Molecular and entomological surveillance of malaria vectors in urban and rural communities of Benguela Province, Angola

Background Malaria is a major public health problem in Angola, with Anopheles gambiae sensu lato (s.l.) and An. funestus s.l. being the primary vectors. This study aimed to clarify the information gaps concerning local Anopheles mosquito populations. Our objectives were to assess their abundance, geographical dispersion, and blood-feeding patterns. We also investigated their insecticide resistance. Molecular methods were used to identify sibling species, determine the origin of blood meals, measure Plasmodium falciparum infection rates, and detect the presence of knockdown resistance (kdr) mutations. Methods Adult mosquitoes were collected indoors using CDC light traps from nine randomly selected households at two sentinel sites with distinct ecological characteristics. The samples were collected from 1 February to 30 June 2022. Anopheles mosquitoes were morphologically identified and subjected to molecular identification. Unfed Anopheles females were tested for the presence of P. falciparum DNA in head and thorax, and engorged females were screened for the source of the blood meals. Additionally, members of An. gambiae complex were genotyped for the presence of the L1014F and L1014S kdr mutations. Results In total, 2226 adult mosquitoes were collected, including 733 Anopheles females. Molecular identification revealed the presence of Anopheles coluzzii, An. gambiae senso stricto (s.s.), An. arabiensis, and An. funestus s.s. Notably, there was the first record of An. coluzzii/An. gambiae s.s. hybrid and An. vaneedeni in Benguela Province. Plasmodium falciparum infection rates for An. coluzzii at the urban sentinel site and An. funestus s.s. at the rural site were 23.1% and 5.7%, respectively. The L1014F kdr mutation was discovered in both resistant and susceptible An. coluzzii mosquitoes, while the L1014S mutation was detected in An. gambiae s.s. for the first time in Benguela Province. No kdr mutations were found in An. arabiensis. Conclusions This study provides valuable insights into the molecular characteristics of malaria vectors from the province of Benguela, emphasising the need for continuous surveillance of local Anopheles populations regarding the establishment of both kdr mutations for tailoring vector control interventions. Graphical abstract


Background
Malaria remains a major public health challenge, with an estimated 249 million cases and 608,000 deaths worldwide in 2022 [1].In sub-Saharan Africa, the primary vectors responsible for malaria transmission are Anopheles gambiae senso stricto (s.s.), An. coluzzii, An. arabiensis, and An.funestus s.s.[2].In Angola, malaria is endemic and mostly caused by Plasmodium falciparum [3].Additionally, cases caused by Plasmodium malariae have been recorded in the province of Benguela [4].In Angola, malaria transmission is complex with vectors occupying different eco-epidemiological areas.Anopheles gambiae s.s., An. coluzzii, An. arabiensis, and An.funestus are considered the primary vectors in Angola [3,[5][6][7].Insecticide-based vector control strategies, such as the use of insecticide-treated nets (ITN) and indoor residual spraying (IRS), have significantly contributed to the reduction of the malaria burden [8].However, the emergence and spread of insecticide resistance in these main vectors [9,10] are threating the effectiveness of these interventions.Two main types of mechanism have been identified to be involved in insecticide resistance: target-site resistance, which involves mutations in the target proteins of insecticides, and metabolic resistance, which involves increased detoxification of insecticides [11].One wellknown target site resistance is the knockdown resistance (kdr) caused by mutations in the voltage-gated sodium channel gene, which compromises its binding to pyrethroid insecticides [12].The identification of Anopheles species and their susceptibility statute to insecticides are essential for effective planning of vector control strategies.Molecular techniques have improved the accuracy of species identification within the An.gambiae complex and Anopheles funestus group [13,14].Furthermore, molecular methods, such as polymerase chain reaction (PCR) assays, have been developed to detect kdr mutations, enabling the assessment of the frequency and distribution of these resistanceassociated mutations in field populations [15].Despite the crucial role of vector control in reducing malaria transmission, little published evidence on Angola malaria vector abundance, behaviour, and insecticide susceptibility has been published in the past 20 years [16][17][18][19][20].This study aimed to describe the local populations of Anopheles mosquito species in two districts of Benguela, with a focus on characterising the vector abundance, behaviour, and insecticide susceptibility of these populations.Using molecular methods to investigate the presence of knockdown resistance (kdr), P. falciparum infectivity rates, and blood meal sources, the outcomes will aid in bridging knowledge gaps concerning malaria vectors in Benguela Province.

Study areas
The study was conducted from 1 February to 30 June 2022 at two sentinel sites in the province of Benguela.Bela Vista (Fig. 1A, urban site, 12°37′27.8ʺS,13°22′23.4ʺE) is an urban neighbourhood located in the city of Benguela at an average altitude of 20 m.The climate is characterised by a dry season (May to September) and a rainy season that extends from October to April with an average rainfall of 12.9 mm (October) to 33.9 mm (March).The average temperature varies from 30 °C (March/April) to 19 °C (July/August).Cavaco River (Fig. 1C, 12°34′10.0ʺS13°25′11.5ʺE) is located at the urban north border of the city of Benguela.It is used by the local population during the wet and dry season to collect water for different uses.During the dry season, the local population dig wells in the dry riverbed to facilitate water collection.Alto Catumbela in the district of Ganda (Fig. 1B, rural site, 12°56′15.3ʺS14°45′19.1ʺE) is a rural community located 210 km east of Bela Vista at an altitude of 1244 m.The predominant activity in this area is agriculture and cattle breeding.Families who own cattle keep them in corrals or fenced off near their houses.Temperatures range from 31 °C (September) to 10 °C (June/July).The rainy season lasts from October to April, with a monthly rainfall average of 37 mm in April to 130.6 mm in December.In both locations, during field work, we observed that households use ITN, aerosol insecticides, and mosquito-repelling coils to protect themselves against mosquito bites.

Adult sampling methods
Adult mosquitoes were collected indoors from the Bela Vista and Alto Catumbela sentinel sites.At each sentinel site, nine houses were randomly selected at least 50 m apart.The selected houses had cement or mud walls, painted or unpainted, with or without ceilings, and roofs made of sheet metal or traditional materials.CDC light traps (CDC-LT; Model 512; John W. Hock Company, Gainesville, FL, USA) were used overnight to collect mosquitoes for 3 consecutive nights in three different houses each night.The CDC-LTs were installed circa 150 cm above the ground at the foot end of the sleeping area where people were sleeping under a mosquito net.Mosquitoes were captured between 18:00 and 07:00.Non-anopheline species were discarded after counting.In households where permission was granted for the collection of adult mosquitoes indoors, the collection procedure was communicated to the household head, who provided their consent.

Larva collections and WHO susceptibility test
To test the insecticide susceptibility of mosquitoes, the WHO tube test was performed using papers impregnated with alpha-cypermethrin at a diagnostic concentration of 0.05% [21].The tests were performed using F0 generation, 2-5-day-old females collected as larvae in June 2022 from shallow wells in the dry riverbed of Cavaco River (Fig. 1C, 12°34′10.0ʺS13°25′11.5ʺE).Collected larvae were brought to the field insectary in Benguela and reared to adult stage.Before the start of the test, a preliminary morphological identification was carried out on live adult Anopheles mosquitoes.Mosquitoes identified as Anopheles gambiae s.l. were sorted for WHO tube tests.At the end of the test, a confirmatory morphological identification was carried out to ensure that only Anopheles gambiae s.l. were included in the mortality analysis.Any other mosquito species found were excluded from the final count.The population was classified as "resistant" if mortality < 90%, "possible resistant" if mortality < 98%, and "susceptible" if mortality ≥ 98%.After the assay, mosquitoes were individually stored in labelled microtubes containing silica gel for further molecular processing.

Anopheles mosquito species identification
Prior to DNA extraction, female Anopheles mosquitoes were identified to species using morphological keys [22,23].Morphological identification was performed by trained entomologists.Abdomens from captured female Anopheles mosquitoes were categorised as unfed or fed.Samples were stored individually in labelled microtubes containing silica gel for molecular analysis at room temperature.Genomic DNA was extracted according to Collins et al. [24].DNA was used for identification of sibling species, the presence of P. falciparum, blood meal source, and presence of kdr mutations.Molecular sibling species identification was conducted on randomly selected morphological identified members of the An.gambiae s.l. and An.funestus s.l.For An. gambiae s.l., molecular identification was conducted by targeting species-specific polymorphisms at the intergenic spacer (IGS) of the ribosomal DNA [25] and by a PCR assay targeting the SINE 200X6.1 retrotransposon insertion [26].Specimens were identified as An.coluzzii, An. gambiae s.s., An. coluzzii/An.gambiae s.s.hybrids, Anopheles melas, or An.arabiensis if they had coincident species-specific patterns for both markers.For the An.funestus s.l.members, which include An. funestus s.s., Anopheles parensis, An. rivulorum, An. leesoni, An. rivulorum-like, and An.vaneedeni, molecular identification was conducted using the protocol described by Koekemoer et al. [14].All PCR assays contained negative controls (no DNA template) and positive controls, consisting of samples of known specimen molecular identification.Anopheles gambiae s.l. and An.funestus s.l.PCR species-specific identifications giving a negative result and randomly selected specimens of less common species of Anopheles mosquitoes were submitted to Cytochrome Oxidase I (COI) barcode sequencing for species identification, following Folmer et al. [27].Comparison between morphological and molecular identification was performed to assess the accuracy of the morphological identification.All COI sequences have been submitted to NCBI and are available online under GenBank accession numbers OR839824-OR839854.

Identification of the blood meal source
The origin of blood meals in blood-fed Anopheles mosquitoes was performed using a multiplex PCR [28], targeting human, cow, dog, goat, and pig DNA.Positive and negative controls were included in all reactions.

Detection of Plasmodium falciparum DNA in Anopheles mosquitoes
Molecular procedure describe by Demas et al. [29] was used to detect the presence of P. falciparum DNA.Since this technique does not allow to distinguish between infected and infectious mosquitoes, only heads and thoraxes of unfed mosquitoes were used.Preference was given to mosquitoes molecularly identified as An.gambiae s.s., An. coluzzii, An. arabiensis, and An.funestus s.s.

Data analysis
The accuracy of morphological identification was determined as the ratio between molecular species confirmations and the total number of morphological identifications subject to molecular identification by species.Human blood index (HBI) was calculated as the number of mosquitoes fed on humans (including mixed blood meal origins) by the total number of blood-fed anopheline mosquitoes analysed.Infection rate (IR%) was estimated as the proportion of unfed Anopheles females positive for P. falciparum DNA.To assess the association between kdr genotypes and resistance phenotypes, in An. gambiae complex, Fisher's exact test was calculated with contingency tables using GraphPad Prism (version 8.0.1).

Anopheles mosquito species composition
From February to June 2022, 2226 adult mosquitoes were collected using indoor CDC-LT.Among the collected mosquitoes, 748 (33.6%; 733 females and 15 males) were identified as anopheline mosquitoes and 1478 (66.4%; 1148 females and 330 males) were classified as non-anopheline mosquitoes.Out of the total Anopheles females collected as adults, 622 (84.9%) were morphologically identified to species or complex of species, while 111 (15.1%) could not be identified because of the absence of anatomical parts.Of the larvae collected from Cavaco River and reared in an insectary to an adult stage, 60 were identified as An.gambiae s.l.The district of Ganda contributed 89.9% (n = 713) while Benguela District contributed 10.1% (n = 80) of the female Anopheles mosquitoes collected and morphologically identified (Table 1).A subsample of 155 An. gambiae s.l. and 375 An. funestus s.l. were molecularly processed for species identification.The molecular identification of the 155 An. gambiae s.l.indicated the presence of An. arabiensis (n = 26), An. coluzzii (n = 75), An. gambiae s.s.(n = 49), and An.coluzzii/An.gambiae s.s.hybrid (n = 2); three showed no amplification.From the 375 An. funestus s.l.mosquitos, 360 were identified as An.funestus s.s. and two as An.vaneedeni, and 13 specimens showed no amplification (Table 2).From those that did not show amplification on species-specific molecular identification (n = 16), five An.funestus s.l. and two An.gambiae s.l., as well as 34 anophelines of other Anopheles species, were identified by COI barcode analysis (Table 3).
The accuracy of the morphological identification was determined by crossing morphological and molecular identification.This reveled an overall accuracy of 94.5% (533/504).Anopheles pretoriensis, An. coustani s.l., and An.maculipalpis had 100% accuracy, followed by An. gambiae s.l.(98.1%),An. funestus s.l.(97.3%), and An.squamosus (85.7%).Morphological identifications failed on An. obscurus, An. rhodesiensis s.l., An. ruarinus, An. argenteolobatus, and An.tenebrosus (Table 4).Molecular species identification revealed a total of 11 species, including the hybrids (Table 5).Overall, more adult Anopheles mosquitoes were caught in the rainy season (n = 388) than in the dry season (n = 164) (Table 5).Overall, in the dry season, the predominant species collected was An. funestus s.s.followed by An. coluzzii, An. arabiensis.and An.gambiae s.s.(Table 5).In the rainy season, An. funestus s.s. was the predominant species recorded followed by An. gambiae s.s., An. coluzzii, and An.arabiensis.In the rainy season, An. squamosus was also registered in relatively high numbers (n = 11; 2.3%) (Table 5).The two specimens of the hybrid An.coluzzii/An.gambiae s.s. were registered in the dry season, in both Cavaco River and Alto Catumbela collection sites, the former as a larva and the latter as an adult (Table 5).In terms of geographical distribution, there were clear differences, with An. coluzzii being found in Bela Vista and Cavaco River sentinel sites (Table 5).

WHO susceptibility test
In the insecticide susceptibility assays conducted on An. gambiae s.l.from Benguela using 0.05% alpha-cypermethrin, a mortality rate of 57.6% was observed.Notably, only 60 An.gambiae s.l.females were exposed to impregnated filter papers.This sample size was not considered robust according to WHO standards.Nonetheless, these preliminary results suggested that this population may exhibit phenotypic resistance to alpha-cypermethrin.Molecular identification revealed that this population was An. coluzzii (n = 59) and An.coluzzii/An.gambiae s.s.(n = 1) (Table 2).

Origin of Anopheles blood meals
The origin of blood meals of Anopheles mosquitoes collected indoors using CDC-LT showed that humans were the main host (71%) in both sites (Table 6).In Bela Vista, An. coluzzii was the only species screened with a human blood index (HBI) of 0.67.In Alto Catumbela, where four species were tested, the highest HBI was registered for An.funestus s.s.(0.84) followed by An. vaneedeni and An.arabiensis (both with 0.50) and An.gambiae s.s.(0.25).

Plasmodium falciparum infection rate
A total of 371 female anopheline mosquitoes were screened for the presence of P. falciparum DNA (Table 7).Infection rate varied depending on the species and season of collections.No P. falciparum DNA was found in An. arabiensis.In the Bela Vista sentinel site, An. coluzzii was the only species screened with an overall infection rate of 23.1%.Overall, at Alto Catumbela, An. funestus s.s. had a higher P. falciparum infection rate of 5.7% compared to the 2.4% rate in An. gambiae s.s.Despite these differences, statistical analysis revealed no significant differences between the two rates (Fisher's exact test, P > 0.05).When comparing the infection rate within the rainy and dry season in Alto Catumbela for An.funestus group and An.gambiae complex, no significant differences were observed between the two (Fisher's exact test, P > 0.05).

Knockdown resistance mutations in Anopheles gambiae s.l.
In Bela Vista, a total of 71 An. coluzzii mosquitoes were analysed for the presence of L1014F and L1014S mutations; among these, 16 were not exposed to insecticides.The genotypic frequency of L1014F in these mosquitoes showed a dominance of the resistant allele genotype, with a frequency of 0.94.In addition, the allele frequencies of L1014F in susceptible or resistant mosquitoes were similar (Fisher exact test, P > 0.05).When considering both phenotypes and unexposed mosquitoes, the frequency of the mutant allele was 0.71.At Alto Catumbela site, 76 mosquitoes, comprising An. arabiensis and An.gambiae s.s., were analysed.The An. arabiensis group, consisting of 26 mosquitoes, did not exhibit the L1014F mutation, whereas the An.gambiae s.s., with 49 mosquitoes, predominantly exhibited the mutant allele with an allele frequency of 0.90.Allele 1014S was reported for the first time to our knowledge in An. gambiae s.s.from the province of Benguela, only in heterozygosity with the 1014F allele.In total, across both sentinel sites, 148 mosquitoes were analysed.The overall frequency of 1014F mutation was 0.65, while the 1014S mutation was found to be 0.01 (Table 8).

Discussion
This study significantly advances our understanding of the Anopheles spp.populations in urban and rural settings in Benguela Province, highlighting key aspects of malaria transmission.Our research confirms the presence of An. coluzzii, An. gambiae s.s., An. arabiensis, and An.funestus s.s., consistent with prior studies in Angola and sub-Saharan Africa [2,[5][6][7].The relatively low number of An. gambiae complex members collected indoors may suggest a behavioural change due to ITN presence as previously reported in other locations [30,31].We confirm the presence of An. coustani s.l., An. squamosus, An. pretoriensis, and the first report to our knowledge of An. vaneedeni in Alto Catumbela sentinel site.These species were found infected with P. falciparum in other sub-Saharan Africa countries [32][33][34][35].These results highlight the importance of continued investigation into the roles that less studied species play in the transmission of malaria in the province.Anopheles coluzzii/An.gambiae s.s.hybrids were reported from Cavaco River and Alto Catumbela sentinel site during the dry season.The use of molecular methods revealed a significant accuracy in morphological identification done in the field, especially for the An.funestus group (97.3%) and An.gambiae complex (98.1%), the primary malaria vectors in Angola.This emphasises the importance of skilled technical staff and robust surveillance networks in malaria control strategies.In Bela Vista, we observed an overall P. falciparum

Conclusion
In conclusion, our findings provide crucial insights into the malaria vector populations in Angola, with significant implications for public health policy, vital for tailoring vector control strategies, ensuring their continued efficacy.This study not only fills a critical knowledge gap but also lays the groundwork for enhancing malaria entomological surveillance and control efforts in Angola.

Fig. 1
Fig. 1 Mosquito collection sites.A Adult collection site at Bela Vista in Benguela District.B Adult collection sites at Alto Catumbela in Ganda District.C Larva collection site at Cavaco River at Benguela District

Table 1
Results of female Anopheles spp.morphological identification Include mosquitoes collected as larvae and reared to adult stage a Only larval collections.b
a Includes mosquitoes collected as larvae and adults.b Only larval collections

Table 3
COI barcoding analysis of Anopheles mosquito

Table 4
Accuracy of morphological identification a Includes mosquitoes collected as larvae and adults.b Anopheles vaneedeni or An.longipalpis.c Anopheles coustani or An.ziemanni

Table 5
Seasonal and geographical distribution Anopheles mosquito in Benguela Province

Table 6
Origin of the blood meal of Anopheles mosquitoes in Benguela and Alto Catumbela a Number of Anopheles mosquitoes screened for the origin of the blood meal

Table 7
Monthly Plasmodium falciparum infection rates in anopheline mosquitoes in the two adult collection sites: Bela Vista and Alto Catumbela insecticide resistance, such as metabolic resistance, were not yet explored in Angola.Future studies should investigate additional resistance mechanisms to provide a more comprehensive understanding of insecticide resistance across malaria vector populations.Additionally, a deeper investigation into the vectors' feeding behaviours is necessary for a complete understanding of malaria transmission dynamics.

Table 8
Genotype frequencies of L1014F and L1014S in Anopheles spp.from a member of An. gambiae complex from Benguela Province NA: not applicable, mosquitoes caught as adults and not exposed to insecticides