Fitness cost of insecticide resistance on the life-traits of a Anopheles coluzzii population from the city of Yaoundé, Cameroon

Background: Pyrethroid resistance is rapidly expanding in An. gambiae s.l. populations across Sub-Saharan Africa. Yet there is still not enough information on the fitness cost of insecticide resistance . In the present study, the fitness cost of insecticide resistance on Anopheles coluzzii population from the city of Yaoundé was investigated. Methods: A resistant An. coluzzii colony was established from field collected mosquitoes resistant to both DDT and pyrethroid and selected for 12 generations with deltamethrin 0.05%. The Ngousso laboratory susceptible strain was used as control. A total of 100 females of each strain were blood fed and allowed for individual eggs laying, and then different life traits parameters such as fecundity, fertility, larval development time, emergence rate and longevity were measured. The TaqMan assay was used to screen for the presence of the L1014F and L1014S kdr mutations. Results:  Field collected mosquitoes from the F0 generation had a mortality rate of 2.05% for DDT, 34.16% for permethrin and 50.23% for deltamethrin. The mortality rate of the F12 generation was 30.48% for deltamethrin, 1.25% for permethrin  and 0% for DDT. The number of eggs laid per female was lower in the resistant colony compared to the susceptible (p <0.0001). Insecticide resistant larvae were found Open Peer Review


Introduction
Malaria prevention mainly relies on the use of vector control measures with indoor residual spraying (IRS) and long-lasting insecticidal nets (LLINs) as the core interventions 1 . Four insecticide families, organophosphates, organochlorines, carbamates and pyrethroids, are used in public health. The massive use of insecticides particularly pyrethroids over the last decades in vector control and in agriculture resulted in rapid expansion of insecticide resistance, which now affects almost all insecticides 2 . Several mechanisms, including metabolic detoxification, target site mutations and cuticular genes are responsible for insecticide resistance 3,4 . Most common mechanisms associated with insecticide resistance in Anopheles gambiae s.l. include target-site resistance, notably knockdown resistance (kdr mutation with the 1014F and 1014S alleles responsible for resistance to DDT and pyrethroids) and the acetylcholinesterase (Ace-1) G119S mutation responsible for resistance to organophosphates and carbamates 4-6 . Metabolic resistance is another major resistance mechanism, occurring through the upregulation of several detoxification genes from three main families, the esterases, cytochrome P450 monooxygenases, and glutathione S-transferases responsible for resistance to different insecticide families and pollutants 3 . There is a growing concern about the negative impact that insecticide resistance could have on malaria control. It increases the survival rate of mosquitoes exposed to insecticides in treated areas which could potentially lead to greater population size, increase in mosquito burden and diseases transmission 7,8 . Studies conducted so far suggested that resistant alleles could be associated with negative pleiotropic effects that could affect mosquito fitness, vectorial competency and disrupts the normal physiological functions of the mosquito 9,10 . Insecticide resistance has always been associated with lower fecundity, longer developmental time, reduced longevity, and lower mating success 11,12 . With the increasing use of insecticides, mosquitoes have been reported over recent years to have become multiresistant to different insecticide compounds [13][14][15][16] . Understanding the influence of insecticide resistance on vector population dynamic is becoming crucial for the implementation of insecticide resistance management strategies. Although resistance is largely expanding in An. gambiae s.l. populations from Cameroon, there has been so far little information on the influence of pyrethroid resistance on An. gambiae s.l. fitness. Experimental infection studies comparing resistant versus susceptible colonies have suggested increased prevalence of Plasmodium infections in An. gambiae s.s. resistant strains [17][18][19] . Studies on the malaria vector An. funestus s.s. indicated that the presence of the L119F-GSTe2 resistant allele was associated with reduced fecundity, increased larval developmental time and adult longevity 20 . Further analysis suggested that this mechanism could also influence the vectorial capacity of resistant An. funestus s.s. populations 21 . During the last decade important changes have been reported in An. gambiae s.l. populations from the city of Yaoundé with populations becoming increasingly tolerant to organic pollution 22 , more resistant to pyrethroids and to different compounds 13,23,24 , changes in the biting behavior was also reported 25 . Yet the influence of these changes on the vectorial capacity or the fitness of An. gambiae s.l. populations has not been fully addressed.
In the present study an insecticide resistant An. coluzzii colony from the city of Yaoundé, was compared to a susceptible An. coluzzii laboratory colony "the Ngousso colony" to determine life-traits parameters affected by the increased expansion of insecticide resistance in this vector population.

Study site
The study was conducted in the city of Yaoundé, the capital of Cameroon (3° 52' 12 N; 11° 31' 12 E) from September 2018 to September 2019. In Yaoundé, An. gambiae s.l. is the main malaria vector. In order to obtained representative sample of the resistant An. gambiae s.l. population from the city of Yaoundé, anopheline larvae used to build the colony, were collected from different districts and locations (Tsinga, Nsam, Nkolbisson, Obobogo, Mvog-beti, Nouvelle Route Bastos, Nouvelle Route Tam-Tam, and Nouvelle Route Nkoldongo). In Yaoundé, malaria is highly prevalent, with the transmission rate varying from 0 to 90 infected bites/man/ year 25 . LLINs is the main method used by the population to prevent from malaria transmission 26 . According to recent record it is estimated that over 75% of households in Yaoundé own at least a net. Urban agriculture is also practiced on a large scale in the city and large quantities of pesticides are used by farmers and these, alongside the use of LLINs, are the main sources of insecticide selection for mosquitoes in the city 27,28 .

Susceptible strain
The laboratory colony used in the study is the Ngousso strain, originating from the district of Ngousso in Yaoundé, reared since 2006. This Ngousso An. coluzzii strain is fully susceptible to both permethrin and deltamethrin (mortality rate: ≥98% after one hour exposition to WHO impregnated papers).
Susceptibility assays and establishment of a resistant laboratory colony Anopheline larvae were collected in standing water collections on the field. Once in the laboratory, larvae were pooled and reared at mean temperature of 30°C and 73-75% relative humidity. After emergence, males were separated from females and adult females aged 3 to 5 days were used to conduct bioassays with deltamethrin (0.05%), permethrin (0.75%), DDT (4%), bendiocarb (0.1%) and Malathion (5%) according to WHO guidelines 12 .
The resistant An. coluzzii colony was established by regular selection (once every two generations) of mosquitoes, exposing 3 to 5 day-old unfed females and males to 0.05% deltamethrin for 12 generations. Batches of 20 to 25 mosquitoes per tube were exposed to 0.05% deltamethrin-impregnated papers for 1 hour. Bioassays were conducted at temperatures of 25±2°C and 70-80% relative humidity. Control tests were conducted using untreated papers. 24 hours after exposure, survivor male and female were pooled for the mating and fed with a 5% glucose solution. After selection, the susceptibility status of the F12 generation was checked for the following insecticides: permethrin (0.75%), DDT (4%), bendiocarb (0.1%), malathion (5%) and PBO (4%) in order to check the implication of P450 metabolic-based mechanisms.

Isofemale rearing and life-trait assessment Blood meal
To ensure that mosquitoes would feed, the glucose solution was removed 24 hours before blood feeding. Anopheline aged 3 to 6 days were placed into three cages (30×30×30 cm) of 100 females each for each strain and blood fed for 20 minutes on an anaesthetized rabbit. After blood feeding, females were provided with glucose solution, to allow maturation of eggs. The engorgement rate was assessed by counting well fed females.

Fecundity and fertility
For each strain (resistant and susceptible), 100 gravid females were placed in individual cups with damp filter paper to enable them lay eggs. After oviposition, the number of eggs laid per female was counted under a stereo microscope and eggs batch from each female were placed in water (plastic basins (17×12×6.5 cm) containing 200 ml of fresh water) 24 hours after the day of oviposition. All females that laid eggs were counted and stored at -20°C into numbered eppendorf tubes containing desiccant for further analyses. The egg from each isofemale line was reared separately.

Larval development
To reduce competition, a maximum of 50 larvae were reared per tray. Additional trays were used for females with more than 50 larvae. Larvae were fed using baby fish food (TetraMin) under standard insectary conditions. During the rearing process, water from the tray was replaced every two days to reduce the influence of evaporation or pollution. At the pupal stage, the number of pupae was recorded every day and they were transferred into paper cup (8.5×10×7cm), up to 30 pupae were placed per cup. Different information were recorded from each colony reared, the number of individual per larval stage, the length of larval development, the number of larvae reaching the pupae stage, the number of pupae emerging as adults and the sex ratio.

Adult longevity
After emergence, the number of males and females was recorded in each cup. Adult mosquitoes were fed with a 5% glucose solution. Each female progeny was placed in separate cup and followed. The survival rate was assessed by recording the number of males and females dying each day in each cup. Dead mosquitoes were removed from each cup daily for each family and kept at -20°C in 1.5-ml Eppendorf tubes containing desiccant.

Molecular identification
Genomic DNA was extracted from whole mosquitoes using a previously described protocol 29 . Molecular identification analyses were performed following the SINE200 PCR method describeb by Santolamazza et al. 30 to identify species of the An. gambiae complex.

Detection of knockdown resistance (kdr)
The presence of the kdr allele was checked in field collected mosquitoes (F0) and the laboratory resistant colony (F12 generation). A set of 119 females of the F0 generation and 111 F12 generation were randomly selected in each group for kdr analysis. L1014F and L1014S mutations were screened using the Mx3005P Real-Time PCR System, as describe previously (Bass et al., 2007(Bass et al., , 2010. The primers kdr-Forward (5'-CAT TTT TCT TGG CCA CTG TAG TGA T-3') and kdr-Reverse (5'-CGA TCT TGG TCC ATG TTA ATT TGC A-3') were used. The probes kdrW (5'-ACG ACA AAA TTT C-3') and kdrE (5'-ACG ACT GAA TTT C-3') labelled with fluorochrome FAM, were used to detect the mutant alleles (1014F and 1014S) while the probe Wildtype (5'-CTT ACG ACT AAA TTT C-3') labelled with fluorochrome HEX, were used for the wild type susceptible alleles detection. The Master Mix solution (9 µl) contained 5 µl of Sensimix (Biogen), 0.125 µl of each probe coupled to allelic specific primer and 3.875 µl of water. The thermocycler was set to run samples following the temperature cycling condition of: 1 hold of 10 minutes at 95°C for initial denaturation, followed by 40 cycles each of 95°C for 10 seconds, and 60°C for 45 seconds. For each experiment, there was one homozygote resistant for L1014S and L1014F kdr; one heterozygote for L1014S and L1014F kdr; and one susceptible L1014L used as positive controls. The negative controls were wells with 1 µl of ddH 2 O.

Statistical analysis
Statistical analyses were performed with the software Excel 2010 and the statistical analysis software R (version 4.0.0) via RCommander (Rcmdr Version 2.6-2) and RStudio (version 1.2.5042). The Shapiro-Wilk test was applied to assess whether eggs and larvae counts conform to a normal distribution. Comparison of proportions between the two strains was conducted using a chi-square (χ 2 ) test. Female fecundity and fertility was assessed by comparing respectively the oviposition and the hatching rate between both strains using Welch's twosample t-test. The life-trait parameters such as duration of larval development and longevity were assessed by comparing means of both susceptible and resistant strains using a Kruskal-Wallis non-parametric test. While those such as pupation, emergence, sex-ratio were assessed by comparing means of both strains using Welch's two sample t-test. To draw survival curves or Kaplan-Meier curve of larvae and adults, we used the package 'survminer' version 0.4.6 that contains the function 'ggsurvplot ()'. The latest allow drawing curve with the 'number at risk' table and the 'censoring count plot'. The level of significance of each test was set at α < 0.05. Scatter plots were obtained from genotypes scored using the MxPro-MX3005P qPCR Software and the fluorescence (ΔR) threshold adjusted manually for each dye, if necessary, to enable the correct scoring of positive controls. The allele frequency of individuals carrying the kdr mutation was calculated using the formula f(R) = (2 × RR + RS)/2N, with RR = total number of homozygote resistant; RS = total number heterozygote resistant; N = total number of mosquitoes screened for the kdr mutation. Genotype frequency was calculated as relative frequencies of the homozygote resistant and heterozygote resistant individuals.

Fecundity and fertility
After choosing 100 fully blood fed females of each strain for individual egg laying, in the susceptible group, 86 females laid. A total of 9604 eggs, corresponding to a mean of 111.67±5.36 eggs/female was recorded (Table 2). In the resistant colony, 40 females laid 3736 eggs (Table 1), which correspond to an average of 93.4±10.77 eggs/female for females which were able to lay (Table 2). In the resistant and the susceptible colonies, eggs count was found in conformity with a normal distribution Shapiro-Wilk normality test (resistant: W = 0.94774, P = 0.06339, susceptible: W = 0.9707, P = 0.05).

Larval and pupae development
The average length of larval development from hatching to pupation was compared between the two colonies. The average length of larval development to the pupa stage was 7.6 days for the susceptible and 10.6 days for the resistant group ( Figure 1). Survival probability analysis indicated that, resistant larvae take much more time to arrive to the pupa stage Abbreviations: N: total number of mosquitoes initially allowed to blood feed and lay eggs; n: number of mosquitoes who successfully blood fed and laid eggs; * significantly different (P<0.05). compared to the susceptible (χ 2 = 2251, df = 1, P = <2×10 -16 ).
In the susceptible group, high pupation rate was recorded between days 7 and 8, whereas for resistant larvae a peak of pupation was detected on day 10 ( Figure 1).

Mean number of pupae and emergence rate in susceptible and resistant strain
In the susceptible colony, out of 7101 first instars larvae, 5434 larvae successfully arrived at the pupae stage and 5226 emerged as adults. The following correspond to 76.52% getting to the pupae stage and 73.6% to the adult stage. Within resistant out of 2459 larvae of the first instar, 1976 successfully arrived at the pupa stage (80.36%) and 1907 (77.55%) emerged as adults. The average number of offspring pupae per female was not significantly different between the two groups (Welch two-sample t-test: t = -1.1822, df = 95.158, P = 0.12) ( Table 2). The average number of adults emerging was also not significantly different between the two groups (Welch two-sample t-test: t = -0.67447, df = 82.33, P = 0.251) ( Table 2). Mortality during emergence was similar between susceptible and resistant groups (Welch two-sample t-test: t = 1.2601, df = 54.338, P = 0.1065) ( Table 2).

Sex-ratio of adults mosquitoes
Of the 5226 mosquitoes who successfully emerged as adults in the susceptible group, 48.41% (N=2530) were males and 51.6% (N=2696) were females. Of the 1907 mosquitoes who successfully emerged as adults in the resistant group, there was 51.49% (N=982) of males and 48.50% (N=925) females. The proportion of emerging female was statistically similar to that of male in each offspring group, susceptible colony (χ 2 = 0.165, df = 1, P = 0.6848) and resistant colony (χ 2 = 0.055, df = 1, P = 0.8145). There were significantly more susceptible female than resistant female in the offspring (Welch two-sample t-test: t = 3.0962, df = 61.992, P = 0.001471) ( Table 2).  for the F12 generation. In the same way, the F12 generation showed 0% mortality to DDT 4%, 1.25% mortality rate to permethrin 0.75%, no change in their susceptibility to bendiocarb 0.1% (mortality rate: 95±1.35%) and malathion 5% (mortality rate: 100%) ( Figure 3A). When mosquitoes of the F12 generation were preexposed to PBO a mortality rate of 67.5±6.87% to deltamethrin 0.05% was recorded whereas, no variation in the mortality to permethrin 0.75% was recorded.

Molecular identification and kdr
Raw data for this study are available, see Underlying data 31 .

Discussion
Insecticide resistance is rapidly expanding in An. gambiae s.l. population from the city of Yaoundé 13,32 . Yet the influence of insecticide resistance on An. gambiae s.l. species life trait and evolution is not well understood 33 . The present study was undertaken to assess the influence of insecticide resistance on the fitness and life trait of An. coluzzii population by comparing a susceptible to a resistant colony. The study indicated high fitness cost in the resistant compared to the susceptible colony. This was consistent with previous observations done on the vector An. funestus s.s. in Cameroon 10,19 . Mosquitoes in the city of Yaoundé have been reported to be resistant to pyrethroids, and also to a large set of compounds including DDT, carbamate and pollutants 23,28,32,34 . Comparison of many life-trait parameters between the two colonies indicated differences at different levels. The blood feeding success was high in susceptible compare to resistant. The difference could have resulted from the fact that the susceptible colony was more adapted to blood feeding on rabbit blood as it has been maintained in laboratory for almost 14 years, whereas this is not the case for the resistant colony which has just been colonized in the insectary. Similar observations have been reported elsewhere 11 . According to Martins et al., the blood meal is a key parameter which directly affects the general fitness, since it influences the number of eggs laid 11 . Eggs from susceptible    The density of pupae and the transition to adult stage were not significantly different between the two strains, suggesting that fitness cost mostly affects the larval stage and has no visible effect on pupae and adult emergence. The following could derive from the fact that pupae do not feed and there is limited influence of exogenic factors such as density or nutrient on this stage 36,40,41 . No significant distortion of the sex ratio was recorded neither in the susceptible nor in the resistant colony. The absence of difference at this level, particularly for the resistant colony, could derive from the reduce number of larvae succeeding at the pupae and adult stage. In the nature, one male can inseminate several females during its lifespan, while females just need a unique insemination to accomplish multiple gonotrophic cycle for the rest of their life 42 . Therefore, the number of females reaching the adult stage is an important parameter of reproduction success and vectorial capacity whereas this is not true for males 42 .
Despite a similar life span between males and females in each colony, the progeny of susceptible mosquitoes was found to live longer in general 6 to 7 days longer than resistant mosquitoes.

Conclusion
The study suggests that increase expansion of insecticide resistance in An. coluzzii populations from the city of Yaoundé, is likely associated with accumulation of deleterious effects affecting the life-traits of An. coluzzii. It appears from the study that the longer development time could render resistant larvae more vulnerable to control measures such as larviciding and to predators. The authors are reporting their study on the "Fitness cost of insecticide resistance on the life-traits of a Anopheles coluzzii population from the city of Yaoundé, Cameroon". Insecticide resistance is the most important factor limiting the success of malaria control program in Africa. More data are needed to understand the effect of insecticide resistance on malaria mosquitoes; the fitness cost involved in getting resistance and how it affects their potential to transmit malaria and other vector-borne diseases. Therefore, the authors have chosen an important subject in malaria research to undertake their study. However, enthusiasm is diminished by weaknesses in the design of the study, and to some extent, the presentation of the results, making it difficult to recommend it for publication. Insectary strains of mosquitoes are known to have reduced resilience compared to wild caught mosquitoes. Being kept in captivity over several generations over the years makes the mosquitoes lose so many life-history traits, including resistant alleles, more especially as they are not under any selection pressure. Wild mosquitoes are under several pressures that could affect their malaria transmission potential. For a colony that has been kept for 14 years since 2006, their life history traits have been affected considerably compared to mosquitoes in the wild. Therefore, to make a comparison of wild-caught resistant mosquitoes that have been selected with deltamethrin for 12 generations with insectary colonized ones is not sound in my opinion. There should have been another wild-caught colony mosquitoes from the same area that were not selected with insecticide which would eventually lose their resistant abilities at some point to make the comparison. This colony would have been the unselected or susceptible strain. The insectary colony should have been a control strain for this experiment. Because of this weakness in the design of the experiment, I am unable to recommend it for publication.

Data availability
Other comments are as follows: The manuscript needs a revision for its language to bring clarity to the text.

Methods:
It is not clear if the same mosquitoes that were hatched from the blood feeding of the adults in the first instance, were the same mosquitoes that were reared to study their development time, pupation rate, emergence rate, and also for their adult longevity. If it is the same that was used then this should be clearly stated.

Results
Under the heading "Fecundity and fertility" There is no information on the susceptible colony. Please compare the two colonies.
○ Under the heading "Life span of the progeny of susceptible and resistant adults Mosquitoes" What was the difference between resistant and susceptible colonies for males and females. This was compared ○ Under the heading "Molecular identification and kdr" Ithe authors report that "molecular identification performed on field-collected mosquitoes (F0 generation) showed that 7.5% (7/93) belong to An. gambiae and 92.47% (86/93) were An. coluzzii." They go on to state that the F12 generation was all An. coluzzii. What happened to the An. gambiae population over time? Did they become fewer, got eliminated or their numbers were increased but were not part of the species ID done at the 12 th generation. In addition, a better clarification of how many colonies involved in the study is needed in that section.
○ Figure 2 legend needs revising. What actually is "B"? There is also no explanation to the figure called "Strata" in both Fig 2A and 2B ○ Is the work clearly and accurately presented and does it cite the current literature? Partly

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility? Yes

Are the conclusions drawn adequately supported by the results? No
Competing Interests: No competing interests were disclosed.

I confirm that I have read this submission and believe that I have an appropriate level of expertise to state that I do not consider it to be of an acceptable scientific standard, for reasons outlined above.
Author Response 04 Sep 2020 Christophe Antonio-N'kondjio, OCEAC, Yaoundé, Cameroon Reviewer 3

Query 1:
The authors are reporting their study on the "Fitness cost of insecticide resistance on the life-traits of a Anopheles coluzzii population from the city of Yaoundé, Cameroon". Insecticide resistance is the most important factor limiting the success of malaria control program in Africa. More data are needed to understand the effect of insecticide resistance on malaria mosquitoes; the fitness cost involved in getting resistance and how it affects their potential to transmit malaria and other vector-borne diseases. Therefore, the authors have chosen an important subject in malaria research to undertake their study. However, enthusiasm is diminished by weaknesses in the design of the study, and to some extent, the presentation of the results, making it difficult to recommend it for publication. Insectary strains of mosquitoes are known to have reduced resilience compared to wild caught mosquitoes. Being kept in captivity over several generations over the years makes the mosquitoes lose so many life-history traits, including resistant alleles, more especially as they are not under any selection pressure. Wild mosquitoes are under several pressures that could affect their malaria transmission potential. For a colony that has been kept for 14 years since 2006, their life history traits have been affected considerably compared to mosquitoes in the wild. Therefore, to make a comparison of wild-caught resistant mosquitoes that have been selected with deltamethrin for 12 generations with insectary colonized ones is not sound in my opinion. There should have been another wild-caught colony mosquitoes from the same area that were not selected with insecticide which would eventually lose their resistant abilities at some point to make the comparison. This colony would have been the unselected or susceptible strain. The insectary colony should have been a control strain for this experiment. Because of this weakness in the design of the experiment, I am unable to recommend it for publication.
Other comments are as follows: The manuscript needs a revision for its language to bring clarity to the text.

Response 1:
We used a laboratory strain as control because it was not possible to have a field susceptible population. With the expansion of insecticide resistance in mosquito populations it has now become difficult to find susceptible field population in Yaoundé. The Ngousso strain used in this case as control originate from Yaoundé and is of the species An. coluzzii and is known for its high susceptibility to all insecticide classes. Several experimental study design comparing laboratory strains to field population have been reported in the literature and are still conducted nowadays. For instance the Kisumu laboratory strain is largely used for experimental studies and also in WHO bioassays to check the quality of impregnated papers. The argument that using a laboratory strain is a major weakness in the design of our study doesn't make sense we totally disagree with the reviewer's point of view.
© 2020 Okumu F. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Fredros O. Okumu
Environmental Health and Ecological Sciences Department, Ifakara Health Institute, Ifakara, Tanzania The work by Nkahe et al. addresses an important question on how pyrethroid resistance might impact several life cycle traits of malaria vectors, and how this may translate to performance of control measures.
The writing is excellent, study methods properly described and the scientific rationale clearly stated.
I found it particularly interesting that susceptible mosquitoes live so much longer than resistant ones (~22 days compared to ~15 days). Coupled with the reduced blood-feeding proportions, this could mean very high levels of impact from insecticidal-interventions even if they do not impart direct mortality. It would be greater if this was discussed further with respect to the expected performance of interventions such as pyrethroid-treated nets.

Two minor comments are as follows:
In the discussion section, the authors indicate that most of the fitness costs were in the aquatic stages, yet we see some great reductions on fecundity and adult survival as well.
Since vector survival has far greater impacts on pathogen transmission than actual densities, it is perhaps incorrect to conclude that the fitness costs are mostly in the preadult stages.

○
Would be great to see clarity on the decision to use 12 generations of the resistant colonies ○ Is the work clearly and accurately presented and does it cite the current literature? Yes

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? The writing is excellent, study methods properly described and the scientific rationale clearly stated.
Query 1: I found it particularly interesting that susceptible mosquitoes live so much longer than resistant ones (~22 days compared to ~15 days). Coupled with the reduced bloodfeeding proportions, this could mean very high levels of impact from insecticidalinterventions even if they do not impact direct mortality. It would be greater if this was discussed further with respect to the expected performance of interventions such as pyrethroid-treated nets. Response 1: The following phrase was added in the discussion section "Although the use of pyrethroid treated nets are considered to induce a low mortality in resistant individuals, resistant mosquitoes were however found to exhibit reduce blood feeding rate, low fecundity and short adult survival rate all this somewhere suggest a long term impact of insecticide base intervention (such as pyrethroid treated nets) selection on vector population. This long term negative impact confirm continuous performance of pyrethroid treated nets interventions on vector populations. This unrecognized impact of treated nets need to be evaluated in different epidemiological settings".
Two minor comments are as follows: Query 2: In the discussion section, the authors indicate that most of the fitness costs were in the aquatic stages, yet we see some great reductions on fecundity and adult survival as well. Since vector survival has far greater impacts on pathogen transmission than actual densities, it is perhaps incorrect to conclude that the fitness costs are mostly in the preadult stages. Response 2: The conclusion was corrected to address this comment.
Query 3: Would be great to see clarity on the decision to use 12 generations of the resistant colonies Response 3: We selected mosquitoes for 12 generations to have highly resistant individuals and avoid susceptible individuals in the colony.
record it is estimated that over 75% of households in Yaoundé own at least a net".

Molecular identification and kdr:
Molecular identification performed on field-collected mosquitoes (F0 generation) showed that 7.5% (7/93) belong to An. gambiae and 92.47% (86/93) were An. coluzzii. While identification of 103 susceptible and 111 resistant (F12 generation) mosquitoes confirms that the two groups belong to An. coluzzii species. This finding is interesting and needs to be discussed in the Discussion section.

○
In Table 3: The sum of the percentage from N/119 is 101.53%, not 100%. This occurs because the percentage of RR should be 20.17% rather than 21.7%. Please check and correct throughout the manuscript.

○
In Table 3: The count of alleles (2N) should be 222 rather than 228 for the F12 ○ ○ Query 5: In the sentence: "Of the 5226 mosquitoes who successfully emerged as adults in the susceptible group, 48.41% (N=2530) were males and 51.59% (N=2696) were females.", the sum of both percentages is 100.1% instead of 100%, therefore it is important to choose either to keep 1 or (48.4% and 51.6%) or two numbers for the decimal (48.41% and 51.59%). Check throughout the manuscript. Response 5: Corrected and checking done throughout the manuscript.
Life span of the progeny of susceptible and resistant adults mosquitoes: Query 6: Line 10: In the sentence "The difference between the average number of males and females in each group was not significant for the susceptible group", replace the average number by average life span number. Response 6: Changes done.
Insecticide resistance profile of F0 and F12 generation: Query 7: What do the error bars mean on Figure 3 (confidence intervals or SEM?). Please add the meaning below the figure.
Response 7: The additional information was added below each figure Figure 1: Error bars= Standard Deviation Figure 3: Error bars= confidence intervals

Molecular identification and kdr:
Query 8: Molecular identification performed on field-collected mosquitoes (F0 generation) showed that 7.5% (7/93) belong to An. gambiae and 92.47% (86/93) were An. coluzzii. While identification of 103 susceptible and 111 resistant (F12 generation) mosquitoes confirms that the two groups belong to An. coluzzii species. This finding is interesting and needs to be discussed in the Discussion section. Response 8: Only An. coluzzii was successfully maintained in laboratory across generations. The findings was discussed in the discussion section. Query 9: In Table 3: The sum of the percentage from N/119 is 101.53%, not 100%. This occurs because the percentage of RR should be 20.17% rather than 21.7%. Please check and correct throughout the manuscript.

○
In Table 3: The count of alleles (2N) should be 222 rather than 228 for the F12 generation ie. 2 x111. Please check and correct.