**1. Introduction**

Mosquitoes represent a major threat to human health because of their role in the transmission of vector-borne diseases (VBDs). Over the past century, the incidence of mosquito-borne diseases has increased significantly around the world [1–3].

To deal with this threat, researchers are developing novel techniques for use in integrated vector managemen<sup>t</sup> (IVM) programmes and are focusing on biological, cultural, physical, mechanical, and genetic control methods [4,5]. However, chemical control, such as insecticide use, remains one of the most reliable strategies [6]. Indeed, the use of insecticides in IVM programmes has increased in recent years, reducing human mortality due to VBDs in many countries and thus playing an essential role in efforts to improve public health [7]. Pyrethroids are a key class of insecticides; they are neurotoxins that interfere with nervous system function in arthropods by blocking the closure of sodium channels. As a result, nerve impulses are prolonged, leading to muscle paralysis and, ultimately, death [8]. Worldwide, pyrethroids are the most frequently used insecticide class because they are relatively less toxic to mammals, have a rapid knockdown (KD) effect on the target arthropods, and break down rapidly in the environment due to their high degree of photodegradation [9]. They are widely deployed against agricultural pests, household pests, store-product pests, ectoparasites found on pets and livestock, and vectors of diseases [10].

Biocidal products (BPs) are strictly regulated by governmental authorities. Regulations are based on the physicochemical properties, efficacy, and environmental and human health risks posed by the active substances (ASs) contained in BPs.

Over recent decades, the European Biocidal Product Regulation (BPR) has drastically reduced the number of ASs used in insecticides, primarily as a result of toxicological and environmental concerns and, secondarily, as a result of the high costs associated with justifying the use of existing ASs or registering new ones [11]. In Europe, there are 22 official biocidal product types (PTs). The category PT18 includes the compounds used in insecticides, acaricides, and other arthropod control products that function by means other than repulsion or attraction. The category PT19 includes compounds that control harmful organisms by acting as repellents or attractants, including those that are used to protect human or animal health via spatial treatments and/or application to the skin [12]. Certain compounds, such as pyrethroids, have a dose-dependent effect: depending on the conditions of use, the substance may kill insects (PT18) [13,14] or repel them (PT19). Personal protection products can be found in both categories [13–18]. In Europe, an AS must be registered in both categories to be authorised for both uses. At present, only two ASs have such a dual status: geraniol (CAS number 106-24-1) and *Chrysanthemum cinerariaefolium* extract (CAS number 89997-63-7) [11].

EU efficacy requirements for insecticides used in space treatments stipulate that a formulation/AS dose must kill 90% of exposed insects within 24 h [19], a threshold known as the LD90. Insecticide doses below the LD90 are considered to be ineffective and, therefore, are not authorised. However, there are other issues to consider. First, high levels of mortality require the use of high doses, which conflicts with the constraints imposed by human health risk assessments (HHRAs), whose results are also required for product authorisation.

In turn, a dose is formally defined as sublethal when it induces mortality in less than 50% of exposed insects [20]. While many studies have characterised the effects of lethal pyrethroid doses on different arthropod taxa [21], much remains unknown about how sublethal pyrethroid doses used in space treatments affect mosquito fitness and behaviour or how such doses could be used in IVM programmes [18,22]. However, some studies have revealed that sublethal doses of insecticides could reduce mosquito survival, population sizes [22–24], and biting rates [25,26].

In this study, the effects of prallethrin 94.7% technical grade (CAS number 23031-36-9; PT18), a synthetic Type I pyrethroid, were assessed using two species of laboratory-reared mosquitoes: *Aedes albopictus* and *Culex pipiens*. Both are commonly used in insecticide efficacy tests across the globe. Prallethrin resulted in rapid knockdown (KD) when deployed against household insect pests via indoor space treatments [27]. The work presented here examined the impacts on three variables in particular: (1) mosquito fitness, (2) protection from mosquito bites in humans, and (3) toxicological risks to humans and the environment. In our analyses, we kept in mind the various constraints associated with EU authorisation standards.

#### **2. Materials and Methods**

The study was conducted in the Henkel Ibérica Research and Development (R&D) Insect Control Department from February 2020 to March 2021. Three experiments were performed using 5 treatments: 3 sublethal doses of prallethrin (0.40 ± 0.01 mg/h, 0.80 ± 0.01 mg/h, and 1.60 ± 0.01 mg/h), an untreated control, and a negative control.

The lowest dose, 0.4 mg/h, was used as a starting point for defining the 2 other doses. Preliminary research determined that this dose resulted in mortality rates of less than 50% 24 h after exposure (Moreno et al., unpublished data) under experimental conditions similar to those in this study (prallethrin applied via a spatial treatment in the laboratory using 12- to 14-day-old female *Ae. albopictus* and *Cx. pipiens*). Consequently, in this study, the starting dose was doubled (0.8 mg/h) and tripled (1.6 mg/h) to assess the effects of using higher levels of the AS.

To achieve accurate dosing, an electric diffuser composed of polypropylene was used (voltage = 220 V; frequency = 50 Hz; maximum power input = 10 W). It is manufactured by Henkel (model EB03) and is commercially available within the EU. The diffuser consisted of a refillable bottle containing the insecticide and a wick connected to a heater that induced evaporation. The release rate of the diffuser could be modulated by adjusting the heater temperature via the diffuser's 2 settings. There was a normal setting, which released a minimum quantity of insecticide (mg of formula/h), and a maximum setting, which released twice that minimum quantity. Thus, to obtain a dose of 0.4 mg/h, the normal setting was used with 1.1% prallethrin in the bottle. To obtain a dose of 0.8 mg/h, the maximum setting was used with 1.1% prallethrin in the bottle. To obtain a dose of 1.6 mg/h, the maximum setting was used with 2.2% prallethrin in the bottle. Solvent types were the same in all 3 cases. The negative control used a formulation that exclusively contained the solvents. In the untreated control, mosquitoes were not exposed to prallethrin or the solvent formulation.

When the electric diffusers were not being used in the efficacy tests, they were kept running (24 h/day) in an evaporation room (temperature: 25 ± 2 ◦C) in the department's chemical laboratory.

The quantities (in mg) of the formulations and the prallethrin that evaporated per hour were calculated based on the change in mass over a series of 24-h periods. Evaporation was monitored for a total of 170 h.

The experiments were carried out in a 30-m<sup>3</sup> chamber, as described in Moreno et al. [28,29].

Two mosquito species—*Ae. albopictus* and *Cx. pipiens*—were used. Representatives of *Ae. albopictus* came from a colony at the Entostudio Test Institute (Italy), which Henkel has maintained for the past 8 years. Representatives of *Cx. pipiens* came from an autogenous strain that Henkel has raised at its own facilities for past 14 years; it was originally collected in the field in Barcelona (Spain). Both colonies are known to be susceptible to pyrethroids.

Mosquito-rearing conditions were as follows: a temperature of 25 ± 2 ◦C, a relative humidity of 60 ± 5%, and a photoperiod of 12:12 (L:D). All the experiments were conducted using 12- to 14-day-old mosquitoes. Although it is standard to estimate mortality in bioassays using mosquitoes of 5–10 days in age, older mosquitoes are more appropriate when changes in biting behaviour need to be evaluated. Thus, mosquito age was standardised for the whole study. Prior to testing, the mosquitoes were separated by species but not by sex. They were allowed to copulate but not to lay eggs. To ensure good activity levels during the experiments, the mosquitoes were given water and a 10% sucrose solution ad libitum before and during the research trials.

#### *2.1. Effects of Sublethal Prallethrin Doses on Mosquito Fitness*

The first experiment examined the effects of sublethal prallethrin doses on mosquito fitness and population dynamics. Female and male mosquitoes of both species were subjected to the 5 treatments. In total, 2500 mosquitoes were used: 1250 mosquitoes of each species, of which 625 were females and 625 were males. Each population of 1250 mosquitoes was divided into 10 subgroups of 125 mosquitoes. Five of the subgroups were composed of females and 5 of the subgroups were composed of males. Each subgroup was randomly assigned to 1 of the 5 treatments.

Every day, the chambers were properly cleaned and, before any experiment was begun, the chamber was checked for insecticide contamination. At least 10 mosquitoes were released into the chamber and left there for 30 min. A piece of cotton wool soaked in a 10% sugar solution was provided. Any mortality or KD during this period was noted, and the chamber was considered to be contaminated or in an unsatisfactory state if KD was higher than 10% [30]. A mosquito was considered to be KD if it was lying on its back and was unable to upright itself [31]. If no contamination was detected, the first set of mosquitoes was removed and the experimental set of 125 mosquitoes was released to initiate testing. These latter mosquitoes were given 30 min to acclimate to the chamber and were also provided with a piece of cotton wool soaked in a 10% sugar solution.

After the mosquito acclimatization period, the electric diffuser was run inside the chamber to begin the treatment. The number of mosquitoes that had been KD was counted every 10 min for up to 90 min. At the end of the trial, the mosquitoes were collected using an entomological aspirator and were taken to an insecticide-free room. There, short-term mortality (STM) was assessed at 24 h and 48 h, then long-term mortality (LTM) was assessed once a week until 100% mortality had been reached or 4 weeks had passed, whichever came first. During this period, the mosquitoes were given water and a 10% sucrose solution ad libitum. Additionally, information on locomotor impairment (i.e., loss of legs) was collected. To this end, mosquitoes were observed and classified for 48 h following a given trial. They were placed in the "living" category if they appeared to be morphologically and/or behaviourally unaffected by the treatment (i.e., they were not found lying on their backs and they had all their limbs). They were placed in the "affected" category if they had lost at least 1 leg. They were placed in the "dead" category if they were lying on their backs and failed to react to any external stimuli [32].

In addition to KD, STM, LTM, and locomotor impairment, fertility, egg laying, the ratio of females to males that emerged, and F1 population size were measured. The exact procedures differed slightly between *Cx. pipiens* and *Ae. albopictus*, as described below.


The larval rearing procedure was the same for both species. The eggs were placed in 6-L plastic trays, which were filled with 5 L of water and then labelled by treatment. The larvae developed in the trays under temperature-controlled conditions (25 ◦C) and were fed rat food (Nanta S.A). Larval density per tray (i.e., 100–120 larvae per litre) was carefully maintained to limit the risk of cannibalism. The water used for larva rearing was not treated with any chemical substances (i.e., anti-algal compounds). The trays were checked every day and additional food was added as needed. Upon reaching the pupal stage, individuals were transferred to the adult emergence containers.

The number of eggs laid over the course of the 4-week post-treatment period was assessed for *Ae. albopictus*, but not for *Cx. pipiens*. In the latter species, eggs are laid in groups (i.e., in egg rafts), making them difficult to count unless separated. For both species, the number of larvae that reached the third/fourth instar and the percentage of females and males that emerged were determined. The ratio of third/fourth instar larvae to females available for egg laying was also calculated.

#### *2.2. Effect of Sublethal Prallethrin Doses on Mosquito Biting Behaviour*

The second experiment examined the effect of sublethal doses on mosquito biting behaviour and, consequently, on host vulnerability. More specifically, it used human volunteers to determine the length of prallethrin exposure that would result in 100% protection.

Six study participants (2 men, 4 women) took part in each trial. They had undergone training to learn how to accurately count mosquito landings. Prior to testing, the skin to be exposed was washed with unscented soap, rinsed with water, rinsed with 70% ethanol or isopropyl alcohol, and then dried with an uncontaminated towel. To ensure that EU guidelines were respected, participants were asked to avoid the use of nicotine, alcohol, fragrances (e.g., perfumes, body lotions, soap), and repellents for 12 h prior to and during testing [19].

Between exposure periods, study participants remained in air-conditioned rooms and kept their activity levels low.

The trials were conducted using only non-blood-fed female *Ae. albopictus*, since the autogenous *Cx. pipiens* strain shows limited interest in feeding on humans.

To ensure good activity levels during the experiment, the mosquitoes were given water and a 10% sucrose solution ad libitum until the trial started.

As in Experiment 1, a preliminary procedure was used to check for insecticide contamination in the chamber. Once the chamber was confirmed to be clean, a pre-treatment trial took place. A total of 20 female mosquitoes were introduced into the chamber [28] and were given 30 min to acclimate. After this period, a study participant entered the chamber with the lower part of their legs exposed; the rest of their body was protected by a light beekeeper's suit. They also wore gloves and white hospital booties [28] (Figure 1). The person remained in the chamber for 3 min [28]. During this time, the number of mosquitoes landing on their exposed skin was recorded. This figure served as a baseline for estimating percent protection following the treatment.

**Figure 1.** (**a**) The 30-m<sup>3</sup> testing chamber at Henkel's R&D Laboratory. (**b**) Participant wearing a protective suit while inside the chamber.

Percent protection expressed the relative reduction in landings/instances of probing attributable to the treatment for each participant [28]. It was calculated as follows:

$$\text{\textquotedblleft protection} = (\text{C} - \text{T}) \times 100 / \text{\textquotedblright} \text{\textquotedblleft} \tag{1}$$

where C = number of landings/instances of probing during the pre-treatment trial and T = number of landings/instances of probing during the treatment trial.

Immediately after the pre-treatment trial, the treatment trial began. First, the electric diffuser was switched on inside the empty chamber. After the diffuser had been running for 5 min, the person who took part in the pre-treatment trial again entered the chamber. They remained inside for 3 min, and the number of mosquitoes landing on their exposed skin was recorded. They then left the chamber. This procedure was repeated 10 min and 15 min after trial initiation.

Each participant was exposed once to each of the 3 prallethrin treatments and the 2 controls.

#### *2.3. Assessments of Human and Environmental Health Risks*

Toxicological risks were assessed in 2 ways: by estimating human health risks using HHRA models and by estimating environmental health risks.

HHRA models were performed for 2 populations: adults and children 2–3 years old. This work was carried out using ConsExpo Web (v. 1.0.7; [33]), a tool designed by the Dutch National Institute for Public Health and the Environment (RIVM). In ConsExpo Web, certain parameters can be set to a chosen value, while others are fixed.

Because an electric diffuser was used in the experiments, only inhalation exposure was considered. However, it is assumed that some of the AS would end up on the floor, where children 2–3 years old might be crawling, so dermal exposure in children was also considered. It was assumed that there was no oral exposure. Thus, the following ConsExpo models were used: "Inhalation exposure: exposure to spray—spray" and "Dermal exposure: direct contact with product—rubbing off".

Within the inhalation exposure model, the inhalation rate was chosen based on Recommendation 14 of the Biocidal Product Committee (BPC) Ad Hoc Working Group on Human Exposure, which describes the default values to use when assessing human exposure to BPs [34]. In this context, here are the key values that were chosen: first, the exposure duration was 24 h per day (a worst-case scenario). Second, it was assumed that night-time respiration in the bedroom was taking place during all those hours (also a worst-case scenario). The volume of that bedroom, 16 m3, was one of the values fixed by ConsExpo and was considered to represent ye<sup>t</sup> another worst-case scenario. To determine the exposure duration that would be considered safe for both adults and children, the 3 experimental doses were examined: 0.4, 0.8, and 1.6 mg/h (Table 1).

Within the dermal exposure model, the dislodgeable amount is the quantity of product applied on a surface area that may potentially be wiped off (per unit of surface area) and that thus may be taken up via contact between surfaces and the human skin. A worst-case scenario was assumed: 10% of the applied AS would end up on the floor, and 10% of that amount would be dislodgeable (Table 2).


**Table 1.** Summary of parameters for the ConsExpo model "Inhalation exposure: exposure to spray— spray".

Chosen and fixed parameter values for the ConsExpo model [33].

**Table 2.** Summary of parameters for the ConsExpo model "Dermal exposure: direct contact with product—rubbing off".


AS, active substance. 1 Chosen and fixed parameter values for the ConsExpo model [33].

To assess risks to environmental health, the following assumptions were made: continuous release (24 h/day) of a vapourised liquid containing prallethrin as its AS and the presence of 2 electric diffusers per household, as per the recommendations in the Technical Agreements for Biocides [35].

The European Chemical Agency (ECHA) Emission Scenario Document (ESD) PT18 spreadsheet (regarding indoor diffusers) was filled out in accordance with the instructions contained in the Organisation for Economic Co-operation and Development (OECD) ESD No. 18 [36]. The results were used to estimate potential product presence in wastewater following treatment and cleaning. Exposure values were calculated using the European Union System for the Evaluation of Substances (EUSES) (software v. 2.2.0).

Any additional risks resulting from metabolites were included in the risk assessment.

For each environmental compartment facing exposure, risk was characterised using the ratio of predicted environmental concentrations (PECs) to predicted no-effect concentrations (PNECs). Of greatest concern was the PEC/PNEC ratio for soils.
