalities (13).Evidence of the presence and exposure to other arboviral diseases such as dengue fever and chikungunya have also been reported in Ghana (14)(15)(16).These reports show that arboviral pathogens are in circulation in Ghana and require the establishment of effective surveillance and vector control management strategies.
Despite the increasing outbreaks of arboviral diseases and the high densities of the arboviral vectors in Africa, its control is given limited attention (8).The control of arboviral diseases majorly relies on vector control using insecticides coupled with larval source management and case management.Increasing insecticide resistance in Aedes mosquitoes poses a major challenge for vector control strategies.Resistance of the Aedes mosquitoes to insecticides has been reported in some West African countries like Senegal, Burkina Faso and Ghana (9,11,17,18).Aedes mosquitoes populations in Ghana have been found to be resistant to several public health insecticides including pyrethroids, organochlorines and carbamates (9,11,12,19).Target-site mutations such as V410L, V1016I and F1534C have been found in pyrethroid-resistant Aedes mosquitoes from Ghana and other countries (9,(20)(21)(22)(23).
Vector and insecticide susceptibility surveillance is still crucial in reducing the global burden of arboviral infections (24).However, there is a paucity of data on the risk of transmission and insecticide susceptibility status of these arboviral disease vectors in Ghana.The current study sought to determine the risk of arboviral transmission and insecticide susceptibility status of the Aedes mosquitoes in the selected yellow fever outbreak and non-outbreak areas in Ghana.

Characterization of Aedes breeding habitats and abundance of Aedes larvae
Larval sampling was performed in each study site, to characterise the breeding habitats and abundance of immature Aedes mosquitoes.During the larval surveys, the habitat type, its location in a household (indoor or outdoor), and its physical characteristics were recorded.Eight container types were classi ed based on their use and material: car tyres, discarded containers, drinking pots, drums, tanks, and buckets.Discarded containers and drinking pots were 50-100 L capacity containers which included broken jars, bottles, small plastic food containers, tins, plates, cans, cooking pots, drinking troughs and broken pots made of clay, plastic or metal.
Drums were de ned as 100-500L capacity plastic water storage containers.Tanks were 100-500 L capacity water storage containers made of metal or concrete.Buckets included 10-25 L water storage containers made of metal or plastic.In almost all the sites, access to pipe-borne water was a big challenge, therefore, households tend to store water in storage containers, pots, drums etc for long-term use.These containers may favour the breeding of the Aedes vector.Coordinates of all collection points were recorded using a GPSMAP® 60CSx geographical position system (GPS) instrument (Garmin International, Inc., Olathe, Kansas, USA).
All larval samples were transported to the insectary at the Department of Medical Microbiology, University of Ghana, where they were raised to adults under suitable conditions (temperature: 27 ± 2°C, 75 ± 10% relative humidity).The larvae were fed on TetraMin Baby sh food (Tetra Werke, Melle, Germany).Emerged adults were morphologically identi ed using standard taxonomic keys (25).

Determination of Stegomyia Indices
The extent of infestation by Aedes mosquitoes was estimated using the classical Stegomyia indices including BI, Breteau Index, (the number of positive containers per 100 surveyed houses); HI; House Index, (percentage of houses positive for Aedes larvae or pupae); CI; Container Index, (the percentage of containers positive per 100 houses inspected).These larval and pupal indices continue to be the predominant and frequent measures used to evaluate vector prevalence because catching adult mosquitoes is time-consuming and requires access to private land (12).Stegomyia indices are quantitative indicators of the risk of transmission.
Using the WHO criteria, the risk of YF at each site was assessed as follows: In an area where BI, HI, and CI exceeded 50%, 35% and 20% respectively, the risk of Ae aegypti-transmitted Viral Haemorrhagic Fever (VHF) was considered to be high; an area where BI was between 5% and 50%, the density of Ae. aegypti was considered to be su cient to promote an outbreak of VHF disease; an area where BI, HI and CI were less than 5%, 4% and 3% respectively, it was considered to be unlikely for YF transmission to occur (26).

Adult Aedes mosquito sampling
The spatio-temporal distribution of adult Aedes mosquitoes was determined by sampling indoors and outdoors using three methods; Biogent sentinel 2 traps (BG traps), Human landing catches (HLC), and prokopack aspiration (PPK) (John W. Hock Company, Gainesville, U.S.A.) (27).Geographical coordinates of each house sampled were taken.Cross-sectional surveys were undertaken in the dry season (March to May 2022) and in the rainy season (August to October 2022).Sixteen houses were randomly selected for each sampling method at each site (4 houses per day).Biogent sentinel traps were set both indoors (living rooms and bedrooms) and outdoors (open, verandas, or under a shed/tree where people sit to chat) about 5 m from the house) during the times 3:00-7:00 pm.The BG traps were baited with carbon dioxide (CO 2 ) which was produced from a mixture of 17.5 g yeast (Angel Yeast (Egypt) Co. Ltd.), 250 g sugar in 1 litre of water (van Loon et al., 2015).After the 4 hours, mosquitoes trapped were carefully removed, placed in a cooler box containing ice and then transported to the insectary.The HLC method was also used to sample hostseeking adult Aedes mosquitoes.On each day, one trained volunteer was positioned indoors and another outdoors and collected mosquitoes from 3:00-6:00 pm.Collected Aedes were placed in well-labelled paper cups, and transported to the insectary for identi cation and further processing.
Prokopack aspirations were used to sample resting mosquitoes.Sampling for Aedes mosquitoes was done indoors and outdoors.Sampled adult Aedes mosquitoes were knocked down with chloroform and preserved in Eppendorf tubes containing silica gel.

Morphological identi cation of Aedes mosquitoes
All adult Aedes mosquitoes from the adult sampling, and those used for the susceptibility testing (Aedes larvae collected in the study sites and raised to adults), were identi ed morphologically using the taxonomic keys by Huang (25).

Insecticide susceptibility tests
Insecticide susceptibility test was conducted using WHO tubes to determine phenotypic resistance according to WHO criteria (28).Adult female Aedes mosquitoes that were 3-5-day-old were exposed to papers impregnated with permethrin (0.75%), deltamethrin (0.05%), and pirimiphos-methyl (0.25%).Larvae and pupae collected from the larval sampling as well as the Ovitraps that were set were used for the WHO susceptibility testing.Though these doses are not the recommended doses for evaluating the susceptibility of Aedes mosquitoes, they are the most commonly used (11,29).The knockdown time was recorded every 10 min during the 60-minute exposure period.Mortalities were recorded after a 24-hour recovery period.Alive (resistant) and dead (susceptible) mosquitoes were stored in absolute ethanol for later DNA analysis.

Genotyping of kdr mutations in Aedes aegypti populations
A sub-sample of 242 phenotypically pyrethroid resistant and susceptible Aedes mosquitoes were genotyped of kdr mutations, F1534C, V1016I and V410L.Total DNA was extracted from whole mosquitoes using the DNeasy Tissue Kit (Qiagen, In USA).Genotyping of the kdr mutations was done using allele-speci c multiplex PCR according to well-described protocols of Villanueva-Segura et al. (30)

Results
Distribution and abundance of Aedes larval habitats.

Genotypic mutations associated with resistance in Aedes aegypti
A subset of 242 Ae. aegypti obtained from the phenotypic assays were genotyped for the F1534C, V1016I and V410L kdr mutations.About 20-30 mosquito samples were selected from each site per insecticide paper per concentration for the genotypic resistance determination.The F1534C mutation was detected in moderate to high allele frequencies in Ae. aegypti mosquitoes exposed to pyrethroid insecticides.Aedes aegypti from Wenchi that were exposed to deltamethrin, had a signi cantly high allele frequency of F1534C mutation (0.92) compared to mosquitoes from the non-outbreak sites, Pagaza (0.19) and Kpalsogu (0.35) (χ2 = 50.50,df = 3, P < 0.001).For the V1016I mutation, low to moderate allele frequencies (0.23 to 0.54) were observed in Ae. aegypti mosquitoes exposed to both deltamethrin and permethrin except for Larabanga, an outbreak site, where mosquitoes had a high allele frequency of 0.77.However, there was a signi cant difference in the frequency of V1016I mutation in outbreak and nonoutbreak sites (P < 0.05).For the V410L mutation, there was no signi cant differences in the frequency of the mutations between outbreak sites and non-outbreak sites with low allele frequencies ranging from 0.05 to 0.15.The genotypes and allele frequencies of each kdr mutation are shown in Table 4.

Discussion
Due to their capacity to spread a variety of arboviral infections like dengue, chikungunya, Zika, and yellow fever, Aedes mosquitoes pose an increasing hazard to public health on a global scale (31).Ghana is endemic for yellow fever with the most recent outbreak occurring in 2021 (13).
Hence, the need to monitor the densities and insecticide susceptibility status of Aedes mosquitoes in the country.This study sought to determine the risk of arboviral transmission and insecticide resistance status of Aedes mosquitoes in a yellow fever outbreak and non-outbreak areas in Ghana.
From the data obtained, car tyres were the most productive habitats for Aedes mosquitoes seasonally.Moreover, the outbreak area (Larabanga) had a high Stegomyia indices value and hence satis ed the WHO criteria for a high-risk zone for yellow fever transmission.Adult Aedes aegypti was the predominant vector sampled in all the study sites, with high abundance in the dry season and from outdoor collection.The distribution of larval habitat types varied signi cantly between seasons.
Findings from this study showed that car tyres were responsible for over 60% of Aedes immatures over the entire sampling period.Car tyres seems to provide an optimal temperature, humidity and su cient light intensity to ensure larvae development (32).This nding is consistent with studies by Owusu-Asenso et al. (11) in Ghana and Kamgang et al. (33) in the Central African Republic who also found car tyres as the most conducive and productive breeding habitat for Aedes aegypti.
In the current study, results from the larval indices indicate that Larabanga had values that exceeded the WHO threshold and hence is a high-risk zone for arboviral pathogen transmission.This is similar to a previous study by Appawu et al. (10) in Larabanga, where the Stegomyia indices exceeded the threshold and was considered as a high-risk zone for yellow fever transmission.Furthermore, Stegomyia indices values within the WHO criteria reported in Wenchi, Kpalsogu and Pagaza were su cient to promote an outbreak within these sites.These ndings imply that inhabitants within these study sites are at risk of yellow fever infection, Hence, there is a need to employ preventive measures through vaccination and effective vector control strategies.
Seasonal variation in population density is common seasonally for Aedes mosquitoes due to their sensitivity to changes in temperature and rainfall (34).This study found a signi cantly higher abundance of Aedes immature during the rainy season.The development of mosquitoes, their survival and the effective transmission of pathogens are in uenced by humidity, temperature and rainfall (35,36).Su cient humidity and rainfall in uence the breeding sites, increasing vector populations (37).Hence, an increase in breeding sites may explain the observed increase in Aedes immatures in the rainy season.
This nding corroborates with studies in Ghana by Owusu-Asenso et al. (11) and in Kenya by Ndenga et al. (38).Their ndings showed high densities of Aedes immatures in the rainy season.However, this nding was contrary to another study in Ghana where Aedes larvae were found predominantly in the dry season.He reported that during drought conditions, the surge in the storage of water creates more breeding habitats for Aedes mosquitoes, causing an increase in their abundance (10).
It was observed that adult Aedes mosquitoes were predominantly collected during the dry season and in outdoor settings.Warmer temperatures enhance the growth of mosquitoes and viral replication (34).The life-limiting elements of latitude, altitude, temperature, rainfall, humidity, season, habitat, and dispersal have an impact on the distribution and population of Aedes mosquitoes (10).Finding more adult Aedes mosquitoes outside may suggest their exophilic nature as reported in previous studies in Ghana and Kenya (11,39).It is important to note that people spending more time outdoors compared to indoors in uences the biting and feeding behaviour of Aedes mosquitoes (38).Studies in Ghana have also suggested that Aedes mosquitoes often rest outdoors before and after blood feeding (40) and were more abundant from outdoor collection in the dry season (10).
It was observed in this study that the majority of the adult Aedes mosquitoes were Aedes aegypti which is responsible for yellow fever transmission in Ghana and can transmit other arboviral pathogens such as dengue fever virus (12).There have been previous reports of dengue viral infections in children in Ghana ( 14) and exposure to dengue and chikungunya (15,16,41) that show the role of Aedes aegypti in the transmission of multiple arboviral pathogens in Ghana, which cannot be overlooked.
In this study, Aedes mosquitoes across the sites showed resistance to Deltamethrin.This might be due to the indirect impact of the use of insecticides for public health vector control such as the use of Long-Lasting Insecticidal Nets and IRS, as well as pesticide use in agriculture (11,42,43).This nding similar to that reported in a study conducted in Ghana (Abdulai et al., 2023).Whereas Aedes mosquitoes collected from Pagaza and Kpalsogu showed suspected resistance to Permethrin, samples from Larabanga were found to be susceptible to Permethrin.Similarly, pyrethroid resistance has also been reported in Aedes aegypti populations from Ghana and other West African countries ( (11,17,18,29) Aedes mosquitoes in this study were also resistant or possibly resistant to Pirimiphos-methyl.The ndings suggest that resistance to this organophosphate by the Aedes population has increased.
In this study, the F1534C and V1016I kdr mutations were high in frequencies in resistant and susceptible Aedes mosquitoes while V410L kdr mutation showed low frequencies.Although suspected resistance and susceptibility to permethrin were recorded in the Aedes mosquitoes from the same population, there is the risk of resistance developing over time due to the high frequencies of the F1534C and V1016I kdr mutations.Pyrethroid resistance in Aedes aegypti is a worldwide challenge for mosquito control due to its use for insecticide-treated nets and indoor residual spraying (44).
Similarly, other studies in Ghana have found high frequencies of V1016I and F1534C kdr mutations in both and resistant Aedes mosquitoes collected in Ghana (9,21).This suggests that other resistance mechanism such as metabolic resistance may be involved in insecticide resistance of Aedes aegypti populations in Ghana.Hence, further studies are needed to elucidate the mechanisms mediating insecticide resistance in Aedes mosquitoes in Ghana.

Conclusion
Our ndings reveal that Aedes immatures were most abundant in the dry season and car tyres were the most productive habitat type.Aedes aegypti were the predominant species found.The results showed that Larabanga, the epicentre of the yellow fever outbreak is a high-risk zone for arboviral pathogen transmission.In addition, high phenotypic and genotypic resistance was observed in Aedes mosquito populations in Ghana.Surveillance of Aedes mosquito bionomics and insecticide susceptibility in Ghana is crucial to help in the development of arboviral vector control strategies to control and prevent arboviral outbreaks in Ghana.
Mortalities of Aedes mosquitoes exposed to different insecticides in the study sites.

Figure 1 :
Figure 1: Map of Ghana indicating the study areas.

Figure 2 :
Figure 2: Larval habitats encountered during larval sampling.a Drinking pot, b Tank, c Car tyre, g Discarded container, e Drum f Bucket

Figure 6 :
Figure 6: Mortalities of Aedes mosquitoes exposed to different insecticides in the study sites.

Figure 1 Map
Figure 1

Figure 3 Stegomyia Indices per study site Figure 4 Page 14 / 15 Seasonal abundance of adult Aedes mosquitoes Figure 5
Figure 3

Table 2
The seasonal distribution of Aedes immatures across the study sites.

Table 4
Number of genotypes and frequencies of the F1534C, V1016I and V410L mutation in the voltage-gated sodium channel gene of Aedes aegypti mosquitoes.