Development of circulating isolates of Plasmodium falciparum is accelerated in Anopheles vectors with reduced reproductive output

Anopheles gambiae and its sibling species Anopheles coluzzii are the most efficient vectors of the malaria parasite Plasmodium falciparum. When females of these species feed on an infected human host, oogenesis and parasite development proceed concurrently, but interactions between these processes are not fully understood. Using multiple natural P. falciparum isolates from Burkina Faso, we show that in both vectors, impairing steroid hormone signaling to disrupt oogenesis leads to accelerated oocyst growth and in a manner that appears to depend on both parasite and mosquito genotype. Consistently, we find that egg numbers are negatively linked to oocyst size, a metric for the rate of oocyst development. Oocyst growth rates are also strongly accelerated in females that are in a pre-gravid state, i.e. that fail to develop eggs after an initial blood meal. Overall, these findings advance our understanding of mosquito-parasite interactions that influence P. falciparum development in malaria-endemic regions.

Anopheles gambiae and its sibling species Anopheles coluzzii are the most efficient vectors of the malaria parasite Plasmodium falciparum.When females of these species feed on an infected human host, oogenesis and parasite development proceed concurrently, but interactions between these processes are not fully understood.Using multiple natural P. falciparum isolates from Burkina Faso, we show that in both vectors, impairing steroid hormone signaling to disrupt oogenesis leads to accelerated oocyst growth and in a manner that appears to depend on both parasite and mosquito genotype.
Consistently, we find that egg numbers are negatively linked to oocyst size, a metric for the rate of oocyst development.Oocyst growth rates are also strongly accelerated in females that are in a pre-gravid state, i.e. that fail to develop eggs after an initial blood meal.Overall, these findings advance our understanding of mosquito-parasite interactions that influence P. falciparum development in malaria-endemic regions.

Author Summary (150-200 words):
Malaria, an infectious disease caused by Plasmodium parasites, continues to affect millions of people each year.These parasites are transmitted by Anopheles mosquitoes during blood feeding, which is required by females to produce eggs.After they are ingested by the female, parasites start developing in the midgut at the same time as the mosquito initiates oogenesis, the process of egg development.Here, we investigated how the concomitant processes of egg and parasite development interact in field-derived Anopheles and Plasmodium isolates.We found that in these naturally occurring pairs, oogenesis is negatively linked to parasite development, such that mosquitoes producing

Introduction
Plasmodium parasites, the causative agents of malaria, are transmitted by female mosquitoes from the Anopheles genus.Like all anautogenous mosquitoes, female Anopheles require a blood meal to produce eggs, and parasites take advantage of this requirement to enable their transmission between human hosts.Blood feeding triggers a vast metabolic and transcriptional program, largely regulated by the insect steroid hormone 20-hydroxyecdysone (20E), which ultimately results in the production of an egg batch (1,2,3).Concurrently, ingested parasites begin their development (sporogony) in the mosquito gut when male and female gametes fuse to form a zygote, which then rapidly transforms into a motile ookinete that breaks across the midgut epithelium and settles beneath the basal lamina to form a sessile oocyst.Oocysts subsequently grow over several days producing thousands of sporozoites, which are eventually released into the mosquito hemolymph and invade the salivary glands, from where they can be transmitted to the next human host when the female blood feeds again.
One critical aspect of Plasmodium infection in the mosquito is the rate at which oocysts develop, as this ultimately determines the extrinsic incubation period (EIP) (4), or the length of time it takes parasites to complete sporogonic development and reach transmission stages.The EIP is a key determinant of malaria transmission due to the lengthy parasite developmental cycle.Indeed, Plasmodium falciparum, the deadliest form of human malaria, requires a minimum of 9 days for development (5,6,7,8), mostly due to the extensive time needed for oocyst growth.This time largely corresponds to the rather limited lifespan of female mosquitoes in the field, which is variable (9) but estimated to last for 2-3 weeks (10,11,12)).
Factors known to affect the length of P. falciparum sporogony are temperature (13,14,15,16), adult diet (17), larval nutrition (18), and multiple blood meals (7,19,20) (21,22).Additionally, we have demonstrated that the rate of P. falciparum development in Anopheles gambiae-a major malaria vector across much of sub-Saharan Africa-is also linked to the hormonally-regulated processes leading to oogenesis.Impairing the mosquito's ability to produce eggs through several methods, including disrupting the function of the steroid hormone 20E via silencing its heterodimeric nuclear receptor Ecdysone Receptor (EcR)/Ultraspiracle, resulted in faster oocyst development.This in turn resulted in sporozoites reaching the salivary glands sooner while, crucially, still maintaining full infectivity to human hepatocytes (23).Moreover, we determined that egg numbers are negatively correlated to P. falciparum growth rates, a finding implying that mosquitoes with reduced reproductive output can transmit parasites sooner, thereby providing parasites with a significant advantage given the short mosquito life span.
Understanding whether this interaction is relevant in malaria-endemic regions where different P. falciparum genotypes and multiple Anopheles species coexist is essential to reveal the complex dynamics of this vector-parasite interplay.
Here, we examine interactions between oogenesis and sporogony in natural settings in Burkina Faso using multiple P. falciparum isolates collected from gametocyte donors and field-derived colonies of two Anopheles species.We determine that silencing the 20E nuclear receptor EcR in An. gambiae and its sibling species Anopheles coluzzii leads to faster parasite development.In agreement with these findings, in both species, we find that oocyst growth rates are negatively linked to oogenesis such that females naturally producing fewer eggs harbor faster developing oocysts.We observe variability in the oocyst growth response to EcR silencing across different P. falciparum isolates and between the two Anopheles species, suggesting that both parasite and mosquito genotype are determinants of growth.Additionally, parasite growth rates are significantly accelerated in field-collected An. coluzzii females that are in a pre-gravid state (i.e.unable to develop eggs after an initial blood meal), a very common phenomenon in natural mosquito populations (24,25).Overall, our data across diverse Anopheles-Plasmodium combinations suggest that parasite development is favored in females that have reduced or impaired oogenesis, expanding our understanding of factors affecting sporogony in the field.These data also have implications for malaria control strategies that aim to interfere with mosquito reproduction.

An. coluzzii and An. gambiae
To investigate if P. falciparum development is affected by Anopheles oogenesis in field settings, we performed infection experiments in Burkina Faso with naturally circulating P. falciparum isolates collected from six gametocyte carriers, here denoted as patient 1-6 (p1-p6) (26).We used colonies of An. coluzzii and An.gambiae that are frequently replenished from local wild populations.In both species, we impaired 20E signaling via RNAi silencing of the ecdysone nuclear receptor (dsEcR), or green fluorescent protein (dsGFP) as a control.Three P. falciparum isolates were used to infect An. coluzzii (p1- gambiae in these conditions (23).We went on to determine parasite growth rates by measuring the cross-sectional area of oocysts at 8 days post-infected blood meal (pIBM) (23) and found a striking increase following dsEcR injections in both Anopheles species.Specifically, in EcR-depleted females P. falciparum oocysts were 55% larger in An. coluzzii and 78% larger in An. gambiae as compared to their respective dsGFP controls (Fig 2A-C).
Our studies using cultured NF54 parasites had shown that larger oocysts are indicative of faster parasite development (7,23).Although we did not measure EIP directly in this study, we found that dsEcR An. gambiae females had a 5.5-fold increase in the number of sporozoites in their salivary glands 12 days after infection relative to controls (Fig Notably, in our experimental settings (23), 12 days pIBM is a time point when sporozoites have started but not yet completed invasion of the salivary glands, so it is likely that this increase in sporozoites reflects faster parasite development.This trend was also To determine a possible mechanism behind accelerated growth, we measured expression levels of the lipid transporter lipophorin (Lp), previously implicated in the faster oocyst growth rates observed in laboratory depletions of EcR (23).We detected elevated Lp expression in dsEcR females of An. coluzzii and An.gambiae (Fig 2E), which could allow greater provisioning of lipids to the developing oocysts (23).Combined, these data show that when 20E signaling is disrupted, growth of P. falciparum field isolates is accelerated, and this effect may be mediated by Lp-transported lipids, consistent with previous findings (23).

Parasite growth is negatively linked to egg development via 20E signaling
We next determined the relationship between oocyst size and egg numbers within our two Anopheles species.To do this, we used a Generalized Linear Mixed-Effects Model (GLMM) (27,28) to analyze paired egg-oocyst data collected from all P. falciparuminfected females (S1 Table , S2 Table).In An. coluzzii, there was a significant two-way interaction between egg number and dsEcR treatment (S1 Table ), suggesting that the effect of egg development on oocyst size differs between control and dsEcR females.
Consistent with this, when analyzed independently, while An.coluzzii controls showed a significant negative association between egg numbers and oocyst size, in EcR-silenced females this association was lost (Fig 3A).These findings mean that in An. coluzzii, oocysts tend to be larger in females that naturally produce fewer eggs, and conversely, females that develop more eggs are likely to have smaller oocysts, but impairing 20Esignaling breaks this link between egg and oocyst development.
Instead in An. gambiae we found a complex three-way interaction between egg number, dsEcR treatment, and oocyst number (S2 Table ), suggesting that both 20E signaling and oocyst density regulate the egg-oocyst relationship.When analyzed independently, both control and EcR-silenced females have a general negative association between egg numbers and oocyst size (Fig 3B), and this association is strongest at low (<10) oocyst densities in controls and at high (>30) oocyst densities in dsEcR females (Fig 3C).As with An. coluzzii, these findings demonstrate that oocyst growth is inversely linked to a female's investment in reproduction, likely because resources that are not utilized for oogenesis become available to parasites.All together these data reveal that egg development is an important factor affecting oocyst growth in two Anopheles species, but also, that signaling by the steroid hormone 20E may differently regulate the egg-oocyst interaction in these mosquito genetic backgrounds.

Parasite isolates vary in their growth response to impaired 20E signaling
As malaria is highly prevalent in the villages surrounding Bobo-Dioulasso (29,30), we reasoned that individual patients may have polyclonal infections comprised of multiple P. falciparum genotypes.Indeed, we found that most of the isolates used in our infections contain two or more distinct parasite genotypes, as determined by analyzing the merozoite surface protein 1 (MSP1) (31) (Table 1, S3 Table ).When we assessed parasite growth in the individual p1-p6 infections (Fig 4A, B), we found that oocyst size does not vary between parasite isolates in dsGFP controls (An.coluzzii dsGFP: GLMM, LRT X 2 2 = 1.52, p = 0.466; An. gambiae dsGFP: GLMM, LRT X 2 4 = 3.47, p = 0.483).On the other hand, in dsEcR females, we detected significant differences between isolates (An.coluzzii dsEcR: GLMM, LRT X 2 1 = 6.3, p = 0.012; An. gambiae dsEcR: GLMM, LRT X 2 1 = 4.67, p = 0.031).Despite some inevitable variability due to using different mosquito batches, all isolates increased in size in dsEcR females when compared to controls, with the notable exception of p3 in An. coluzzii (Fig 4A  An. gambiae and An.coluzzii, these data suggest that the mosquito genotype may also play a role in determining parasite growth rates.Further supporting this idea, in controls infected with p1 or p3 parasites, we see that oocyst growth is significantly different between our two mosquito species, whereby p1 and p3 oocysts grow larger in An. coluzzii than in An. gambiae (GLMM, LRT X 2 1 = 4.03, p = 0.045, Fig 4C).P. falciparum parasites were collected from six different gametocyte carriers (p1-p6) and used for infections in An. coluzzii and An.gambiae females.Prior to infection, a blood smear was used to count the number of gametocytes present in each sample.Gametocytes were counted per 1,000 leukocytes and then, using an estimated conversion factor of 8,000 leukocytes per uL of blood, converted to gametocytes per μL of blood (Gams/μL).MSP1-typing was performed on dried blood spots or recently thawed frozen blood samples and used to determine the (minimum) number of unique parasite genotypes, or complexity of infection (COI), present in each isolate.Oocyst growth is accelerated in pre-gravid mosquitoes

Parasite
Given the observed negative association between mosquito oogenesis and parasite growth rates (Fig 3), we next set out to investigate oocyst development in pre-gravid mosquitoes (females that fail to develop eggs after taking an initial blood meal), using field-derived Anopheles because this phenomenon is highly common in wild populations and considerably less so in laboratory colonies (24,25,32).To this aim, we collected An.
coluzzii larvae from natural breeding sites and infected emerged adult females with the same six gametocyte carriers as above (p1-p6).Out of the 323 mosquitoes analyzed in these experiments, the vast majority (86%, N = 277) failed to develop eggs despite taking a blood meal (S4A Fig) .This effect could be driven by the fact that field mosquitoes are not well-adapted to membrane feeding and may take smaller blood meals, but additionally, it may also suggest that the pre-gravid rate is high in this population as previously observed (24,33).
Altogether, the field mosquitoes averaged 72% infection prevalence with a mean of 11 Therefore, pre-gravid status in An. coluzzii has two contrasting effects on parasite infection within the mosquito -it reduces infection prevalence but also strikingly promotes oocyst growth, with probable effects on the EIP.

Discussion
In this study, we investigated the interplay between mosquito oogenesis and parasite development using multiple P. falciparum field isolates and two different Anopheles species, including field collected mosquitoes.We found that oocyst size, a good metric for oocyst growth rates (7,23), is negatively linked to oogenesis in An. coluzzii and An.)), parasites that develop faster will have a greater chance of being transmitted (7).Although here we could not perform hepatocyte infections due to restrictions in the experimental set-up, our previous laboratory studies showed that faster developing parasites are just as transmissible to human hepatocytes as slower developing parasites (23).
Consistent with accelerated parasite growth in dsEcR females, we also found that in both An. coluzzii and An.gambiae controls, egg development is negatively linked to oocyst size (Fig 3A-B), confirming that P. falciparum growth is influenced by female reproductive processes (23).In An. gambiae, the effect of oocyst density on this interaction suggests that at low density infections, any change in the pathways leading to egg development results in a greater per-oocyst change in available resources, therefore affecting oocyst growth more profoundly than at high density infections (Fig 3C).Disruption of 20Esignaling perturbs the effect of parasite density on these egg-oocyst size interactions, but in contrast to An. coluzzii, the negative link between egg numbers and oocyst size remains intact (Fig 3A, B).Overall, these results demonstrate that there is a strong, conserved interaction between Anopheles oogenesis and oocyst growth, but also that 20E signaling regulates this interaction differentially, even in closely related species, indicating that P. falciparum interacts uniquely with the post-blood meal environments of these two anophelines.Previously, we demonstrated that the negative association between egg development and oocyst growth is dependent on midgut lipids that the parasite may access via Lipophorin (23).Given that we observed elevated Lp expression in the dsEcR-treated females (Fig 2E), it is likely that resource allocation and lipids/lipid transporters also play a role in the oocyst growth phenotypes we observed here in the field, although this remains to be tested directly.
Further supporting the link between oogenesis and parasite growth, in infections with field-collected An. coluzzii, pre-gravid females that failed to develop eggs had significantly larger oocysts compared to females that produced an egg batch (Fig 4E).On the other hand, pre-gravid females were also less likely to become infected initially (pie charts in

Fig 4D
), as also seen previously ( 17), perhaps because these females take smaller blood meals and therefore are less exposed to infectious parasites.These contrasting effects of the pre-gravid status on two major determinants of vectorial capacity (vector competence and speed of parasite development) make it difficult to conclude how pregravid mosquitoes may contribute to malaria transmission dynamics, and models of transmission would be needed to assess this fully.Despite this limitation, given that the pre-gravid condition is quite common and pre-gravid rates over 60% have previously been reported for certain Anopheles field populations (24,33), possibly due to low teneral reserves (32,33,34), our observations that pre-gravid females favor oocyst growth may suggest that the EIP of parasites in field settings is actually shorter than that typically observed with lab-adapted colony mosquitoes.Furthermore, the effect of pre-gravid status on oocyst growth rates could be even further exacerbated by these females needing to seek an additional blood meal to complete production of their first egg batch, since it has been shown that a second blood meal significantly accelerates sporozoite formation (7).
In our experiments, oocyst intensity and prevalence were not affected by EcR silencing in either An.coluzzii or An.gambiae females (Fig 1E-F).This is partly in contrast to our previous study where we found a reduction in parasite numbers (but not prevalence) during the ookinete-oocyst transition.Considering that parasite numbers are roughly 4 times lower here than those achieved in laboratory conditions (23), it is possible that the effects of 20E signaling on parasite survival are dependent on the intensity of infection.
Given that oocyst density was also a significant factor affecting egg-oocyst size relationships in An. gambiae (Fig 3C), it would be worthwhile to further explore how infection load may influence these multiple interactions between parasites and oogenetic processes.Interestingly, dsEcR treatment did not significantly reduce egg numbers in An.
coluzzii, while it did in An. gambiae (Fig 1C-D).Due to the low levels of EcR expression in females prior to blood feeding, knockdown efficiency was confirmed via increased expression of the lipid transporter Lp (one of the genes that is strongly induced by EcR silencing ( 23))(Fig 2E), suggesting that silencing was comparable in these species.A possible explanation for these results is that 20E regulation of post-blood feeding processes is slightly different in these two species, as also indicated by the differential effects of dsEcR treatment on the egg-oocyst size interaction in An. coluzzii compared to significantly influence oocyst growth rates.Considering that high levels of genetic diversity exist in parasite populations in regions with high malaria transmission (36,37,38) such as Burkina Faso (39), there is likely natural variability in parasite genes regulating development (e.g.genes involved in metabolism, or nutrient transporters).It is plausible that these differences may be caused by the parasite's need to adapt to and be transmitted by different sympatric Anopheles species populating the same geographical area (40,41,42), thus maintaining polymorphisms in genes affecting oocyst growth rates.
Interestingly, we found that oocyst growth significantly varied across parasite isolates in This may be because the effects of parasite genotype on oocyst growth are only detectable through our method of oocyst measurement in conditions where a great surplus of nutrients are accessible to the parasite throughout development, as when 20E signaling is impaired (23).Isolate-specific differences in parasite growth may emerge more clearly following additional blood meals, when growth rates are profoundly accelerated (7,19,20).Providing a second blood meal to these mosquitoes would be most representative of field conditions, as pre-gravid mosquitoes will require a second feed into order to produce eggs.
Altogether, this study expands our knowledge of naturally occurring mosquito-parasite interactions that influence the rate of sporogony in the field.These findings are important for understanding the dynamics of P. falciparum transmission in sub-Saharan Africa, and for informing the development of future vector control strategies that seek to interfere with mosquito reproduction.Indeed, our data emphasize that mosquito species may differ in their interactions with parasite isolates.Manipulations to fundamental processes like mosquito reproduction will need to be evaluated for their effects within the specific contexts in which control strategies are implemented and cannot necessarily be extrapolated from laboratory experiments that use single combinations of lab-adapted mosquito and parasite strains.
conditions (27 °C, 70-80% humidity, 12h day/night cycle).Female mosquitoes were maintained as virgins throughout adulthood.Prior to blood feeding, mosquitoes were mouth aspirated from cages into large cups sealed with netting and sugar-starved for ~24hr.After blood feeding, engorged mosquitoes were returned to cages and again maintained as before on 10% glucose solution under standard conditions.Before any dissections, mosquitoes were cold-anesthetized and decapitated.
dsRNA injections: Adult virgin females <24h post-eclosion were anesthetized on ice and then injected in the thorax with 69 nL of dsRNA (dsGFP or dsEcR) at a concentration of 10 μg/μl using a Nanoject II (Drummond).Immediately following injection, females were transferred to a large cup for recovery before being moved to a new cage where they were maintained under standard rearing conditions as described above; only females surviving injections (>90%) were used in experiments.3-4 days post-injection, females were provided with a P. falciparum-infected blood meal.

Gene expression analysis: Sample collection:
On the same day that mosquitoes were provided a P. falciparum-infected blood meal (3-4 days post-injection), non-blood fed females were collected for gene expression analysis.Females were coldanesthetized, de-capitated, and then the rest of body samples were stored in 500 μl of RNAlater (Invitrogen).2 pools of 5 mosquitoes were collected from both dsGFP and dsEcR groups, with collections done for each batch of mosquitoes that was infected (3 batches of An. gambiae and 2 batches of An. coluzzii (Fig 1A, B)).Samples in RNAlater were kept for ~1 day at room temperature or 4°C, before being frozen at -20°C.During transport back to Boston, USA from Bobo-Dioulasso, Burkina Faso, the samples were at room temperature for ~3 days, before again returning to -20°C for long-term storage.
RNA extraction & cDNA synthesis: Using sterile forceps, mosquito samples were removed from the collection tubes containing RNAlater, blotted on a Kimwipe, and then transferred to a fresh set of tubes containing 400 μl of TRI Reagent (Thermo Fisher Scientific).A bead-beater was used to homogenize the samples using two 2-mm beads per tube and beating at 2400 rpm for 2 rounds of 90 seconds each (chilling samples at -20°C for 5 min between rounds of homogenization).Following homogenization, RNA extraction was performed according to TRI Reagent manufacturer instructions.After extraction, Turbo DNAase (Thermo Fisher Scientific) was used to DNAse treat the samples.NanoDrop Spectrophotometer 2000c (Thermo Fisher Scientific) was used both before and after DNAase treatment to assess the quality and quantity of RNA.About 3 µg of RNA was used for cDNA synthesis, performed as described in (47).
qRT-PCR: EcR expression is very low in non-blood fed females and unreliable for direct assessment of gene knockdown efficiency; therefore, transcript levels of the mosquito lipid transporter Lipophorin (Lp) were used as a reporter for EcR knockdown, because Lp is known to be strongly induced by EcR silencing independent of blood feeding ( 23 targeting Lipophorin (Lp) (AGAP001826) or the ribosomal gene RpL19 (AGAP004422), used as a housekeeping control, were previously published (23).Relative gene expression was determined using the delta-delta-Ct method, normalizing Lp expression against RpL19, and dsEcR samples against dsGFP.

Infections with P. falciparum field isolates
Identification of gametocyte carriers: P. falciparum gametocyte carriers were identified in the villages surrounding Bobo-Dioulasso, as described in (48,49).
Briefly, children 5-13 years old were surveyed for the presence of P. falciparum parasites by fingerprick blood smears, which were analyzed by Giemsa staining and light microscopy.Subjects were asked if they would participate in the study if they were found to have a gametocytemia between 5-40 gametocytes per 1000 leukocytes (or 40-320 gametocytes/µl blood, based on an estimated conversion factor of 8000 leukocytes/µl blood).All persons found to be infected with malaria from this screen were offered treatment regardless of enrollment in the study.

Membrane feeding:
If parental consent was granted, gametocyte carriers were brought to the lab where < 6 ml of venous blood was drawn by a trained technician.
Blood was spun down at 1800 rpm for 5 min at 37°C to separate the red blood cells from the serum, and then the sample serum was removed and replaced with purchased serum from persons naïve to malaria (48,50).After resuspension, the blood was introduced into 20-30 glass membrane feeders heated to 37°C with a Haake water pump.Mosquitoes contained in large cups were allowed to feed through a parafilm membrane sealing the bottom of the glass feeders for 60-120 mins (49,50,51).Remaining sample blood was maintained at 37°C and supplemented into the feeders as needed.Once feeding time was complete, any unfed or partially-fed females were mouth aspirated and removed from the experiment.

Assessing fecundity and pre-gravid status
Since females were maintained as virgin throughout experimentation, oviposition did not occur following blood feeding and fecundity/pre-gravid status could be determined at the same time that P. falciparum midgut infection was assessed 7-8 days pIBM.At this time, ovaries were dissected and developed eggs were counted (dsRNA-injected An. gambiae, An. coluzzii) or females were scored as gravid/pre-gravid (VK5 An. coluzzii).Assessing rMAL-R: 5'-AAAATTCCCATGCATAAAAAATTATACAAA-3'.
Gel electrophoresis was used to evaluable all PCR products and confirm that only P.
falciparum parasites were present in our samples (p1-p6).The resulting PCR products were run on a gel, identifying blocks 4A and 4P as: -MM-, -MK-, -KM-, or -KK-.Combining the typing results from all four blocks (i.e.MMMM, KMMK, RMKK, etc), allows for the identification of up to 24 distinct MSP1 genotypes.These nested PCR reactions were performed on blood sample DNA (p1-p6) to determine which MSP1 genotypes were present in each isolate (S1 Table ).The number of unique MSP1 genotypes identified in each sample was then used to determine the sample's (minimum) complexity of infection (COI) (Table 1).NF54 cultured parasites were used a control for these experiments, as they are clonal and known to contain only a single parasite genotype.

Statistical Analyses
GraphPad work included in this submission.Review the submission guidelines for detailed requirements.View published research articles from PLOS Neglected Tropical Diseases for specific examples.This statement is required for submission and will appear in the published article if the submission is accepted.Please make sure it is accurate.Funded studies Enter a statement with the following details: Initials of the authors who received each award • Grant numbers awarded to each author • The full name of each funder • URL of each funder website • Did the sponsors or funders play any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript?• Did you receive funding for this work?Please add funding details. as follow-up to "Financial Disclosure Enter a financial disclosure statement that describes the sources of funding for the work included in this submission.Review the submission guidelines for detailed requirements.View published research articles from PLOS Neglected Tropical Diseases for specific examples.This statement is required for submission and will appear in the published article if the submission is accepted.Please make sure it is accurate.This work was funded by the National Institutes of Health (NIH) (R01 AI124165, R01 AI104956) and by a Faculty Research Scholar Award by the Howard Hughes Medical Institute (HHMI) and the Bill & Melinda Gates Foundation (BMGF) (Grant ID: OPP1158190) to FC. FC is a HHMI Investigator.The funders had no role in the study design, data collection, data analysis, data interpretation, decision to publish or preparation of the manuscript.The findings and conclusions herein do not necessarily reflect positions or policies of the NIH, HHMI or BMGF.Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation Funded studies Enter a statement with the following details: Initials of the authors who received each award • Grant numbers awarded to each author • The full name of each funder • URL of each funder website • Did the sponsors or funders play any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript?• Did you receive funding for this work?"Please select the country of your main research funder (please select carefully as in some cases this is used in fee calculation).as follow-up to "Financial Disclosure Enter a financial disclosure statement that describes the sources of funding for the work included in this submission.Review the submission guidelines for detailed requirements.View published research articles from PLOS Neglected Tropical Diseases for specific examples.This statement is required for submission and will appear in the published article if the submission is accepted.Please make sure it is accurate.Funded studies Enter a statement with the following details: Initials of the authors who received each award • Grant numbers awarded to each author • The full name of each funder • URL of each funder website • Did the sponsors or funders play any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript?• Did you receive funding for this work?"UNITED STATES -US On behalf of all authors, disclose any competing interests that could be perceived to bias this work.This statement will be typeset if the manuscript is accepted for publication.Review the instructions link below and PLOS NTDs' competing interests policy to determine what information must be disclosed at submission.Data Availability Provide a Data Availability Statement in the box below.This statement should detail where the data used in this submission can be accessed.This statement will be typeset if the manuscript is accepted for publication.Before publication, authors are required to make all data underlying their findings fully available, without restriction.Review our PLOS Data Policy page for detailed information on this policy.Instructions for writing your Data Availability statement can be accessed via the Instructions link below.The datasets generated and analyzed in the current study will be available in the Harvard Dataverse repository under the identifier https://doi:10.7910/DVN/AVVLVIPowered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation Abstract (<300 words): Fig).We next counted oocyst numbers and found that EcR silencing had no effect on the Fig).
) and p6 in An. gambiae (Fig 4B).It is important to note that p6 parasites infected the same batch of An. gambiae mosquitoes as p5, which instead showed the characteristic size increase in response to EcR silencing (Fig 4B).As p5 and p6 differ in both the number and combination of genotypes they contain ( oocysts per blood feeding event (S4B Fig).The pre-gravid females were slightly less likely to become infected than gravid females (pie charts in Fig 4D), yet among infected individuals, there was no difference in oocyst intensity (Fig 4D).We next determined oocyst size, for practical reasons at 7 days pIBM, one day earlier than in previous experiments.Strikingly, oocysts were significantly larger in pre-gravid females as compared to females that produced an egg batch, with a 48% size increase (Fig 4E)(S4 Table).Although limited in number, the same trend was observed for the colony mosquitoes (S4C Fig).
gambiae, such that disruption of 20E signaling resulted in accelerated oocyst growth in both species (Fig 2A-C) and across several parasite isolates (Fig 4A-B).In An. gambiae, dsEcR treatment also resulted in a greater number of salivary gland sporozoites at the relatively early time point of 12 days piBM (Fig 2D, S3 Figure).While additional studies assessing sporozoite load across time are still needed to precisely quantify the effects onthe EIP, these data confirm that impairing the processes regulating mosquito oogenesis can result in earlier arrival of sporozoites to the salivary glands.Given the lengthy sporogonic cycle of P. falciparum (minimum 9 days (5, 6, 7)) coupled with the relatively limited lifespan of mosquitoes in the field (~2-3 weeks(10,11,12 An. gambiae (Fig 3A-B).Regardless, disrupting 20E signaling resulted in significantly larger oocysts in both anophelines (Fig 2A-C), demonstrating that 20E regulated pathways have a conserved influence on parasite growth.While most parasite isolates responded to impaired 20E signaling by accelerating development, the comparison between p5 and p6 in the same batch of An. gambiae revealed that p6 parasites did not increase in size (Fig 4B).Furthermore, p3 oocysts did not respond to EcR silencing in An. coluzzii (Fig 4A), yet they had a large growth response in An. gambiae (Fig 4B).Taken together, these results suggest that mosquitoparasite genotype-by-genotype interactions may regulate sporogonic development, as postulated for dengue virus EIP in Aedes (35).This conclusion is further supported by our GLMM analyses, which revealed that mosquito genotype (dsGFP An. gambiae vs dsGFP An. coluzzii) (Fig 4C) and parasite isolate (p1-p6 in dsEcR females) (Fig 4A-B) can EcR-silenced mosquitoes (An.gambiae and An.coluzzii) (Fig 4A-B, dsEcR), yet we failed to detect any isolate-driven effect on oocyst growth in dsGFP controls (Fig 4A-B, Cntrl) or wild VK5 An. coluzzii (Fig S4D).
) (Fig 2E).To this end, quantitative real-time PCR (qRT-PCR) was performed on a QuantStudio 6 Pro thermocycler (Thermo Fisher Scientific) in 15 µl reactions containing 1X PowerUp SYBR Green Master Mix (Thermo Fisher Scientific), primer dilutions, and 5 µl of sample cDNA diluted 1:10.Primers (Integrated DNA Technologies) Extracted sample DNA was used to determine the MSP1 genotype(s) present in each sample by nested PCR as described by(31).This nested PCR approach was used to identify the allelic type of 4 variable regions on the MSP1 gene termed block 2, 4A, 4P, and 6.Each block can either have a MAD20 or K1 allelic family type, with the exception of block 2, which can alternatively have an RO33 type.The first PCR reaction was used to type blocks 2 and 6.To do this, the following primers (3 forward; 2 reverse) were used in six unique combinations for each then run on a gel in order to identify blocks 2 and 6 as: M--M, M--K, K--M, K--K, R--M, or R--K.PCR products from the first reaction were then used in subsequent nested PCRs where blocks 4A and 4P were typed.This second round of PCRs was done using the following primers (2 forward; 2 reverse) in four unique combinations:

Fig 1 :
Fig 1: An. coluzzii and An.gambiae females were infected with P. falciparum field

Fig 2 :
Fig 2: Oocyst growth is accelerated after dsEcR treatment in both An.coluzzii and

Fig 3 :
Fig 3: P. falciparum oocyst growth is negatively linked to egg development.(A) In

Fig 4 :
Fig 4: Individual P. falciparum isolates vary in their growth response to EcR-

Table 1 ,
S3 Table), these observations suggest that parasite genetic factors may influence the interactions between mosquito oogenesis and P. falciparum growth rates.Interestingly, we observed that p3 parasites did not respond to EcR silencing in An. coluzzii (Fig 4A), yet they increased in size in An. gambiae (Fig 4B), emphasizing that the same parasite isolate can behave differently in different Anopheles species.The only other parasite isolate tested in both species, p1, responded robustly to EcR silencing in both vectors.With the caveat that we could perform only two direct comparisons between

Table 1 : Gametocytemia and COI for P. falciparum field isolates
Prism 8, JMP Pro 14, and R were used to analyze the data with a significance cut-off of p = 0.05.Oocyst size data was analyzed using Generalized Linear Mixed Effects Models (GLMM) and Likelihood Ratio Tests (LRT), allowing for nesting of individual oocyst measurements by mosquito, and assessment of both fixed and random effects.normallydistributeddata was analyzed by either Mann-Whitney or Kruskal-Wallis.Prevalence data (for oocysts and sporozoites) was analyzed using Fisher's Exact test.Specific statistical tests and p values are indicated in the text, figures, and figure legends.All graphs display mean values with standard error of the mean (SEM).
by either unpaired t-test or one-way ANOVA (Welch's correction when appropriate).Non-