*2.2. Timeline*

The timeline below, Figure 2, indicates the sequence of activities in this study across generations of *An. gambiae* Muleba-Kis.

**Figure 2.** The timeline for activities, indicating wild mosquito collection, insectary colonization, out-crossing, selection, and strain characterization across *An. gambiae* Muleba-Kis generations.

#### *2.3. Collection of Wild Mosquitoes and Introduction into the Insectary*

Indoor resting blood-fed *Anopheles* were collected in house bedrooms using mouth aspirators. Mosquitoes were transferred in paper cups supplemented with glucose and transported to field insectaries located in Muleba. They were held under ambient relative humidity and temperature conditions in 30 × 30 × 30 mosquito cages containing a petri dish of moistened cotton wool overlaid with damp filter paper for egg laying. After laying, adult *An. gambiae* s.l. were stored individually and subsequently identified by Polymerase Chain Reaction (PCR) [47]. Collections were done over two months and eggs (approximately 500 eggs) were sent to the KCMUCo-PAMVERC Test Facility. Eggs were introduced into plastic bowls (6 L capacity) filled with 4 L of water. Larvae were reared under ambient temperature and relative humidity and fed with cereal for infants (Cerelac®, Nestlé Kenya Limited, Pate, Kenya) mixed with ground sardines at a 2:1 ratio. Adult mosquitoes were reared at 60–90% RH and 20–35 ◦C in cages (30 cm × 30 cm × 30 cm) covered with untreated netting material and provided with glucose solution 10%. To ensure optimal rearing conditions, insectary larval density was restricted to 200–300 per bowl (3 L capacity), water for mosquito rearing was pre-boiled to avoid bacterial infections, and environmental conditions (water and air temperature, relative humidity) were monitored and maintained.

#### *2.4. Crossing for "Insectary Vigor"*

When F1 mosquitoes were five days old, a restrained guinea pig was introduced into the cages of mosquitoes that were starved for one hour prior to blood-feeding. To overcome difficulties of adaptation to insectary conditions, out-crossing between female *An. gambiae* Kisumu and male *An. gambiae* Muleba mosquitoes were conducted. The main difficulties encountered were low blood-feeding, egg-laying, and survival, otherwise known as "insectary vigor." The *An. gambiae* Kisumu strain was obtained in 2008 through BEI Resources, NIAID, NIH: *Anopheles gambiae*, Strain KISUMU1, Eggs, MRA-762, contributed by Vincent Corbel. This strain is originating from Kisumu, Kenya, and was successfully established at our insectary and feeds well on guinea pigs. The Muleba and Kisumu strain pupae were collected separately, and males were separated from females on the first day after emerging. Adult male Muleba and female Kisumu mosquitoes were mixed at a ratio of 50:50 in a mosquito cage. These were reared at 20–35 ◦C, 60–90% RH, and a natural 12:12 h L:D photoperiod, and were provided with a guinea pig for blood-feeding and filter paper medium for egg-laying. This successful outcrossed mosquito was then named "*An. gambiae* s.s. Muleba-Kis strain" and has been reared at the KCMUCo-PAMVERC test facility since 2013.

#### *2.5. Selection to Maintain Pyrethroid Resistance*

In this study, selection was based on the exposure of larval mosquitoes to pyrethroid insecticides, and pyrethroids were chosen due to intensive usage in public health and having the most widespread resistance among mosquito vectors across Africa [48,49]. Artificial selection for pyrethroid resistance was started in the 15th generation. Six bowls each with around 100 larvae of 3rd to 4th instars were used initially, adopting a modified method by Shidrawi [50]. One mL of insecticide solution was added to 1 L of tap water at 27–32 ◦C, stirred for 1 min using a Pasteur pipette, and then left for 10 min to allow evaporation of acetone which was used as a solvent for insecticide solution preparation. Larvae were transferred into the glass bowl with the dissolved insecticide solution, each bowl with around 100 larvae. A small amount of larvae food was added and larvae were left for 24 h in the selection bowl. After 24 h, mortality was estimated. Mortality was estimated in three categories: high mortality, 67–100%; moderate mortality, 34–66%; or low mortality, 0–33%. The initial selection was done using permethrin, and later alphacypermethrin was used for colony selection. The initial permethrin concentration used for the section was 0.1 mg/L and increased to 0.2 mg/L at a time when larvae mortality was in a low category. The initial alphacypermethrin concentration was 0.025 mg/L and it increased to

0.05 mg/L when larvae mortality was in a low category. The larvae were sieved when still alive from the selection bowl, rinsed with 500 mL water (temp 27–32 ◦C) and returned to their original six bowls, and reared, while the dead larvae were removed. The selection was conducted intermittently. The availability of technical grade insecticide to make up the selection solutions and a need for mosquitoes for ongoing laboratory bioassays were the main constraints preventing routine artificial section of the colony.

#### *2.6. Authentication of the Outcrossed An. Gambiae s.s. Muleba-Kis Strain*
