Investigation on key aspects of mating biology in the mosquito Aedes koreicus

Aedes koreicus Edwards, 1917 (Hulecoetomyia koreica) is a mosquito (Diptera: Culicidae) from Northeast Asia with a rapidly expanding presence outside its original native range. Over the years, the species has been discovered in several new countries, either spreading after first introduction or remaining localised to limited areas. Notably, recent studies have demonstrated the ability of the species to transmit zoonotic parasites and viruses both in the field and in laboratory settings. Combined with its invasive potential, the possible role of Ae. koreicus in pathogen transmission highlights the public health risks resulting from its invasion. In this study, we used a recently established population from Italy to investigate aspects of biology that influence reproductive success in Ae. koreicus: autogeny, mating behaviour, mating disruption by the sympatric invasive species Aedes albopictus Skuse, 1894, and the presence of the endosymbiont Wolbachia pipientis Hertig, 1936. Our laboratory population did not exhibit autogenic behaviour and required a bloodmeal to complete its ovarian cycle. When we exposed Ae. koreicus females to males of Ae. albopictus, we observed repeated attempts at insemination and an aggressive, disruptive mating behaviour initiated by male Ae. albopictus. Despite this, no sperm was identified in Ae. koreicus spermathecae. Wolbachia, an endosymbiotic bacterium capable of influencing mosquito reproductive behaviour, was not detected in this Ae. koreicus population and, therefore, had no effect on Ae. koreicus reproduction.


INTRODUCTION
After its first detection in Belgium in 2008 (Versteirt et al., 2012), the mosquito Aedes koreicus (Hulecoetomyia koreica), commonly known as the invasive Korean bush mosquito, has invaded and established in several states in Europe and European neighbouring countries (ECDC, 2022).In some areas, such as Italy, the species is currently expanding its distribution (Arnoldi et al., 2022), but in others, such as Germany, it has shown a relatively low tendency to spread despite suspected repeated introductions (Kurucz et al., 2022).
While the role of Ae. koreicus in arthropod-borne diseases transmission is still largely unclear, the species is known to vector dog heartworm Dirofilaria immitis Leidy, 1856 (Filarioidea: Onchocercidae) under laboratory conditions (Montarsi, Ciocchetta, et al., 2015), a finding later supported by field evidence of filarial DNA in Ae. koreicus sampled near the city of Pécs (Baranya County) in Hungary (Kurucz et al., 2018).The species may also have a role as an intermediate host for Brugia malayi S.L. Brug, 1927 (Filarioidea: Onchocercidae) to infect humans (KCDC, 2007).The potential of A. koreicus to transmit chikungunya virus was demonstrated for the first time under laboratory conditions by Ciocchetta et al. (2018).This study showed how virus transmission was temperature-dependent, and results were further confirmed by Jansen et al. (2021).The same study reported a low vector competence for Zika virus and no transmission of West Nile virus.Japanese encephalitis virus was not detected in Ae. koreicus collected in Korea during recent monitoring activities; however, its possible role as a vector could not be completely excluded (Jegal et al., 2020).
In some hematophagous arthropods, such as mosquitoes, completion of an ovarian cycle and the production of viable offspring can occur in the absence of a bloodmeal in a process called autogeny, most likely as a survival strategy when hosts are rare (Lucius et al., 2017).Autogeny is hypothesised to allow the persistence of a population when the presence of vertebrate hosts is low or to allow for rapid growth of a mosquito population at the start of a season (Reisen & Milby, 1987).This allows mosquitoes to persist in uncertain environments and rapidly exploit optimal conditions; however, the number of eggs laid might vary considerably compared with eggs laid after a bloodmeal (O'Meara & Edman, 1975).Furthermore, this behaviour may delay contact with infected hosts and could, therefore, affect transmission of human pathogenic viruses by mosquito vectors early in the season.Autogeny may be facultative or obligate depending on the species and environmental conditions (O'Meara & Edman, 1975).
The autogeny phenotype has been demonstrated both in the Culicinae and Anophelinae mosquitoes (Clements, 2013).Within the Culicinae subfamily, limited levels of autogeny have been observed in numerous species of Aedes mosquitoes, including some of the main mosquito threats of this century, Aedes albopictus and Aedes aegypti Linnaeus in Hasselquist, 1762 (Aardema & Zimmerman, 2021;Gulia-Nuss et al., 2015).An essential component of autogeny is the female mating status (evidence that sperm transfer occurred): egg development in certain mosquito species does not initiate unless mating occurs, and male accessory gland products can play a central role for oogenesis (O'Meara & Evans, 1976).The ability to identify sperm in the Ae.koreicus female reproductive tract (mating status) is necessary to identify whether the absence of autogeny is simply the result of non-mated females.It is also fundamental in evaluating mating behaviour, reproductive success and the subsequent spread of invasive species in a new territory.
The establishment of an exotic species may be hampered by the disruption of conspecific mating by the aggressive mating behaviour of males of different species (Tripet et al., 2011) and by interspecific cross-insemination (satyrization, a form of sterility caused by interspecific mating) (Bargielowski & Lounibos, 2016).For example, the transfer of Ae. albopictus male accessory gland product to Ae. aegypti females causes them to become refractory to further mating (including with conspecific males) (Tripet et al., 2011).Although Ae. albopictus males are particularly efficient in satyrizing Ae. aegypti females, similar interactions have been noted between Ae. albopictus and other Aedes species such as Aedes polynesiensis Marks, 1951 and members of the Aedes scutellaris Walker, 1859 group (Ali & Rozeboom, 1971;Gubler, 1970).
Additionally, mosquito reproductive behaviour can be influenced by the presence of the endosymbiotic bacteria Wolbachia pipientis.
Wolbachia are small (0.5-1 μm), intracellular, α-proteobacteria known to infect the reproductive organs of 40%-60% of insect species (Weinert et al., 2015).They can affect host reproduction by increasing the reproductive success of infected females, thus enhancing the bacteria's maternal transmission and changing male sperm structure such that only mating with a male infected by the same bacterial strain will lead to progeny (a mechanism called cytoplasmic incompatibility) (Werren et al., 2008).In some cases, Wolbachia can induce parthenogenesis (Stouthamer et al., 1999), and influence fecundity (Alexandrov et al., 2007) and oogenesis (Dedeine et al., 2003).
Our aim here was to provide the basis for further studies on the reproductive behaviour of Ae. koreicus and its potential to become established when introduced in new territories.Even though Ae. koreicus was first detected in Europe more than 14 years ago, its mating biology remains largely unknown.Reproductive success plays a fundamental role in mosquito establishment and population growth (Takken et al., 2006), and an assessment of the reproductive biology of Ae. koreicus could assist in determining its invasive potential.In this study, we investigated several important aspects that may influence the mating biology and reproductive success of the Korean bush mosquito in Italy, such as autogeny, mating behaviour and competitive mating with a sympatric invasive mosquito species (Ae.albopictus).We also screened the mosquito population used to derive our colony for the presence of the endosymbiont Wolbachia pipientis.

Determination of autogeny in Ae. koreicus
Ae. koreicus adult mosquitoes were obtained from a colony maintained at the QIMR Berghofer Medical Research Institute (QIMRB) and derived from an Ae.koreicus population established in north-eastern Italy, Europe (Ciocchetta et al., 2017).Colony eggs laid on Masonite ® sticks were hatched in rainwater to obtain larvae.Due to the low hatching rate of this species (Ciocchetta et al., 2017), larvae were obtained from eggs pooled in order to produce sufficient adults for experimentation.Pupae developed from larvae after 9 days and were sexed using the method previously described (Ciocchetta et al., 2017).To generate three experimental replicates, male and female pupae were placed together in three different cages (BugDorm ® Insect Rearing Cage, 30 Â 30 Â 30 cm) at approximate 1:1 sex ratios (cage 1: 161 males/163 females, cage 2: 161 males/174 females, cage 3: 161 males/170 females).Emerging adult ratios deviated from 1:1 because of differential survival in manipulated pupae (see the Results section).The number of adults per square centimetre of vertical resting surface in a cage was kept low (0.09 mosquitoes/cm 2 ) to avoid overcrowding, which can affect fitness (Montarsi, Ciocchetta, et al., 2015).
The cages of adults were maintained in environmental chambers (Panasonic, Osaka, Japan), as described previously (Ciocchetta et al., 2017).A 10% w/v sucrose solution was provided ad libitum, and each cage was equipped with one egg collection tray (© 2014 Genfac Plastics Pty Ltd, 18.3 Â 15.2 Â 6.5 cm) with rainwater and Masonite ® sticks as oviposition substrates (Figure 1).The position of the cages, stacked on top of each other, was rotated twice per week to minimise positional bias.The number of emerging adults was counted, and cages were checked daily for eggs.After 3 weeks of caging, to allow for the observation of any oviposition activity, one of the three cages was randomly chosen (cage 2) to proceed to blood-feeding on human volunteers (QIMRB Human Research Ethics Committee approval HREC361) in order to confirm the successful completion of the gonotrophic cycle in the presence of a bloodmeal.
The percentage of fed mosquitoes was recorded.Two weeks after blood-feeding (and 7 days from the start of oviposition), eggs were collected, counted and stored in an anti-leak plastic bag.Additionally, five female mosquitoes from the blood-fed cage and 10 female mosquitoes from the remaining two cages were killed (using CO 2 ), and their ovaries were dissected in a drop of phosphate-buffered saline (PBS) on a glass slide at a magnification of 10Â in order to identify mature follicles (stage IVb and V) (Hugo et al., 2003).Differential survival between males and females and male survival between the cages at the time of these observations were tested with a set of chi-square tests using the GraphPad Prism Software (GraphPad Prism version 9.5.1, www.graphpad.com).
The viability of a subsample of eggs collected from cage 2 (n = 1189) was measured after 14 days of storage (Ciocchetta et al., 2017) to verify the successful completion of the gonotrophic cycle in that cage.Observation of Masonite ® sticks for the presence of eggs in the non-blood-fed cages continued until all adult mosquitoes had died and the absence of autogeny was confirmed.

Conspecific Ae. koreicus mating behaviour
Ae. koreicus pupae were derived from mosquito eggs laid on Masonite ® sticks from the QIMRB colony and sexed according to Ciocchetta et al. (2017).A total of 190 males and 240 females were separated into two different BugDorm ® cages placed in environmental chambers for emergence, at the previously described colony rearing temperature and relative humidity (Ciocchetta et al., 2017).
Preliminary observations demonstrated that Ae. koreicus mosquitoes mate under conditions of scarce illumination (S.Ciocchetta, personal observation).As a result, the light/dark cycle was reversed so that mosquito behaviour could be observed under crepuscular and dark conditions.The observation cage was a modified BugDorm ® one with transparent plexiglass used on one side of the cage instead of mesh.
Male mosquitoes require a sufficient period of time for genitalia and sperm development before mating (Oliva et al., 2014), whereas females are often receptive as soon as they emerge (Takken et al., 2006).As a result, 6-7-day-old virgin males and 2-3-day-old virgin females were caged together, and their behaviour was recorded.
At 12-13 h intervals, 25 females were aspirated from the experiment cage, anaesthetised with CO 2 and dissected in a drop of PBS on a glass slide at 10Â magnification.A cover slip was used to rupture the spermathecae and allow for sperm visualisation at an increased magnification of 40Â.Prior to the behavioural observation of mating activity, two samples of female mosquitoes were taken from the observation cage and dissected to determine whether mating was occurring unobserved.After mating was observed, and after sufficient time had passed to allow sperm to reach the spermathecae, another batch of mosquitoes were dissected to confirm that a proportion had mated.
Preliminary observations of Ae. albopictus and Ae.koreicus mating disruption Ae. koreicus larvae were reared as previously described (Ciocchetta et al., 2017).Ae. albopictus larvae (from a colony established at QIMR from eggs collected on Hammond Island, Torres Strait, Australia, in May 2014) were similarly reared, but at a temperature of 27 ± 1 C.
The colonies of both species were synchronised to pupate at the same time.Pupae were individually placed in Falcon ® tubes containing 5-10 mL of rainwater to allow the collection of emerging virgin males or females.3-4 days old Ae.albopictus males (n = 27) ready for copula (Oliva et al., 2014) and 2-3 days old virgin Ae. koreicus females (n = 22) were introduced in a BugDorm ® cage containing a solution of 10% w/v sucrose.The interaction between the two mosquito species was recorded utilising a GoPro ® Hero 3 camera.After 5 days, all female mosquitoes were anaesthetised with CO 2 , and the spermathecae were dissected in a drop of PBS, crushed under a cover slip and scanned at 40Â magnification for the presence of sperm.

Determination of autogeny in Ae. koreicus
The proportion of male:female totals was 123:134, 103:138 in the two non-blood-fed cages (cages 1 and 3) and 116:146 in the 3rd blood-fed cage (cage 2).No eggs were observed in the oviposition trays of the three cages for 21 days after co-caging.After this period, a volunteer fed the mosquitoes in the cage designated to be bloodfed (97.2% fed, n = 109), and oviposition on the Masonite ® sticks in that cage occurred 7 days post-feeding.Mature follicles were observed in all five mosquito females dissected from that cage.No eggs were observed on Masonite ® sticks in cages 1 and 3, and no mature follicles were found in the female mosquitoes dissected from those cages.The percentage of male and female mosquitoes still alive at the time of these observations is reported in Table 1.Under our conditions, 60%-70% of females remained alive after 35 days.
Male longevity is always poorer than female longevity (Bellini et al., 2014), as males only need to survive long enough to mate.This differential was seen in our caged experiments, although there was significant variation in survival between the cages (Table 1).A total of 4925 eggs were counted under the stereoscope from the Masonite ® sticks collected from the blood-fed cage; the average eggs/ female = 50.25 was consistent with a previously reported fecundity index (Ciocchetta et al., 2017).

Conspecific Ae. koreicus mating behaviour
No sperm was observed in spermathecae from female mosquitoes dissected 12 ± 0.5 and 25 ± 0.5 h after co-caging.Mating activity was observed after 25.5 h showing Ae. koreicus males and females in the act of copula and documented with a smartphone device (Oppo F1 Android smartphone; Supplemental File 1).Evidence of motile sperm in Ae. koreicus female spermathecae (Figure 2) was found in 28% of females (n = 25) sampled 31 h after co-caging with males (females were sampled approximately 5 h after evidence of mating activity in the cage to allow the sperm a sufficient period to reach the spermathecae) (Oliva et al., 2014).2021) reported an Ae.albopictus male wing length of approximately 2 mm when the species was reared at a temperature close to our experiment (28 ± 1 C).When reared at the same rearing conditions, these two species maintained their differences in dimensions, with Ae. koreicus bigger in size when compared with Ae. albopictus (Baldacchino et al., 2017), supporting our observations.
Wolbachia presence in field-collected Ae. koreicus No Wolbachia was identified in the Ae.koreicus field samples.The DNA extraction was validated by running a PCR analysis using RpS17 housekeeping gene primers for mosquito DNA, and Wolbachia was detected by the wsp and 16S primers in all positive controls.

DISCUSSION
Defined as the ability to produce offspring in the absence of a bloodmeal, autogeny can influence the vector potential of a mosquito by affecting the abundance or persistence of vectors, even in the absence of immediate hosts (Reisen & Milby, 1987).Conversely, autogeny may limit contact with hosts and reduce transmission risks (Reisen & Milby, 1987).Our results suggest that Ae. koreicus mosquitoes do not display this phenotype under the conditions of our experiment.There was no oviposition when mosquitoes were deprived of a blood source.In early studies with the mosquito Aedes taeniorhynchus Wiedemann, 1821, O'Meara and Evans (1976) showed that mating may increase the levels of autogeny and that the expression of autogeny is correlated with the environmental conditions in which the larval stages develop and the geographical origin of the population (O'Meara, 1979).In Ae. taeniorhynchus, mating was necessary only when larvae were exposed to conditions unfavourable to their development and was otherwise not required for the production of viable eggs (O'Meara, 1979).The observation of Ae. koreicus mating behaviour and the detection of sperm in Ae. koreicus spermathecae confirmed that the absence of autogeny was not due to a lack of mosquito mating.Moreover, autogenic populations of Aedes japonicus Theobald, 1901, a species phylogenetically close to Ae. koreicus, have never been reported in the literature.We hypothesised that Ae. koreicus may be an anautogenous mosquito species; however, although autogeny was not present in the studied colony, the phenotype could still be present in different Ae.koreicus populations, as previously found for instance in Ae. albopictus (Mori et al., 2008).
The delay of 25.5 h being observed before mosquito mating could be due to different factors.Although adult female mosquitoes are ready to be inseminated once they emerge, male antennae and genitalia at the moment of imaginal stage emergence are not in the correct morphological conformation to allow copula.Physical changes must occur for the males to become sexually active (Oliva et al., 2014).
These changes include the erection of fibrillar hairs in the antennae and the permanent 180 rotation of terminalia part of the genitalia to correctly orient the male genital structure for mating.In particular, the time required for this rotation varies among mosquito species and can take up to 4 days, for example, as reported in the species Aedes provocans Walker, 1848 (Smith & Gadawski, 1994).The time of Ae. koreicus male genitalia rotation is not known, which justifies the choice to cage females with 6-7 days old virgin males.Moreover, mating may be encouraged by behaviours displayed in the wild, such as swarming (Cabrera & Jaffe, 2007), that are challenging to create in a laboratory colony.Further observations clarifying the time of genitalia rotation for Ae.koreicus males and the introduction of bigger cages to facilitate swarming could assist in better understanding the mating behaviour of the species.
In this preliminary exploration of Ae. koreicus and Ae.albopictus mating interactions, Ae. albopictus males showed repeated and aggressive mating attempts towards Ae.koreicus females but were unable to transfer sperm to Ae. koreicus.The different sizes of the two species might be one explanation of the outcome of this experiment, with the wing length for females of Ae. koreicus being reported over 3 mm and Ae.albopictus male wing length approximately 2 mm (Baldacchino et al., 2017;Ciocchetta et al., 2017;Pudar et al., 2021).Yet, the lack of sperm does not necessarily exclude a satyrization effect produced by Ae. albopictus males, because the transfer of male accessory gland products (responsible for the satyrization effect) may occur even in the absence of sperm in the spermathecae, as demonstrated by Carrasquilla and Lounibos (Carrasquilla & Lounibos, 2015).Although satyrization between these two species seems unlikely, we speculate  (Alfano et al., 2019;Rosso et al., 2018).It should be noted that in our preliminary investigation, we tested a small sample (n = 21) of Ae. koreicus females; nonetheless, our results suggest that even if present in the mosquito population initially established in Italy, the prevalence of Wolbachia was low.Interestingly, a more recent study (Damiani et al., 2022) Preliminary observations on Ae. albopictus and Ae.koreicus mating disruption Despite repeated interactions between Ae. koreicus females and Ae.albopictus males (Supplemental File 2), no sperm was detected in the 22 individuals dissected (Figure 3).Differences in the size of the T A B L E 1 Percentage of male and female mosquitoes still alive 35 days after co-caging Differential male and female survival χ 2 , p value 72.3, p < 0.0001 13.1, p = 0.0003 49.5, p < 0.0001 Male survival between cages χ 2 , p value Cage 1 and 2; 39.6, p < 0.0001 Cage 2 and 3; 9.1, p = 0.003 Cage 1 and 3; 10.9, p = 0.001 F I G U R E 2 Aedes koreicus sperm (long and slender filaments, (a) visible after spermathecae (b) rupture [microscope magnification 40Â]).mosquitoes (Ae.koreicus females were visibly bigger than Ae.albopictus males; Figure 4) may have been a possible cause for the failure in interspecific insemination.Although the specific size of each individual was not measured, Ae. koreicus female wing length for individuals reared according to previous work (Ciocchetta et al., 2017) has been reported to be over 3 mm; in their work, Pudar et al. (

F
I G U R E 3 No evidence of Aedes albopictus sperm in Aedes koreicus spermathecae (a) before and (b) after rupture (microscope magnification 10Â).F I G U R E 4 Difference in size between Aedes koreicus female (left) and Aedes albopictus male (right).
that the continuous aggressive mating attempts shown by Ae. albopictus males towards Ae.koreicus females could prevent less aggressive Ae. koreicus males from mating and interrupt female behaviours such as feeding and oviposition.This might lead to a decrease in Ae. koreicus numbers in the field.Further studies evaluating the potential for satyrization of males of both species are needed to fully characterise interactions between these sympatric species.Furthermore, since mosquito populations from different geographic regions can show differences in behaviour, additional experiments are needed to test mating interactions and satyrization between Ae. albopictus and Ae.koreicus from diverse geographical origins.Samples tested for Wolbachia were collected during the early stages of Ae. koreicus invasion in Italy.Wolbachia was not detected in Ae. koreicus from the first established field population in Belluno from which the studied colony was derived.Therefore, the bacterial endosymbiont is unlikely to have affected Ae. koreicus reproductive behaviour in the initial establishment of this mosquito in Italy.The absence of Wolbachia in Ae. koreicus has been confirmed by subsequent studies in immature and adult stages sampled in the Province of Trento in 2015 and 2017 on samples collected between 2019 and 2020 in several villages in North-East Italy detected Wolbachia in two of the 85 females examined.Further research with larger sample sizes could help establish whether Wolbachia is present at a very low prevalence in the Ae.koreicus population established in Italy or whether the recent discovery is due to introgression (Bargielowski et al., 2015) from other mosquito species carrying the endosymbiont (Wolbachia strains isolated in Ae. koreicus are closely related to the wAlbB strain, one of the two native strains of Ae. albopictus; McMeniman et al., 2009).In the Province of Trento for instance, Ae. albopictus had 2.5% prevalence of Wolbachia (Rosso et al., 2018).Considering the importance played by reproductive success in ensuring the establishment and growth of invasive mosquito populations during the colonisation of new territories, our preliminary results are aimed at informing further studies to assist in determining the invasive potential of Ae. koreicus and the public health risk posed in areas of recent introduction.