elifeeLifeeLife2050-084XeLife Sciences Publications, Ltd9098910.7554/eLife.9098910.7554/eLife.90989.3Research ArticleMicrobiology and Infectious DiseaseQuantification of sporozoite expelling by Anopheles mosquitoes infected with laboratory and naturally circulating P. falciparum gametocytesAndolinaChiara1†GraumansWouterhttps://orcid.org/0000-0003-3952-64911†GuelbeogoMoussa2van GemertGeert-Janhttps://orcid.org/0000-0002-2740-53371RamijthJordache1HarounaSoré2SoumanabaZongo2StoterRianne1Vegte-BolmerMarga1PangosMartina3SinnisPhotini4CollinsKatharine1StaedkeSarah G5TionoAlfred B2DrakeleyChrishttps://orcid.org/0000-0003-4863-075X6LankeKjerstin1BousemaTeunhttps://orcid.org/0000-0003-2666-094Xteun.bousema@radboudumc.nl16https://ror.org/05wg1m734Department of Medical Microbiology, Radboud University Nijmegen Medical CentreNijmegenNetherlandshttps://ror.org/03y3jby41Centre National de Recherche et de Formation sur le PaludismeOuagadougouBurkina FasoDepartment of Plastic and Reconstructive Surgery, Azienda Ospedaliero Universitaria GiulianoIsontina TriesteTriesteItalyDepartment of Molecular Microbiology and Immunology, Johns HopkinsBloomberg School of Public HealthBaltimoreUnited Stateshttps://ror.org/03svjbs84Liverpool School of Tropical MedicineLiverpoolUnited Kingdomhttps://ror.org/00a0jsq62Department of Immunology and Infection, London School of Hygiene and Tropical MedicineLondonUnited KingdomLevashinaElena AReviewing Editorhttps://ror.org/0046gcs23Max Planck Institute for Infection BiologyGermanySoldati-FavreDominiqueSenior Editorhttps://ror.org/01swzsf04University of GenevaSwitzerland
It is currently unknown whether all Plasmodium falciparum-infected mosquitoes are equally infectious. We assessed sporogonic development using cultured gametocytes in the Netherlands and naturally circulating strains in Burkina Faso. We quantified the number of sporozoites expelled into artificial skin in relation to intact oocysts, ruptured oocysts, and residual salivary gland sporozoites. In laboratory conditions, higher total sporozoite burden was associated with shorter duration of sporogony (p<0.001). Overall, 53% (116/216) of infected Anopheles stephensi mosquitoes expelled sporozoites into artificial skin with a median of 136 expelled sporozoites (interquartile range [IQR], 34–501). There was a strong positive correlation between ruptured oocyst number and salivary gland sporozoite load (ρ = 0.8; p<0.0001) and a weaker positive correlation between salivary gland sporozoite load and number of sporozoites expelled (ρ = 0.35; p=0.0002). In Burkina Faso, Anopheles coluzzii mosquitoes were infected by natural gametocyte carriers. Among salivary gland sporozoite positive mosquitoes, 89% (33/37) expelled sporozoites with a median of 1035 expelled sporozoites (IQR, 171–2969). Again, we observed a strong correlation between ruptured oocyst number and salivary gland sporozoite load (ρ = 0.9; p<0.0001) and a positive correlation between salivary gland sporozoite load and the number of sporozoites expelled (ρ = 0.7; p<0.0001). Several mosquitoes expelled multiple parasite clones during probing. Whilst sporozoite expelling was regularly observed from mosquitoes with low infection burdens, our findings indicate that mosquito infection burden is positively associated with the number of expelled sporozoites. Future work is required to determine the direct implications of these findings for transmission potential.
gametocytesporozoiteanophelessporogonyexpellingResearch organismP. falciparumhttp://dx.doi.org/10.13039/100019180HORIZON EUROPE European Research CouncilERC-CoG 864180AndolinaChiaraGraumansWouterramijthJordacheBousemaTeunAMMODO Science Award2019BousemaTeunhttp://dx.doi.org/10.13039/100000002National Institutes of HealthAI132359SinnisPhotiniThe funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.Author impact statementThe number of sporozoites expelled by mosquitoes that are infected with lab-cultured or naturally circulating Plasmodium falciparum gametocytes is heterogeneous and associated with the sporozoite load in their salivary glands.publishing-routeprcIntroduction
Malaria transmission to mosquitoes depends on the presence of mature gametocytes in human peripheral blood that are ingested by a mosquito during blood feeding. Ingested parasites undergo several developmental transformations in a process called sporogony. After ingestion, gametocytes transform into male and female gametes that fuse to form a zygote. The zygote differentiates into a motile ookinete that penetrates the midgut epithelium to form an oocyst. Multiple rounds of mitotic replication result in the formation of sporozoites inside an oocyst. Upon oocyst rupture, sporozoites are released into the haemocoel and invade the salivary glands (Mueller et al., 2010). These sporozoites penetrate the distal portion of the two lateral and medial lobes of the glands and accumulate extracellularly inside secretory cavities before entering the salivary ducts (Wells and Andrew, 2019). Despite the large number of sporozoites in the cavities, only a small proportion pass through the proximal part of the lobes where the salivary ducts become narrow (Frischknecht et al., 2004; Bradley et al., 2018), and only tens or low hundreds of sporozoites are assumed to be inoculated per mosquito bite (Graumans et al., 2020). Developmental bottlenecks during sporogony, as well as the size of the sporozoite inoculum, remain incompletely understood (Graumans et al., 2020).
The density of gametocytes in human peripheral blood is an important determinant of human to mosquito transmission. Though infections with parasite densities below the microscopic threshold for detection can infect mosquitoes (Schneider et al., 2007; Bousema et al., 2006), the likelihood and infection intensity increase with the number of ingested gametocytes (Bradley et al., 2018). Because of the abundance of low-density gametocyte carriers among infected populations, these are considered important drivers of malaria transmission (Andolina et al., 2021; Slater et al., 2019; Ouédraogo et al., 2016). Importantly, this conclusion is based on the assumption that all infected mosquitoes are equally infectious regardless of oocyst densities.
In apparent support of this assumption, single oocyst infections can result in thousands of salivary gland sporozoites (Rosenberg and Rungsiwongse, 1991; Pringle, 1965). A positive correlation between oocyst densities and salivary gland sporozoites was previously observed in Plasmodium falciparum (Vaughan et al., 1992; Miura et al., 2019) as well as in Plasmodium vivax (Solarte et al., 2011), suggesting that on average of 1000–2000 sporozoites reach the salivary glands per single ruptured oocyst. The few studies that quantified sporozoite inoculum by allowing mosquitoes to salivating into capillary tubes containing mineral oil, sucrose solution, or blood (Beier et al., 1991a; Beier et al., 1991b; Rosenberg et al., 1990) estimated median inocula ranging between 8 and 39 sporozoites with a minority of mosquitoes expelling >100 sporozoites (reviewed in Graumans et al., 2020).
While these studies provide some insights into sporozoite expelling and transmission dynamics, they do not reflect natural feeding conditions. Also, microscopy techniques used to quantify sporozoites may have underestimated the number of sporozoites (Medica and Sinnis, 2005). Subsequent studies with rodent Plasmodium models allowed natural mosquito probing through skin and showed a positive association between sporozoite density in salivary glands and expelled sporozoite numbers (Medica and Sinnis, 2005; Aleshnick et al., 2020). A recent study using a Plasmodium yoelii malaria model demonstrated that mosquitoes with over 10,000 salivary gland sporozoites were 7.5 times more likely to initiate a malaria infection in mice (Aleshnick et al., 2020). This finding has not been replicated for human malarias but is broadly consistent with post hoc analysis of infection likelihood in controlled human malaria infections (CHMI), where only mosquitoes that had >1000 P. falciparum salivary gland sporozoites remaining after probing (the residual sporozoite load) were capable of establishing an infection in malaria-naïve volunteers (Churcher et al., 2017).
If low oocyst/low sporozoite densities in mosquitoes are unlikely to initiate infections in humans, this may have profound consequences for our understanding of transmission (WHO, 2017). If mosquitoes with low-infection burdens have limited transmission potential, then the rationale for targeting low-density infections in humans that give rise to low-infection burdens in mosquitoes (Bradley et al., 2018) may be diminished.
Here, we examined the progression of sporozoite development and the number of sporozoites expelled into artificial skin by individual Anopheles mosquitoes infected either with P. falciparum gametocyte cultures or in experimental infections with naturally circulating parasite strains. We directly assessed the association between oocyst burden, salivary gland infection intensity, and the number of sporozoites expelled.
ResultsLow number of <italic>P. falciparum</italic> sporozoites are quantifiable by qPCR
Multicopy mitochondrial COX-1 and 18S rRNA gene targets were analyzed in octuplicate on serial dilutions of sporozoites to assess qPCR performance and select the target that achieved highest sensitivity and most consistent sporozoite detection. COX-1 outperformed 18S in detecting sporozoites (Figure 1—figure supplement 1), the limit of detection (LOD) and limit of quantification (LOQ) for COX-1 qPCR were determined at 20 sporozoites per sample (8/8 sample positivity with a coefficient of variation <2) (Figure 1A). Next, we confirmed the qPCR performance in combination with the matrix that was used for expelling experiments by spotting serial dilutions of sporozoites in whole-blood on Integra dermal substitute artificial skin (Agostinis et al., 2021) prior to nucleic acid extraction. The matrix had no apparent impact on sporozoite detectability and quantification (Figure 1—figure supplement 1), unlike previously used filter paper matrices (Brugman et al., 2018; Holzschuh and Koepfli, 2022).
The detectability of sporozoites by molecular methods and oocysts by immunolabeling.
(A) qPCR performance for P. falciparum sporozoites. Serial dilutions of sporozoites (x-axis) were prepared in phosphate-buffered saline (PBS) and assessed in octuplicate on a single plate to determine qPCR COX-1 limit of detection and quantification. Dots represent sample cycle threshold (left y-axis), and bars coefficient of variation (right y-axis). For each serial dilution, Ct sample positivity is shown as the percentage of total tested. (B) The relationship between oocyst density versus infection prevalence from 457 membrane feeding experiments using cultured gametocytes. Colors represent regular feeds (red) and those selected for performing experiments on the extrinsic incubation period (EIP; green) or sporozoite-expelling experiments (blue). (C) Immunofluorescence staining with 3SP2-Alexa 488 anti-CSP. Empty sheet (left) and intact oocyst (right). (D) Violin plots of oocyst staining – day 8 post infection by mercurochrome (purple) and – day 18 by 3SP2-Alexa 488 anti-CSP immunostaining (yellow) for cultured parasites. Box plots show interquartile range, whiskers show the 95% intervals.
qPCR performance on extracted <italic>P. falciparum</italic> sporozoites (SPZ) targeting the 18S rRNA gene, and DNA extraction/quantification of <italic>P. falciparum</italic> SPZ from artificial skin.
A serial dilution of SPZ was prepared as described in ‘Materials and methods’ and run in octuplicate to assesses the limit of detection (LOD) and limit of quantification (LOQ) for qPCR targeting 18S. (A) Dots represent sample cycle threshold (left y-axis) and bars the coefficient of variation (right y-axis). For each serial dilution, Ct sample positivity is shown as the percentage of the total number of replicates analyzed. The LOD was set at 100% sample positivity, which was 50 SPZ per sample. The LOQ was set at 100% sample positivity, with a coefficient of variation (COV) <2. For 18S, this was 1000 SPZ per sample (COV 0.71%). (B) Next to test if Integra bovine collagen artificial skin would be suitable as a mosquito feeding membrane to collect expelled SPZ we first tested DNA extraction and quantification from this material. Serial dilutions of SPZ were prepared in phosphate-buffered saline (PBS) and human blood. To compare DNA extraction efficiency, three conditions were tested in duplicate and compared by COX-1 qPCR; SPZ diluted in PBS, SPZ diluted in whole-blood, and SPZ diluted in whole-blood spotted on artificial skin. The amount of SPZ in the sample (x-axis, log 10) plotted against the qPCR Ct value (y-axis) for tested conditions. (C) The amount of SPZ in sample and mean Ct of the three assessed conditions. SPZ serial dilutions in PBS and human blood showed comparable Ct values. SPZ in blood spotted on artificial skin showed slightly lower Ct values, indicating that SPZ can be efficiently extracted from skin without loss of signal. Detection was possible down to approximately five SPZ.
A comparative analysis of oocyst densities using mercurochrome staining and immunostaining
Subsequently, mosquito feeding assays were performed by offering diluted in vitro cultured gametocytes to mosquitoes to obtain a broad range of oocyst densities. The association between log10 oocyst intensity and infection prevalence in mosquitoes was assessed using a logistic regression model (using data from 457 mosquitoes, Figure 1B). Mosquito infection prevalence was strongly associated with oocyst intensity, corroborating earlier work (Churcher et al., 2012), with a strong positive sigmoidal association and a 14.68 (95% CI, 8.18–26.35, p<0.0001) times higher odds of infection prevalence associated with a tenfold higher oocyst density. In this analysis, oocysts were enumerated microscopically following standard mercurochrome staining. We previously used 3SP2-Alexa 488 anti-circumsporozoite (CSP) immunostaining to visualize ruptured and intact oocysts (Stone et al., 2013). The concordance between oocyst prevalence by standard oocyst mercurochrome staining (day 8 post infection [PI]) and anti-CSP immunostaining on day 18 PI (Figure 1C) was investigated. For this, oocyst density distributions by both methods were compared within batches of mosquitoes that were fed on cultured gametocytes during the same standard membrane feeding assay (Figure 1D). We observed no statistically significant difference in oocyst densities determined by day 8 mercurochrome staining (median five oocysts, interquartile range [IQR], 2–20, N = 252) and day 18 immunolabeling (median six oocysts, IQR, 2–14, N = 167; Student’s t-test on log densities, p=0.944).
Highly infected mosquitoes become salivary gland sporozoite positive earlier
Following assay validation, the extrinsic incubation period (EIP) was compared between mosquitoes with low and high oocyst densities. Batches of high- and low-infected mosquitoes were generated using standard membrane feeding assay with standard concentrations of cultured gametocytes or culture material that was five- or tenfold diluted (Figure 1B). On day 8 PI, 20 mosquitoes were dissected and batches that had ≥70% oocyst infection prevalence and means of ≤5 or >20 oocysts were selected for subsequent dissections (Figure 1B). On days 9–11, salivary glands and the remaining mosquito body (that included the mosquito midgut) were collected separately and analyzed for sporozoite density by COX-1 qPCR. Mosquitoes were then binned into four categories of sporozoite infection intensity, defined as the sum of mosquito body and salivary gland sporozoite density (Figure 2A). This total sporozoite density was examined in relation with the likelihood of being salivary gland sporozoite positive and thus having completed sporogonic development.
Extrinsic incubation period in high- versus low-infected mosquitoes.
(A) Total sporozoites (SPZ) per mosquito in body plus salivary glands (x-axis) were binned by infection load <1k; 1k–10k; 10k–50k; >50k and plotted against the proportion of mosquitoes (%) that were SPZ positive (y-axis) as estimated from an additive logistic regression model with factors day and SPZ categories. In total, 120, 120, and 40 mosquitoes were dissected on days 9 (blue), 10 (dark green), and 11 (light green), respectively. Error bars show the 95% confidence intervals. (B) Violin plots of SPZ density in single oocysts dissected on days 9 (purple) and 10 (green). The box indicates the interquartile range (IQR) (Q1 and Q3 quartiles) and the median. Lines extending Q1 and Q3 quartiles indicate the range of the data within 1.5× IQR.
On day 9 PI, 54.3% of highly infected mosquitoes (>50,000 sporozoites) were salivary glands sporozoite positive (Figure 2A) and had 3.17 (95% CI 95%, 0.7–14.4, p=0.4278) times the odds of being salivary gland positive compared to low-infected mosquitoes with <1000 sporozoites (27.3% salivary gland sporozoite positive). By day 10 PI, 82.2% of mosquitoes with 10,000–50,000 (10k–50k) sporozoites were salivary glands positive and had 11.56 (95% CI, 1.83, 73.25, p=0.0449) times the odds of being salivary glands positive compared to low-infected mosquitoes with <1000 sporozoites (28.6% salivary glands sporozoite positive). On day 11 PI, all 15 highly infected mosquitoes (>50,000 sporozoites) were salivary glands sporozoite positive and meaningful odds ratios and 95% CIs could not be determined. When considering the entire period over which EIP experiments were conducted, mosquitoes with >50,000 sporozoites had a 13.44 times higher odds of being salivary glands positive compared to low-infected mosquitoes (<1000 sporozoites; 95% CI, 4.02–44.88, p<0.0001). Mosquitoes harboring 10,000–50,000 sporozoites had a 5.98 times higher odds of being salivary glands positive compared to low-infected mosquitoes (95% CI, 1.88–19.07, p=0.0119). These data demonstrate that EIP is shorter in high-infected compared to low-infected mosquitoes in a temperature- and humidity-controlled environment.
Sporozoite densities increase with oocyst age
To quantify the number of sporozoites per oocyst, individual oocysts were isolated from midguts on days 9 and 10 PI (Soontarawirat et al., 2017) and stained with 1% mercurochrome. The median sporozoite density was 10,485 (IQR, 9171.3–12,322.5; 12 examined mosquitoes) per oocyst for day 9 and 15,390 (IQR, 10,600–20,887, 19 examined mosquitoes) for day 10 (Figure 2B, p=0.04995, by Welch’s two-sample t-test on log10-transformed data).
Oocyst density, salivary gland density, and the size of the sporozoite inoculum are positively associated in mosquitoes infected with cultured gametocytes
We performed artificial skin feeding experiments with individual mosquitoes on day 15 PI to assess sporozoite expelling. To avoid interference of residual blood with oocyst immunolabeling, we assessed mosquito oocyst density (ruptured and intact oocysts) and sporozoite density in the salivary glands on day 18, allowing 3 d for bloodmeal digestion. This approach allowed us to determine the density of intact and ruptured oocysts and associate this to sporozoite density in the same mosquito. It was noted that a minority of oocysts failed to rupture during this time span; 5% (93/1854) of all oocysts were visually intact and 88.3% (166/188) of examined mosquitoes had at least one unruptured oocyst on day 18. While we observed good concordance between oocyst densities by mercurochrome staining on day 8 and immunostaining on day 18 PI (Figure 1D), oocysts sporadically did not take up the 3SP2-Alexa 488 anti-CSP antibody labeling. In 54% (12/22) of mosquitoes without evidence of ruptured oocysts, we observed salivary gland sporozoites. Nevertheless, there was a strong positive association observed between ruptured oocysts and salivary gland sporozoite load (ρ = 0.80, p<0.0001; N = 185) (Figure 3B). When intact oocysts were also included, this association was nearly identical (ρ = 0.80, p<0.0001; N = 185). We estimated a median of 4951 (IQR, 3016–8318) salivary gland sporozoites per ruptured oocyst in Anopheles stephensi. Next, the association between sporozoite density and the number of expelled sporozoites (inoculum size) was determined. For this, individual mosquitoes in miniature cages were allowed to probe for a maximum of 8 min on blood-soaked artificial skin. In preparation of expelling experiments, we assessed if sporozoites may migrate away from the examined area of artificial skin (surface feeding area 201 mm2) using a batch of high-infected mosquitoes (35,6 oocyst mean). For this, we examined the part of the artificial skin normally unexposed to mosquitoes beneath the rubber band. In three experiments with either 5 (N = 2) or 20 mosquitoes (N = 1) probing and feeding on the artificial skin, we detected a total of 3264, 3599, and 10,340 expelled sporozoites in the exposed skin part, respectively, but no qPCR signal in the unexposed part of artificial skin. This indicates no or negligible migration of sporozoites to the outer sides of the skin and affirms that our procedure reliably captures all expelled sporozoites.
Sporozoite expelling in relation to infection burden in <italic>Anopheles stephensi</italic> mosquitoes infected with cultured gametocytes.
(A) Binning of mosquitoes by total sporozoite load and expelling prevalence (N = 186). (B) The number of ruptured oocysts stained by 3SP2-Alexa 488 anti-CSP and fluorescent microscopy (x-axis) in relation to total salivary gland sporozoite density (y-axis), assessed by COX-1 qPCR; ρ = 0.80 (CI, 0.74–0.85, p<0.0001). The red dot indicates a mosquito which had 9 ruptured oocysts but only 126 residual salivary glands sporozoites while it expelled 1567 sporozoites. Considering the high number of ruptured oocysts in the midgut, it is possible that some lobes of salivary glands were missed during dissection and sporozoite load was underestimated by qPCR. (C) Total sporozoite density (residual salivary gland sporozoites + sporozoites expelled, x-axis) in relation to the number of expelled sporozoites (y-axis) by COX-1 qPCR ρ = 0.35 (CI, 0.17–0.50, p=0.0002). The dotted line on the x-axis shows the threshold of qPCR detection of 20 sporozoites. The line represents the fitted linear regression line, the intercept is forced to zero for biological plausibility, and the gray shaded area is the 95% CI.
Source data include re-analyses of correlation coefficients when different thresholds for data inclusion are imposed to determine whether associations are robust.
Sporozoite expelling in relation to infection burden.
Total sporozoite density (residual salivary gland sporozoites + sporozoites expelled, x-axis) in relation to the number of expelled sporozoites (y-axis) by COX-1 qPCR (ρ = 0.016, 95% CI, –0.12 to 0.16, p=0.8321). This figure includes 26 observations from mosquitoes that did not expel any sporozoites; these observations were excluded from the main text. The line represents the fitted linear regression line, the intercept is forced to zero for biological plausibility, and the shaded area is the 95% CI.
Among all mosquitoes used in skin feeding experiments, 53% (116/216) expelled sporozoites at any density, and 45% (97/216) expelled sporozoites above our threshold for reliable detection and quantification of 20 sporozoites/skin. In line with previous work with rodent malaria species Plasmodium berghei (Matsuoka et al., 2002), sporozoite expelling was observed in mosquitoes that did not take a bloodmeal; 33% (5/15) of mosquitoes that probed but failed to take a bloodmeal expelled sporozoites (sporozoite range 5–1802). To examine sporozoite expelling in relation to infection burden, mosquitoes were binned into four categories of salivary gland infection intensity that was estimated by combining the residual sporozoite load in the salivary glands and the sporozoites successfully expelled into the skin. In this way also heavily infected mosquitoes that expelled the majority of sporozoites were categorized as heavily infected. We observed a strong positive association between oocyst sheets and total salivary gland sporozoite load (Spearman’s correlation coefficient ρ = 0.80, p<0.0001; N = 111; Figure 3B). When examining this association for different ranges of oocyst intensity (<5, <10, <20 oocysts), correlation estimates remained highly similar and statistical significance was retained (Figure 3—figure supplement 1).
We observed no statistically significant association between salivary gland infection intensity and the prevalence of expelling sporozoites (Figure 3A; 95% CI, 0.74–0.85; p=0.1880). Among mosquitoes that expelled sporozoites, the medians of expelled and residual salivary gland sporozoites were 136 (IQR, 34–501) and 23,947 (IQR, 9127–78,380), respectively, while the highest number of sporozoites detected in skin was 4166. We observed a weak but statistically significant positive association between total sporozoite load and the number of expelled sporozoites (ρ = 0.35, 95% CI, 0.17–0.50; p=0.0002; N = 112; Figure 3C). When examining this association for different ranges of total sporozoite load (<10,000; <50,000; <100,000 sporozoites), correlation estimates remained highly similar although this correlation lost statistical significance when only including low total sporozoite loads <10,000 sporozoites (ρ = 0.29; 95% CI, –0.07 to 0.58, p=0.1094) (Figure 3—source data 1). When we included 26 observations from mosquitoes that did not expel any sporozoites, we observed no statistically significant association between total sporozoite load and the number of expelled sporozoites (ρ = 0.016, 95% CI, –0.12 to 0.16; p=0.8321).
We observed no evidence for a sharp increase in sporozoite expelling at sporozoite densities ≥10,000, as was previously described in rodent malaria models Aleshnick et al., 2020; 28% (53/186) of our mosquitoes harbored sporozoites below this density. Among these low-infected mosquitoes, 64% (34/53) expelled sporozoites and the median number of expelled sporozoites was 67 (IQR, 13–128).
Infected mosquitoes in Burkina Faso show comparable correlations between oocyst density, salivary gland density, and sporozoite inoculum
Seven gametocyte donors (age 5–15 y; median 48 gametocytes/µl [range 40–167]) were recruited in Balonghin, Burkina Faso. Their blood was offered to locally reared Anopheles coluzzii via membrane feeders at the gametocyte density observed and following gametocyte enrichment by magnetic activated cell-sorting whereby gametocyte concentration was increased by approximately three- to fourfold. Five donors infected mosquitoes in at least one of these two conditions (Figure 4A); as expected, mosquito infection rates and oocyst densities were significantly increased after gametocyte enrichment. Mosquito batches with ≥50% infection prevalence were used for artificial skin feeding experiments as described above. From a total of 53 mosquitoes, salivary gland sporozoites were detected in 69% (37/53). Also, 31 mosquito midguts were available for immunolabeling, of which 67.7% (23/31) had oocyst sheets detected by immunolabeling on day 19 and nearly half of these (48% [15/31]) had <5 oocyst sheets. Two mosquito midguts were negative by immunolabeling while their residual salivary gland sporozoite loads were 7958 and 14,750 respectively, suggesting oocyst staining failure or rupture followed by complete oocyst disappearance. 87.8% (370/421) of all detected oocysts ruptured while 39% (9/23) of oocyst positive mosquitoes harbored at least one intact oocyst (range 1–19). Failure to rupture was uncommon in low-infected mosquitoes (≤5 oocysts, N = 80) where only two midguts had one intact oocyst. Among 53 mosquitoes used for artificial skin feeding on day 16 PI, 89% (33/37) salivary gland sporozoite-positive mosquitoes with a median of 45,100 residual salivary gland sporozoites (IQR, 20,310–164,900) expelled a median of 1035 sporozoites (IQR, 171–2969). We estimated a median of 6350 (IQR, 4225–8475) salivary gland sporozoites per ruptured oocyst in these experiments. Three skin samples were positive by COX-1 qPCR (range 1–64) while salivary glands were negative, suggesting either all sporozoites were expelled or a technical failure in DNA extraction from the salivary glands. AMA-1 amplicon sequencing was conducted on extracted salivary glands and artificial skins for mosquitoes that expelled sporozoites at any quantity. Among 22 mosquitoes that were infected from three gametocyte donors, we identified 10 unique clones; 68.2% (15/22) of the infected mosquitoes that were tested harbored multiclonal infections. Following probing and successful sporozoite expelling, 10 skin samples contained more than one P. falciparum clone (45.5%; 10/22) (Figure 5).
Sporozoite expelling in relation to infection load in <italic>Anopheles coluzzii</italic> mosquitoes infected by naturally acquired gametocyte infections in Burkina Faso.
(A) Direct feeding (blue) vs magnetic-activated cell sorting (MACS; green). Bars show the infection prevalence for each of the seven gametocyte carriers. Scatter plots with median lines show the midgut oocyst density as a result of direct feeding (blue) and MACS (green). (B) Binning of total sporozoite load and expelling prevalence (N = 25). (C) Scatter plot of absolute numbers of ruptured oocyst (sheet) density assessed by fluorescent microscopy vs total salivary gland sporozoite density assessed by COX-I qPCR; ρ = 0.90 (95% CI, 0.80–0.95). The line represents the fitted linear regression line and the gray shaded area is the 95% CI. (D) Scatter plot of absolute numbers of total sporozoite density (residual salivary gland sporozoites + sporozoites expelled) and sporozoites expelled into the artificial skin assessed by COX-I qPCR; ρ = 0.70 (CI, 0.52–0.82). The line represents the fitted linear regression line, the intercept is forced to zero for biological plausibility, and the gray shaded area is the 95% CI.
Source data include re-analyses of correlation coefficients when different thresholds for data inclusion are imposed to determine whether associations are robust.
Clonal complexity of <italic>P.</italic> <italic>falciparum</italic> infections in salivary glands and artificial skins following probing by mosquitoes infected by gametocyte carriers who were naturally infected in Burkina Faso.
Clonal data for three donors: 01646 IR (top row), 01661 NJ (middle row), and 011690 CC (bottom row). Left panel shows scatter plots for the association between sporozoite salivary gland load and expelled sporozoites in skin, with a Spearman’s rank correlation across samples from each donor. Right panel shows clonal data for each donor sample in a heatmap plot. The colored numbers on y-axes correspond to the color of the sample in the scatter plot. Purple indicates the presence of a clone in the salivary gland only, turquoise indicates the presence of a clone in the skin only, and green indicates the presence of a clone in both salivary gland and skin.
During the artificial skin feeding, 30% (16/53) of probing mosquitoes did not ingest blood, of which 68% (11/16) expelled sporozoites (range 1–11,970). There appeared to be a trend toward higher prevalence of expelling with increasing sporozoite density (Figure 4B). There was a strong association between ruptured oocyst density and total salivary gland sporozoite density (ρ = 0.84, 95% CI, 0.80–0.95; p<0.0001) (Figure 4C); when intact oocysts were also included, the association was very similar (ρ = 0.86, p<0.0001; N = 30). When examining these associations for different ranges of oocyst intensity (<5, <10, <20 oocysts), correlation estimates remained highly similar and statistical significance was retained (Figure 4—source data 1). There was also a strong positive association between total sporozoite load and the number of sporozoites expelled (ρ = 0.71, 95% CI, 0.52–0.82; p<0.0001) (Figure 4D). When examining this association between total sporozoite load and expelling for different ranges of total sporozoite load (<10,000, <50,000, <100,000 sporozoites), correlation estimates remained similar and statistically significant (Figure 4—source data 1).
Discussion
We examined P. falciparum sporogony in different mosquito species by employing sensitive molecular and staining techniques in conjunction with mosquito probing experiments on artificial skin. Through the visualization of ruptured oocysts and the simultaneous quantification of sporozoites that are expelled by individual mosquitoes, we observed that 95% of oocysts rupture to release sporozoites and that higher salivary gland sporozoite load is associated with shorter time to colonization of salivary glands and larger inoculum size.
The proportion of infectious mosquitoes is a central component of malariometric indices both in terms of quantifying the force of infection and the human infectious reservoir. The entomological inoculation rate (EIR) is defined as the number of infectious bites per person per time unit and is the product of human biting rate and the proportion of sporozoite-positive mosquitoes (Tusting et al., 2014; Shaukat et al., 2010). While EIR is a common measure of human malaria exposure, mosquito infection prevalence is used in this calculation and thus assumes all sporozoite-positive mosquitoes are equally infectious (Smith et al., 2012). Similarly, assessments of the human infectious reservoir for malaria typically take the number of oocyst-positive mosquitoes as a measure of transmission (Ouédraogo et al., 2016; Gonçalves et al., 2017) and thereby not only assume that all oocysts will lead to salivary gland sporozoites but also that all oocyst-positive mosquitoes have equal transmission potential. Recent work with a rodent malaria model challenged this central assumption (Aleshnick et al., 2020); we provide the first direct evidence for P. falciparum that sporozoite burden may indeed be a relevant determinant in efficient sporozoite expelling.
In this study, we assessed sporozoite expelling by mosquitoes carrying low- and high-infection burdens. Our findings confirm that the vast majority (~95%) of oocysts rupture to release sporozoites. This estimate is higher than a previous study with cultured gametocytes (~72%) (Stone et al., 2013) that did not provide a second bloodmeal that may accelerate oocyst maturation (Shaw et al., 2020); a second bloodmeal also better mimics natural feeding habits where multiple bloodmeals are taken within the period required for sporogony. Moreover, we observed a strong positive association between sporozoite salivary gland load and ruptured oocyst density in mosquitoes infected with both cultured and naturally circulating gametocytes, with similar median numbers of 4951 (IQR, 3016–8318) and 6350 (IQR, 4225–8475) salivary gland sporozoites per ruptured oocyst in An. stephensi and An. coluzzii, respectively. Previous studies have shown that despite the substantial release of sporozoites per oocyst, only a proportion of sporozoites successfully reach the salivary glands (Rosenberg and Rungsiwongse, 1991; Hillyer et al., 2007). In line with this, we observed considerably higher sporozoite estimates in intact oocysts (10,485–15,390 sporozoites per oocyst). Whilst these sporozoite estimates are higher than commonly reported (Graumans et al., 2020), an early study from the 1960s reported microscopy-detected sporozoite densities above 9000 sporozoites per oocyst (Pringle, 1965) and a recent study using qPCR similarly observed up to 12,583 sporozoites from a single oocyst (Wang et al., 2018). These high sporozoite numbers per oocyst also mean that, different from P. yoelii (Aleshnick et al., 2020) and P. berghei (Churcher et al., 2017), only a minority of infected mosquitoes had sporozoite densities below the threshold values previously reported in relation to a very low likelihood of achieving secondary infections. Even among infected An. stephensi mosquitoes with only one ruptured oocyst, 92.9% (26/28) had >1000 and 35.7% (10/28) had >10,000 sporozoites in their salivary glands. In our experiments in natural gametocyte carriers in Burkina Faso, only two mosquitoes were observed with single oocysts and both had >10,000 salivary gland sporozoites. Although P. falciparum sporozoite densities below these lower thresholds were thus uncommon, we observed that 39% of sporozoite-positive mosquitoes across a wide range of infection densities failed to expel sporozoites upon probing. This finding broadly aligns with an earlier study of Medica and Sinnis that reported that 22% of P. yoelii-infected mosquitoes failed to expel sporozoites (Medica and Sinnis, 2005). For highly infected mosquitoes, this inefficient expelling has been related to a decrease in apyrase in the mosquito saliva (Rossignol et al., 1984; Thiévent et al., 2019).
Importantly, we observed a positive association between salivary gland sporozoite density and the number of expelled sporozoites. For unknown reasons, this association was markedly stronger in experiments where An. coluzzii mosquitoes were infected using blood from gametocyte carriers who were naturally infected with P. falciparum in Burkina Faso compared to An. stephensi mosquitoes infected with cultured gametocytes. In our experiments with natural gametocyte carriers, sporozoite density appeared associated with both the likelihood of expelling any sporozoites and the inoculum size. The most heavily infected mosquitoes expelled ~14-fold higher sporozoite numbers compared to the lowest quartile. In addition, we found evidence that the EIP – the period required for a mosquito to become salivary gland sporozoite positive – is shorter for heavily infected mosquitoes. Previous studies have not found consistent effects of parasite burden on EIP (Guissou et al., 2023; Stopard et al., 2021) but, unlike our experiments, were also not specifically designed to examine this association across a broad range of infection intensities. Our associations of a shorter EIP in highly infected mosquitoes and, in separate experiments, a larger sporozoite inoculum size for highly infected mosquitoes make mosquito infection intensity a plausible factor in determining onward transmission potential to humans. Heavily infected mosquitoes may be infectious sooner and be more infectious. On the other side of the infection spectrum, it is conceivable that submicroscopic gametocyte carriers that typically result in low oocyst burdens in mosquitoes give rise to infected mosquitoes with a reduced transmission potential. This would greatly reduce their importance for sustaining malaria transmission and make it less important to identify individuals with low parasite densities for malaria elimination purposes (Slater et al., 2015). At the same time, we occasionally observed high sporozoite inocula from mosquitoes with low infection intensity. This argues against a clear threshold sporozoite or oocyst density below which mosquitoes are truly irrelevant for transmission. Moreover, the high number of sporozoites per oocyst makes it plausible that even the low oocyst burdens that are typically observed from asymptomatic parasite carriers (in the range of 1–5 oocysts/infected gut; Andolina et al., 2021; Gonçalves et al., 2017) would be sufficient to result in salivary gland sporozoite loads that are sufficient to result in secondary infections.
Our study leaves a number of questions and has several limitations. While the use of two mosquito and gametocyte sources was a relevant strength of our study, an uncertainty relates to the choice of artificial skin that has a realistic 1.33 mm thickness but is arguably less natural than microvascularized skin with all the natural cues for mosquito probing. Whilst genuine skin might have improved natural feeding behavior, probing and blood feeding were highly efficient in our model and we see no reasons to assume bias in the comparison between high- and low-infected mosquitoes. Our assessments of EIP and sporozoite expelling did not demonstrate the viability of sporozoites. Whilst the infectivity of sporozoites at different time points PI has been examined previously (Yang et al., 2017), these experiments have never been conducted with individual mosquitoes. In vitro experiments that aim to determine the infectivity of single mosquito bites would ideally retain the skin barrier that may be a relevant determinant for invasion capacity and use primary hepatocytes. These experiments were beyond the scope of the current work and would also not provide conclusive evidence on the likelihood of achieving secondary infections. Given striking differences in sporozoite burden between different Plasmodium species – low sporozoite densities appear considerably more common in mosquitoes infected with P. yoelii and P. berghei (Graumans et al., 2020; Churcher et al., 2017; Ahmed et al., 2021), and the inherent limitations of in vitro studies – the association between sporozoite inoculum size and the likelihood of achieving secondary infections may be best examined in controlled human infection studies. CHMI experiments can be specifically designed to estimate the likelihood that probing by (individual) high- and low-infected mosquitoes results in blood-stage infection in malaria-naïve volunteers. Whilst laborious, CHMI studies are unique in allowing definitive evidence on the possible differences in infectiousness between high- and low-infected mosquitoes.
In conclusion, we observed that the majority of oocysts rupture and contribute to salivary gland infection load. We further observe that this sporozoite load is highly variable and an important determinant of the number of sporozoites that is expelled into the skin upon probing.
Materials and methods<italic>P. falciparum</italic> in vitro culture and mosquito infection
P. falciparum gametocytes, NF54 (West Africa) and NF135 (Cambodia) (Conway et al., 2000; Teirlinck et al., 2013) were cultured in an automated culture system (Ponnudurai et al., 1982) and maintained as previously described at Radboudumc, Nijmegen (van de Vegte-Bolmer et al., 2021). An. stephensi mosquitoes, Nijmegen Sind-Kasur strain (Feldmann and Ponnudurai, 1989), were reared at 30°C and 70–80% humidity with a 12 hr reverse day/night cycle. To have a range of infection intensities in mosquitoes, undiluted and diluted cultured gametocytes (0.3–0.5% gametocytes) were generated in heparin blood. In total, 100–150, 1- to 3-day-old female mosquitoes were fed using glass membrane mini-feeders (Ponnudurai et al., 1989).
Mosquito feeding on gametocyte carriers who were naturally infected with <italic>P. falciparum</italic>
P. falciparum gametocyte donors were recruited at schools in Saponé Health District, 45 km southwest of Ouagadougou. Following informed consent, finger-prick blood was examined for gametocytes by counting against 500–1000 white blood cells (WBCs) in thick-blood films. The gametocyte counts were done by two independent microscopists and expressed as density/µl by assuming 8000 WBCs per µl of blood. If gametocyte densities were above 16/µl, 2–5 ml of venous blood was drawn by venepuncture in lithium heparin tubes (BD Vacutainers, ref 368496) and transported to the Centre National de Recherche et de Formation sur le Paludisme (CNRFP) insectary in Ouagadougou in thermos flasks filled with water at 35.5°C (Soumare et al., 2021).
Then, 700 µl of whole blood was used for immediate feeding, which was performed as described elsewhere, using 3- to 5-day-old An. coluzzii mosquitoes per glass mini feeder (Coelen Glastechniek, the Netherlands) that was attached to a circulating water bath set up at 38°C (Isotemp, Fisher Scientific) (Ouédraogo, 2013; Musiime et al., 2019). Fifty mosquitoes per cup (starved for 12 hr) were allowed to feed in the dark for 15–20 min through a Parafilm membrane. To increase mosquito infection prevalence and intensity, 1 ml of blood was used to enrich gametocytes with magnetic cells sorting columns (MACS), as described by Graumans et al., 2019. In this process, a prewarmed 23G hypodermic needle (Becton Dickinson, Franklin Lakes, NJ) was attached to an LC column on a QuadroMACS separator (Miltenyi Biotec, UK) that was placed inside a temperature-monitored incubator (37°C). The flow-through was collected in a 15 ml Falcon tube. Following hydration with 1 ml RPMI, 1 ml donor blood (lithium heparin) was added to MACS column. The column was rinsed with 2 ml warm RPMI. The 15 ml Falcon tube was replaced with a new tube, and the needle was removed. Enriched (bound) gametocytes were washed off the column with 4 ml of warm RPMI, the plunger was used to press the last 1 ml of medium through the column. The two 15 ml Falcon tubes, one with the blood mixed to RPMI and the second with the gametocyte suspension, were spun down at 2000 rpm for 5 min in a temperature-controlled centrifuge at 37°C (Eppendorf 5702R). RPMI supernatant was removed from both tubes with 3 ml disposable Pasteur pipettes. To prepare the mosquito bloodmeal, the small visible pellet of concentrated gametocytes (~30 µl) was resuspended with 150 µl of prewarmed malaria-naïve serum (Sanquin Bloodbank, Nijmegen, the Netherlands) with a blunt needle and 200 µl of the patients’ packed red cells were added and mixed. About 350 µl gametocyte enriched bloodmeal was added to a water-jacket glass feeder as described above.
Mosquito husbandry and oocyst detection by mercurochrome staining
In both insectaries, at Radboudumc (the Netherlands) and CNRFP (Burkina Faso), following membrane feeding, unfed mosquitoes were immediately removed from cups with an aspirator. On days 4–6 PI, mosquitoes were given a second bloodmeal to synchronize oocyst development. Mosquitoes were kept at 27–29°C in the insectaries on 5–10% glucose and dissected for 7–8 d to assess infection prevalence. Twenty mosquito midguts were stained with 1% mercurochrome and oocysts were examined and confirmed by two independent microscopists at ×400 under an optic microscope (CX 40 Olympus). If oocyst prevalence was above 40%, the infected mosquitoes were transferred to the bio-secure insectary in Nijmegen, whilst in Ouagadougou cups with infected mosquitoes were placed into secured metal cages (30 × 30 × 30 cm) and kept in a temperature- and relative humidity-controlled environment (27–29°C and 70–80% HR) with double-screened doors to prevent sporozoite-positive mosquitoes from escaping.
Extrinsic incubation period
The EIP, defined as the duration of sporogony, was estimated in An. stephensi mosquitoes in order to assess (i) the best day to perform the expelling experiment and (ii) if EIP differed in low- vs high-infected mosquitoes. Cages containing 150 mosquitoes were fed with P. falciparum NF54 and NF135 gametocytes, and 20 midguts were dissected on days 6–8 PI to assess oocyst infection intensity. Groups of low- (mean of 5 oocysts) and high (mean of 20 oocysts)-infected mosquitoes (prevalence of infection > 70%) were maintained in a secure insectary until salivary gland dissections were performed on days 9–11. Salivary glands for individual mosquitoes were collected in 1.5 ml Eppendorf tubes containing 180 µl oocyst lysis buffer (NaCl 0.1 M: EDTA 25 mM: Tris–HCl 10 mM), and stored at –20°C for further molecular analysis. Mosquito bodies were first homogenized by beat-beating as previously described (Graumans et al., 2017) in 100 µl phosphate-buffered saline (PBS).
Sporozoite-expelling experiments
On 15 and 16 d post feeding, at Radboudumc and CNRFP respectively, infected mosquitoes were used to quantify the number of expelled sporozoites. To prevent contamination, all instruments and equipment were cleaned from nucleic acids by 30 min exposure to sodium hypochlorite (10% in H2O), rinsed with water, and paper-dried on the day before the experiment. New gloves were used each time the experiments were performed. Integra dermal substitute (Dermal Regeneration Template, single layer 20 × 25 cm, ref 68101), hereafter referred to as artificial skin, was cut into 3.5 cm squares (Figure 6). Squares were transferred to Petri dishes filled with sterile nuclease-free water (VWR, E476) and left overnight at room temperature (RT). Mosquitoes were individually collected in small Perspex cages (5 × 5 × 7 cm, covered with netting material on the top and bottom sides). Mosquitoes were starved 14–16 hr prior to feeding. On the day of the expelling experiment, artificial skin was transferred with gloves to an inverted positioned glass membrane mini-feeder (convex bottom, 15 mm diameter, ref 70172000) connected to a heated circulating water bath (CORIO C-B5, Julabo) set to 39°C. A rubber band was wrapped around the feeder to secure the artificial skin. Paper tissue was gently pressed on the skin four times to absorb water. Then, 100 µl of naïve donor blood (EDTA, BD Vacutainers, ref 367525) was pipetted on the circular artificial skin area and spread evenly across the surface with the horizontal side of the tip. The feeder was then turned around and placed on top of the cage, without touching the netting, with a maximum of 8 min for mosquito probing. Following mosquito probing, a scalpel (Dalhausen präzisa plus, no. 11) was used to cut the artificial skin above the rubber band around the entire feeder. The artificial skin was transferred with tweezers to a 1.5 ml Eppendorf tube containing 180 µl oocyst lysis buffer, and stored at –20°C. After feeding, mosquitoes were transferred to metal cages. Mosquitoes were kept at 27–29°C in the insectaries on 5% glucose.
Artificial skin feeding procedure.
(A) Petri dish filled with sterile nuclease-free water, containing the artificial skin. (B) Skin folded around a mini-feeder and supplemented with 100 µl of human blood. (C) Feeding setup with adjustable table and a small Perspex cage containing a single mosquito underneath. (D) A blood-feeding mosquito on artificial skin.
Immunolabeling of intact and ruptured oocysts
Mosquitoes were allowed to digest blood for 3 d to prevent interference with immunolabeling; mosquito salivary glands and midguts were dissected on day 18 PI for An. stephensi (Nijmegen), and day 19 PI for An. coluzzii (Burkina Faso). Salivary glands were collected in 1.5 ml Eppendorf tubes containing 180 µl oocyst lysis buffer (NaCl 0.1 M: EDTA 25 mM: Tris–HCl 10 mM), and stored at –20°C for further molecular analysis. Midguts were dissected in 20 µl PBS (pH 7.2) without mercurochrome. For experiments performed in Nijmegen, An. stephensi midguts were transferred to a fresh drop of (1:400) 3SP2-Alexa 488 anti-CSP antibodies (DyLight 488 Anitibody Labeling Kit, Thermo Fisher Scientific, ref 53024) and incubated for 30 min at RT in a slide humidity incubation box. Following staining, midguts were washed twice with 10 µl of PBS for 10 min. Midguts were transferred to glass slides and secured with a cover slip. Intact/degenerated and ruptured oocysts were counted using an incident light fluorescence microscope GFP filter at 400×. Due to the lack of fluorescence microscopes at CNRFP, we combined a formalin fixation method with immunostaining. An. coluzzii midguts were transferred into individual screw cap tubes (Eppendorf) filled with 400 µl of 4% formalin and stored at 4°C until shipped to the Netherlands. Midguts stored in 4% formalin were collected by using a p1000 Gilson pipette with the point of the tip cut and placed on a slide. Midguts were rinsed from formalin three times in PBS 1X-Tween 0.05% by moving the midgut with a needle from drop to drop. They were then transferred to a fresh drop of 3SP2-Alexa 488 anti-CSP antibodies (1:400) and incubated for 20 min. Midguts were rinsed in PBS and examined as described above.
Sample extraction, sporozoite quantification by qPCR, and amplicon deep sequencing
Serial dilutions of P. falciparum sporozoites were generated to prepare standard curves for qPCR. Therefore, pooled salivary glands from highly infected mosquitoes were collected in a glass pestle grinder that contained 500 µl PBS. The sample was homogenized and subsequently diluted 100 times in PBS. Sporozoites were transferred to a hemocytometer and counted under a phase-contrast light microscopy (×400 magnification), by two independent microscopists. Serial dilutions were prepared in PBS using glass test tubes and low binding tips. For each concentration, 100 µl was filled out in a 1.5 ml Eppendorf tube. Eppendorf tubes were stored at –70°C for at least 1 d before sample processing. Prior to DNA extraction, 30 μl proteinase K (QIAGEN, Cat# 19133) was added to Eppendorf tubes containing collected artificial skin and salivary glands samples. PBS was added to all samples to have an equal volume (410 µl) before incubation overnight at 56°C. The following day, total nucleic acids (NA) were extracted with the automated MagNA Pure LC instrument (Roche) using the MagNA Pure LC DNA Isolation Kit – High performance (Roche, product no. 03310515001), and eluted in 50 μl. Samples were used immediately or stored at –20°C. P. falciparum sporozoites were quantified by qPCR, targeting the mitochondrial gene COX-1. A previously published primer set (Boissière et al., 2013) was modified to improve template annealing, forward primer 5′-CATCAGGAATGTTATTGCTAACAC-3′ and reverse primer 5′-GGATCTCCTGCAAATGTTGGGTC-3′, resulting in an amplicon length of 112 bp. A probe was designed for amplicon detection 6FAM-ACCGGTTTTAACTGGAGGAGTA-BHQ1. qPCR reactions were prepared with TaqMan Fast Advanced Master Mix (Applied Biosystems, ref 4444557). For each reaction, 12.5 µl mix, 0.4 µl primers (stock 50 µM, final concentration 800 nM), 0.1 µl probe (stock 100 µM, final concentration 400 nM), 7 µl PCR grade water, and 5 µl template DNA were used. In each run, standard curves and negative controls (water) were included. Melt curves were visually inspected. Samples were run on a Bio-Rad CFX 96 real-time System at 95°C for 15 s, followed by 30 cycles of 95°C for 15 s, 60°C for 60 s. To identify unique clones in mosquitoes infected by gametocyte donors with naturally acquired P. falciparum gametocytes, samples were genotyped by apical membrane antigen 1 (AMA-1) amplicon sequencing as previously described (Briggs et al., 2020).
Statistical analysis
Statistical analyses were performed in R, version 3.1.12 (Team RC, 2019). Associations between log10 oocyst intensity and infection prevalence were modeled using a logistic regression model (using N = 457, Figure 1B). The difference in mean log10 oocyst densities between staining types was compared using a t-test (t = 0.070075, df = 405, p-value=0.9442, N = 406, Figure 1D). The association between total sporozoite density and experiment day with the prevalence of salivary gland sporozoite was modeled using a mixed-effects logistic regression with a random intercept for the different experiments (using N = 266, Figure 2A). Welch’s t-test was used to compare sporozoite density between days 9 and 10 (t = –2.0467, df = 28.66, p=0.04995, N = 31, Figure 2B). The association between total sporozoite load and sporozoite expelling prevalence was modelled using a logistic regression (using N = 186, Figure 3A).
Spearman’s correlation coefficient ρ was used to assess the association between oocyst sheets and salivary gland sporozoite load (one outlier not included) (ρ = 0.80, 95% CI, 0.74–0.85; p<0.0001; N = 111, Figure 3B); the association between total sporozoite density and the number of sporozoites that was expelled into the artificial skin (ρ = 0.35, 95% CI, 0.17–0.50; p=0.0002; N = 112, Figure 3C) (one outlier not included); the associations between ruptured oocyst density and total sporozoite load, and between total sporozoite load and skin expelling (ρ = 0.9, 95% CI, 0.80–0.95; p<0.0001; N = 25, Figure 4C) and (ρ = 0.71, 95% CI, 0.52–0.82; p<0.0003; N = 25, Figure 4D).
Additional informationCompeting interests
No competing interests declared
Author contributions
Conceptualization, Formal analysis, Investigation, Methodology, Writing – original draft
Conceptualization, Formal analysis, Investigation, Methodology, Writing – original draft
Investigation, Methodology, Writing – review and editing
Investigation, Methodology, Writing – review and editing
Formal analysis, Visualization, Methodology, Writing – review and editing
Investigation, Writing – review and editing
Writing – review and editing
Investigation
Investigation
Resources, Methodology
Conceptualization, Writing – review and editing
Supervision, Writing – review and editing
Supervision, Writing – review and editing
Supervision, Investigation, Writing – review and editing
Conceptualization
Investigation, Methodology, Writing – original draft
Conceptualization, Formal analysis, Supervision, Funding acquisition, Methodology, Writing – original draft, Writing – review and editing
Ethics
The study protocol in Burkina Faso was approved by the London School of Hygiene and Tropical Medicine ethics committee (Review number: 14724), the Centre National de Recherche et de Formation sur le Paludisme institutional review board (Deliberation N° 2018/000,002/MS/SG/CNRFP/CIB) and the Ethics Committee for Health Research in Burkina Faso (Deliberation N° 2018-01-010). Donors were recruited at schools in Sapone00E9; Health District, following informed consent. Experiments with in vitro cultured parasites and An. stephensi mosquitoes at Radboud University Medical Center were conducted following approval from the Radboud University Experimental Animal Ethical Committee (RUDEC 2009-019, RUDEC 2009-225).
Data availability
All data generated or analysed during this study are deposited on Dryad: https://doi.org/10.5061/dryad.dbrv15f89.
The following dataset was generated:
AndolinaCGraumansWGuelbeogoMGemertGJVRamjithJHarounaSSoumanabaZStoterRVegte-BolmerMPangosMSinnisPCollinsKStaedkeSGTionoABDrakeleyCLankeKBousemaT2024Data from: Quantification of sporozoite expelling by Anopheles mosquitoes infected with laboratory and naturally circulating P. falciparum gametocytesDryad Digital Repository10.5061/dryad.dbrv15f89
Acknowledgements
We thank all the study participants for their willingness to support the study and donate blood. We would like to thank all the lab staff, drivers, and procurement for their dedication in the study; Jacqueline Kuhnen, Laura Pelser-Posthumus, Astrid Pouwelsen, and Jolanda Klaassen (Radboudumc) for all mosquito husbandry; Jared Honeycutt and Sophie Maxfield (University of California – San Francisco, USA) for their work on AMA-1 amplicon deep sequencing; Claudia Bin for all the tester units of IntegraDermal Regeneration Template. Funding was provided by the fellowship from the European Research Council to TB (ERC-CoG 864180; QUANTUM) and an AMMODO Science Award (2019) to TB. PS was supported by the National Institutes of Health through an R01 grant (AI132359).
ReferencesAgostinisCSpazzapanMVuerichRBalduitAStoccoCMangognaARicciGPapaGZacchignaSBullaR2021Differential capability of clinically employed dermal regeneration scaffolds to support vascularization for tissue bioengineeringBiomedicines9145810.3390/biomedicines910145834680575AhmedAPignatelliPElaagipAAbdel HamidMMAlrahmanOFWeetmanD2021Invasive malaria vector anopheles stephensi mosquitoes in sudan, 2016-2018Emerging Infectious Diseases272952295410.3201/eid2711.21004034670658AleshnickMGanusovVVNasirGYenokyanGSinnisP2020Experimental determination of the force of malaria infection reveals a non-linear relationship to mosquito sporozoite loadsPLOS Pathogens16e100818110.1371/journal.ppat.100818132453765AndolinaCRekJCBriggsJOkothJMusiimeARamjithJTeyssierNConradMNankabirwaJILankeKRodriguez-BarraquerIMeerstein-KesselLArinaitweEOlwochPRosenthalPJKamyaMRDorseyGGreenhouseBDrakeleyCStaedkeSGBousemaT2021Sources of persistent malaria transmission in a setting with effective malaria control in eastern uganda: a longitudinal, observational cohort studyThe Lancet. Infectious Diseases211568157810.1016/S1473-3099(21)00072-434146476BeierJCDavisJRVaughanJANodenBHBeierMS1991aQuantitation of Plasmodium falciparum sporozoites transmitted in vitro by experimentally infected Anopheles gambiae and Anopheles stephensiThe American Journal of Tropical Medicine and Hygiene4456457010.4269/ajtmh.1991.44.5642063960BeierJCOnyangoFKKorosJKRamadhanMOgwangRWirtzRAKoechDKRobertsCR1991bQuantitation of malaria sporozoites transmitted in vitro during salivation by wild Afrotropical AnophelesMedical and Veterinary Entomology5717910.1111/j.1365-2915.1991.tb00523.x1768903BoissièreAGimonneauGTchioffoMTAbateLBayibekiAAwono-AmbénéPHNsangoSEMorlaisI2013Application of a qPCR assay in the investigation of susceptibility to malaria infection of the M and S molecular forms of an: gambiae S.S. in cameroonPLOS ONE8e5482010.1371/journal.pone.005482023349974BousemaJTSchneiderPGouagnaLCDrakeleyCJTostmannAHoubenRGithureJIOrdRSutherlandCJOmarSASauerweinRW2006Moderate effect of artemisinin-based combination therapy on transmission of Plasmodium falciparumThe Journal of Infectious Diseases1931151115910.1086/50305116544256BradleyJStoneWDaDFMorlaisIDickoACohuetAGuelbeogoWMMahamarANsangoSSoumaréHMDiawaraHLankeKGraumansWSiebelink-StoterRvan de Vegte-BolmerMChenITionoAGonçalvesBPGoslingRSauerweinRWDrakeleyCChurcherTSBousemaT2018Predicting the likelihood and intensity of mosquito infection from sex specific Plasmodium falciparum gametocyte densityeLife7e3446310.7554/eLife.3446329848446BriggsJTeyssierNNankabirwaJIRekJJagannathanPArinaitweEBousemaTDrakeleyCMurrayMCrawfordEHathawayNStaedkeSGSmithDRosenthalPJKamyaMDorseyGRodriguez-BarraquerIGreenhouseB2020Sex-based differences in clearance of chronic Plasmodium falciparum infectioneLife9e5987210.7554/eLife.5987233107430BrugmanVAKristanMGibbinsMPAngrisanoFSalaKADessensJTBlagboroughAMWalkerT2018Detection of malaria sporozoites expelled during mosquito sugar feedingScientific Reports8754510.1038/s41598-018-26010-629765136ChurcherTSBlagboroughAMDelvesMRamakrishnanCKapuluMCWilliamsARBiswasSDaDFCohuetASindenRE2012Measuring the blockade of malaria transmission--an analysis of the Standard Membrane Feeding AssayInternational Journal for Parasitology421037104410.1016/j.ijpara.2012.09.00223023048ChurcherTSSindenREEdwardsNJPoultonIDRamplingTWBrockPMGriffinJTUptonLMZakutanskySESalaKAAngrisanoFHillAVSBlagboroughAM2017Probability of transmission of malaria from mosquito to human is regulated by mosquito parasite density in naïve and vaccinated hostsPLOS Pathogens13e100610810.1371/journal.ppat.100610828081253ConwayDJFanelloCLloydJMAl-JouboriBMBalochAHSomanathSDRoperCOduolaAMMulderBPovoaMMSinghBThomasAW2000Origin of Plasmodium falciparum malaria is traced by mitochondrial DNAMolecular and Biochemical Parasitology11116317110.1016/s0166-6851(00)00313-311087926FeldmannAMPonnuduraiT1989Selection of Anopheles stephensi for refractoriness and susceptibility to Plasmodium falciparumMedical and Veterinary Entomology3415210.1111/j.1365-2915.1989.tb00473.x2519646FrischknechtFBaldacciPMartinBZimmerCThibergeSOlivo-MarinJCShorteSLMénardR2004Imaging movement of malaria parasites during transmission by Anopheles mosquitoesCellular Microbiology668769410.1111/j.1462-5822.2004.00395.x15186404GonçalvesBPKapuluMCSawaPGuelbéogoWMTionoABGrignardLStoneWHellewellJLankeKBastiaensGJHBradleyJNébiéINgoiJMOriangoRMkabiliDNyaurahMMidegaJWirthDFMarshKChurcherTSBejonPSirimaSBDrakeleyCBousemaT2017Examining the human infectious reservoir for Plasmodium falciparum malaria in areas of differing transmission intensityNature Communications8113310.1038/s41467-017-01270-429074880GraumansWTadesseFGAndolinaCvan GemertGJTeelenKLankeKGadisaEYewhalawDvan de Vegte-BolmerMSiebelink-StoterRReulingISauerweinRBousemaT2017Semi-high-throughput detection of Plasmodium falciparum and Plasmodium vivax oocysts in mosquitoes using bead-beating followed by circumsporozoite ELISA and quantitative PCRMalaria Journal1635610.1186/s12936-017-2011-928877707GraumansWAndolinaCAwanduSSGrignardLLankeKBousemaT2019Plasmodium falciparum gametocyte enrichment in peripheral blood samples by magnetic fractionation: gametocyte yields and possibilities to reuse columnsThe American Journal of Tropical Medicine and Hygiene10057257710.4269/ajtmh.18-077330608048GraumansWJacobsEBousemaTSinnisP2020When is a plasmodium-infected mosquito an infectious mosquito?Trends in Parasitology3670571610.1016/j.pt.2020.05.01132620501GuissouEDaDFHienDYameogoKBYerbangaSROuédraogoGADabiréKRLefèvreTCohuetA2023Intervention reducing malaria parasite load in vector mosquitoes: no impact on Plasmodium falciparum extrinsic incubation period and the survival of Anopheles gambiaePLOS Pathogens19e101108410.1371/journal.ppat.101108437195964HillyerJFBarreauCVernickKD2007Efficiency of salivary gland invasion by malaria sporozoites is controlled by rapid sporozoite destruction in the mosquito haemocoelInternational Journal for Parasitology3767368110.1016/j.ijpara.2006.12.00717275826HolzschuhAKoepfliC2022Tenfold difference in DNA recovery rate: systematic comparison of whole blood vs dried blood spot sample collection for malaria molecular surveillanceMalaria Journal218810.1186/s12936-022-04122-935292038MatsuokaHYoshidaSHiraiMIshiiA2002A rodent malaria, Plasmodium berghei, is experimentally transmitted to mice by merely probing of infective mosquito, Anopheles stephensiParasitology International51172310.1016/s1383-5769(01)00095-211880224MedicaDLSinnisP2005Quantitative dynamics of Plasmodium yoelii sporozoite transmission by infected anopheline mosquitoesInfection and Immunity734363436910.1128/IAI.73.7.4363-4369.200515972531MiuraKSwihartBJDengBZhouLPhamTPDioufAFayMPLongCA2019Strong concordance between percent inhibition in oocyst and sporozoite intensities in a Plasmodium falciparum standard membrane-feeding assayParasites & Vectors1220610.1186/s13071-019-3470-331060594MuellerAKKohlheppFHammerschmidtCMichelK2010Invasion of mosquito salivary glands by malaria parasites: prerequisites and defense strategiesInternational Journal for Parasitology401229123510.1016/j.ijpara.2010.05.00520621627MusiimeAKOkothJConradMAyoDOnyigeIRekJNankabirwaJIArinaitweEKamyaMRDorseyGvan GemertGJStaedkeSGDrakeleyCBousemaTAndolinaC2019Is that a real oocyst? Insectary establishment and identification of Plasmodium falciparum oocysts in midguts of Anopheles mosquitoes fed on infected human blood in Tororo, UgandaMalaria Journal1828710.1186/s12936-019-2922-831455343OuédraogoAL2013A protocol for membrane feeding assays to determine the infectiousness of P. falciparum naturally infected individuals to Anopheles gambiaePLOS ONE7004282110.1371/journal.pone.0042821OuédraogoALGonçalvesBPGnéméAWengerEAGuelbeogoMWOuédraogoAGerardinJBeverCALyonsHPitroipaXVerhaveJPEckhoffPADrakeleyCSauerweinRLutyAJFKouyatéBBousemaT2016Dynamics of the human infectious reservoir for malaria determined by mosquito feeding assays and ultrasensitive malaria diagnosis in burkina fasoThe Journal of Infectious Diseases213909910.1093/infdis/jiv37026142435PonnuduraiTLensenAHLeeuwenbergADMeuwissenJH1982Cultivation of fertile Plasmodium falciparum gametocytes in semi-automated systems. 1. static culturesTransactions of the Royal Society of Tropical Medicine and Hygiene7681281810.1016/0035-9203(82)90116-XPonnuduraiTLensenAHVan GemertGJBensinkMPBolmerMMeuwissenJH1989Infectivity of cultured Plasmodium falciparum gametocytes to mosquitoesParasitology98 Pt 216517310.1017/s00311820000620652668861PringleG1965A count of the sporozoites in an ooecyst of Plasmodium falciparumTransactions of the Royal Society of Tropical Medicine and Hygiene5928929010.1016/0035-9203(65)90009-x14298032RosenbergRWirtzRASchneiderIBurgeR1990An estimation of the number of malaria sporozoites ejected by a feeding mosquitoTransactions of the Royal Society of Tropical Medicine and Hygiene8420921210.1016/0035-9203(90)90258-g2202101RosenbergRRungsiwongseJ1991The number of sporozoites produced by individual malaria oocystsThe American Journal of Tropical Medicine and Hygiene4557457710.4269/ajtmh.1991.45.5741951866RossignolPARibeiroJMSpielmanA1984Increased intradermal probing time in sporozoite-infected mosquitoesThe American Journal of Tropical Medicine and Hygiene33172010.4269/ajtmh.1984.33.176696175SchneiderPBousemaJTGouagnaLCOtienoSvan de Vegte-BolmerMOmarSASauerweinRW2007Submicroscopic Plasmodium falciparum gametocyte densities frequently result in mosquito infectionThe American Journal of Tropical Medicine and Hygiene7647047417360869ShaukatAMBremanJGMcKenzieFE2010Using the entomological inoculation rate to assess the impact of vector control on malaria parasite transmission and eliminationMalaria Journal912210.1186/1475-2875-9-12220459850ShawWRHolmdahlIEItoeMAWerlingKMarquetteMPatonDGSinghNBuckeeCOChildsLMCatterucciaF2020Multiple blood feeding in mosquitoes shortens the Plasmodium falciparum incubation period and increases malaria transmission potentialPLOS Pathogens16e100913110.1371/journal.ppat.100913133382824SlaterHCRossAOuédraogoALWhiteLJNguonCWalkerPGTNgorPAguasRSilalSPDondorpAMLa BarrePBurtonRSauerweinRWDrakeleyCSmithTABousemaTGhaniAC2015Assessing the impact of next-generation rapid diagnostic tests on Plasmodium falciparum malaria elimination strategiesNature528S94S10110.1038/nature1604026633771SlaterHCRossAFelgerIHofmannNERobinsonLCookJGonçalvesBPBjörkmanAOuedraogoALMorrisUMsellemMKoepfliCMuellerITadesseFGadisaEDasSDomingoGKapuluMMidegaJOwusu-AgyeiSNabetCPiarrouxRDoumboODoumboSNKoramKLucchiNUdhayakumarVMoshaJTionoAChandramohanDGoslingRMwingiraFSauerweinRPaulRRileyEMWhiteNJNostenFImwongMBousemaTDrakeleyCOkellLC2019The temporal dynamics and infectiousness of subpatent Plasmodium falciparum infections in relation to parasite densityNature Communications10143310.1038/s41467-019-09441-130926893SmithDLBattleKEHaySIBarkerCMScottTWMcKenzieFE2012Ross, macdonald, and a theory for the dynamics and control of mosquito-transmitted pathogensPLOS Pathogens8e100258810.1371/journal.ppat.100258822496640SolarteYManzanoMRRochaLHurtadoHJamesMAArévalo-HerreraMHerreraS2011Plasmodium vivax sporozoite production in Anopheles albimanus mosquitoes for vaccine clinical trialsThe American Journal of Tropical Medicine and Hygiene84283410.4269/ajtmh.2011.09-049921292875SoontarawiratIAndolinaCPaulRDayNPJNostenFWoodrowCJImwongM2017Plasmodium vivax genetic diversity and heterozygosity in blood samples and resulting oocysts at the Thai-Myanmar borderMalaria Journal1635510.1186/s12936-017-2002-x28870214SoumareHMGuelbeogoWMvan de Vegte-BolmerMvan GemertGJSoumanabaZOuedraogoAOuattaraMSAbdullahiAJadamaLCamaraMMGayePMMendyMDavisNTionoABD’AlessandroUDrakeleyCBousemaTMorenoMCollinsKA2021Maintaining Plasmodium falciparum gametocyte infectivity during blood collection and transport for mosquito feeding assays in the fieldMalaria Journal2019110.1186/s12936-021-03725-y33879163StoneWJRElderingMvan GemertG-JLankeKHWGrignardLvan de Vegte-BolmerMGSiebelink-StoterRGraumansWRoeffenWFGDrakeleyCJSauerweinRWBousemaT2013The relevance and applicability of oocyst prevalence as a read-out for mosquito feeding assaysScientific Reports3341810.1038/srep0341824301557StopardIJChurcherTSLambertB2021Estimating the extrinsic incubation period of malaria using a mechanistic model of sporogonyPLOS Computational Biology17e100865810.1371/journal.pcbi.100865833591963Team RC2019The R project for statistical computingCranhttps://www.r-project.org/TeirlinckACRoestenbergMvan de Vegte-BolmerMScholzenAHeinrichsMJLSiebelink-StoterRGraumansWvan GemertGJTeelenKVosMWNganou-MakamdopKBorrmannSRozierYPAErkensMAALutyAJFHermsenCCSimBKLvan LieshoutLHoffmanSLVisserLGSauerweinRW2013NF135.C10: a new Plasmodium falciparum clone for controlled human malaria infectionsThe Journal of Infectious Diseases20765666010.1093/infdis/jis72523186785ThiéventKZilioGHauserGKoellaJC2019Malaria load affects the activity of mosquito salivary apyraseJournal of Insect Physiology116101610.1016/j.jinsphys.2019.04.00330986373TustingLSBousemaTSmithDLDrakeleyC2014Measuring changes in Plasmodium falciparum transmission: precision, accuracy and costs of metricsAdvances in Parasitology8415120810.1016/B978-0-12-800099-1.00003-X24480314van de Vegte-BolmerMGraumansWStoterRvan GemertGJSauerweinRCollinsKABousemaT2021A portfolio of geographically distinct laboratory-adapted Plasmodium falciparum clones with consistent infection rates in Anopheles mosquitoesMalaria Journal2038110.1186/s12936-021-03912-x34565372VaughanJANodenBHBeierJC1992Population dynamics of Plasmodium falciparum sporogony in laboratory-infected Anopheles gambiaeThe Journal of Parasitology787167241635032WangCYTMcCarthyJSStoneWJBousemaTCollinsKABialasiewiczS2018Assessing Plasmodium falciparum transmission in mosquito-feeding assays using quantitative PCRMalaria Journal1724910.1186/s12936-018-2382-629976199WellsMBAndrewDJ2019Anopheles salivary gland architecture shapes plasmodium sporozoite availability for transmissionmBio10e0123810.1128/mBio.01238-19WHO2017Meeting report of the WHO Evidence Review Group on Low-Density Malaria Infectionshttps://www.who.int/publications/m/item/WHO-HTM-GMP-MPAC-2017.10July 17, 2023YangASPLopatickiSO’NeillMTEricksonSMDouglasDNKnetemanNMBoddeyJA2017AMA1 and MAEBL are important for Plasmodium falciparum sporozoite infection of the liverCellular Microbiology19274510.1111/cmi.127452837116810.7554/eLife.90989.3.sa0eLife assessmentLevashinaElena AReviewing EditorMax Planck Institute for Infection BiologyGermanyCompellingImportant
This important study combines experimental infections with laboratory and field Plasmodium falciparum isolates to quantify the force of human malaria parasite transmission. By using compelling methodological approaches, the authors establish clear positive correlations between mosquito infection levels (as determined by oocyst numbers), sporozoite loads in salivary glands, and sporozoites expelled during feeding. The link between heterogeneous infection levels in the mosquitoes and malaria transmission would be of interest to vector biologists, parasitologists, immunologists, and mathematical modelers.
There is a long-believed dogma in the malaria field; a mosquito infected with a single oocyst is equally infectious to humans as another mosquito with many oocysts. This belief has been used for goal setting (and modeling) of malaria transmission-blocking interventions. While recent studies using rodent malaria suggest that the dogma may not be true, there was no such study with human P. falciparum parasites. In this study, the numbers of oocysts and sporozoite in the mosquitoes and the number of expelled sporozoites into artificial skin from the infected mosquito was quantified individually. There was a significant correlation between sporozoite burden in the mosquitoes and expelled sporozoites. In addition, this study showed that highly infected mosquitoes expelled sporozoites sooner.
Strengths:
• The study was conducted using two different parasite-mosquito combinations; one was lab-adapted parasites with Anopheles stephensi and the other was parasites, which were circulated in infected patients, with An. coluzzii. Both combinations showed statistically significant correlations between sporozoite burden in mosquitoes and the number of expelled sporozoites.
• Usually, this type of study has been done in group bases (e.g., count oocysts and sporozoites at different time points using different mosquitoes from the same group). However, this study determined the numbers in individual bases after multiple optimization and validation of the approach. This individual approach significantly increases the power of correlation analysis.
Weaknesses:
• In a natural setting, most mosquitoes have less than 5 oocysts. Thus, the conclusion is more convincing if the authors perform additional analysis for the key correlations (Fig 3C and 4D) excluding mosquitoes with very high total sporozoite load (e.g., more than 5-oocyst equivalent load).
• As written as the second limitation of the study, this study did not investigate whether all expelled sporozoites were equally infectious. For example, Day 9 expelled sporozoites may be less infectious than Day 11 sporozoites, or expelled sporozoites from high-burden mosquitoes may be less infectious because they experience low nutrient conditions in a mosquito. Ideally, it is nice to test the infectivity by ex vivo assays, such as hepatocyte invasion assay, and gliding assay at least for salivary sporozoites. But are there any preceding studies where the infectivity of sporozoites from different conditions was evaluated? Citing such studies would strengthen the argument.
• Since correlation analyses are the main points of this paper, it is important to show 95%CI of Spearman rank coefficient (not only p-value). By doing so, readers will understand the strengths/weaknesses of the correlations. The p-value only shows whether the observed correlation is significantly different from no correlation or not. In other words, if there are many data points, the p-value could be very small even if the correlation is weak.
The malaria parasite Plasmodium develops into oocysts and sporozoites inside Anopheles mosquitoes, in a process called sporogony. Sporozoites invade the insect salivary glands in order to be transmitted during a blood meal. An important question regarding malaria transmission is whether all mosquitoes harboring Plasmodium parasites are equally infectious. In this paper, the authors investigated the progression of P. falciparum sporozoite development in Anopheles mosquitoes, using a sensitive qPCR method to quantify sporozoites and an artificial skin system to probe for parasite expelling. They assessed the association between oocyst burden, salivary gland infection intensity, and sporozoites expelled.
The data show that higher sporozoite loads are associated with earlier colonization of salivary glands and a higher prevalence of sporozoite-positive salivary glands and that higher salivary gland sporozoite burdens are associated with higher numbers of expelled sporozoites. Intriguingly, there is no clear association between salivary gland burdens and the prevalence of expelling, suggesting that most infections reach a sufficient threshold to allow parasite expelling during a mosquito bite. This important observation suggests that low-density gametocyte carriers, although less likely to infect mosquitoes, could nevertheless contribute to malaria transmission.
Strengths:
The paper is well written and the work is well conducted. The authors used two experimental models, one using cultured P. falciparum gametocytes and An. stephensi mosquitoes, and the other one using natural gametocyte infections in a field setup with An. coluzzii mosquitoes. Both studies gave similar results, reinforcing the validity of the observations. Parasite quantification relies on a robust and sensitive qPCR method, and parasite expelling was assessed using an innovative experimental setup based on artificial skin.
Weaknesses:
There is no clear association between the prevalence of sporozoite expelling and the parasite burden. However, high total sporozoite burdens are associated with earlier and more efficient colonization of the salivary glands, and higher salivary gland burdens are associated with higher numbers of expelled sporozoites. While these observations suggest that highly infected mosquitoes could transmit/expel parasites earlier, this is not directly addressed in the study. In addition, whether all expelled sporozoites are equally infectious is unknown. The central question, i.e. whether all infected mosquitoes are equally infectious, therefore remains open.
This study uses a state-of-the-art artificial skin assay to determine the quantity of P. falciparum sporozoites expelled during feeding using mosquito infection (by standardised membrane feeding assay SMFA) using both cultured gametocytes and natural infection. Sporozoite densities in salivary glands and expelled into the skin are quantified using a well-validated molecular assay. These studies show clear positive correlations between mosquito infection levels (as determined by oocyst numbers), sporozoite numbers in salivary glands, and sporozoites expelled during feeding. This indicates potentially significant heterogeneity in infectiousness between mosquitoes with different infection loads and thus challenges the often-made assumption that all infected mosquitoes are equally infectious.
Strengths:
Very rigorously designed studies using very well validated, state-of-the-art methods for studying malaria infections in the mosquito and quantifying load of expelled sporozoites. This resulted in very high-quality data that was well-analyzed and presented. Both sources of gametocytes (cultures vs. natural infection) show consistent results further strengthening the quality of the results obtained.
Weaknesses:
As is generally the case when using SMFAs, the mosquito infections levels are often relatively high compared to wild-caught mosquitoes (e.g. Bombard et al 2020 IJP: median 3-4 ), and the strength of the observed correlations between oocyst sheet and salivary gland sporozoite load even more so between salivary gland sporozoite load and expelled sporozoite number may be dominated by results from mosquitoes with infection levels rarely observed in wild-caught mosquitoes. This could result in an overestimation of the importance of these well-observed positive relationships under natural transmission conditions.
The results obtained from these excellently designed and executed studies very well supported their conclusion - with a slight caveat regarding their application to natural transmission scenarios
This work very convincingly highlights the potential for significant heterogeneity in the infectiousness between individual P. falciparum-infected mosquitoes. Such heterogeneity needs to be further investigated and if again confirmed taken into account both when modelling malaria transmission and when evaluating the importance of low-density infections in sustaining malaria transmission.
10.7554/eLife.90989.3.sa4Author ResponseAndolinaChiaraAuthorRadboudumcNijmegenNetherlandsGraumansWouterAuthorRadboud University Medical CenterNijmegenNetherlandsGuelbeogoMoussaAuthorCentre National de Recherche et de Formation sur le PaludismeOuagadougouBurkina Fasovan GemertGeert-JanAuthorRadboud University Medical CenterNijmegenNetherlandsRamijthJordacheAuthorRadboud University Nijmegen Medical CentreNijmegenNetherlandsHarounaSoréAuthorCentre National de Recherche et de Formation sur le PaludismeOuagadouguBurkina FasoSoumanabaZongoAuthorCentre National de Recherche et de Formation sur le PaludismeOuagadouguBurkina FasoStoterRianneAuthorRadboud University Nijmegen Medical CentreNijmegenNetherlandsvan de Vegte-BolmerMargaAuthorRadboud University Medical CenterNijmegenNetherlandsPangosMartinaAuthorDepartment of Plastic and Reconstructive Surgery, Azienda Ospedaliero Universitaria Giuliano IsontinatriesteItalySinnisPhotiniAuthorJohn Hopkins Bloomberg School of Public HealthBaltimoreUnited StatesCollinsKatharine AAuthorRadboud University Medical CenterNijmegenNetherlandsStaedkeSarah GAuthorLiverpool School of Tropical Medicine of Tropical MedicineLiverpoolUnited KingdomTionoAlfred BAuthorCentre National de Recherche et de Formation sur le PaludismeOuagadougouBurkina FasoDrakeleyChrisAuthorLondon School of Hygiene and Tropical MedicineLondonUnited KingdomLankeKjerstinAuthorRadboud University Medical CenterNijmegenNetherlandsBousemaTeunAuthorRadboud University Nijmegen Medical CentreNijmegenNetherlands
We would like to thank the editor and the reviewers for their constructive comments and the chance to revise the manuscript. The suggestions have allowed us to improve our manuscript. We have been able to fulfil all reviewer comments and added new statistical analyses to examine associations for subsets of data. Whilst suggested by a reviewer, we did not perform large-scale experiments to confirm the viability of low sporozoite densities at different time-points post salivary gland colonization. For these assays there are currently no satisfactory in vitro models for sporozoites harvested from single mosquitoes and setting up and validating such experiments could be a PhD project in itself. We do consider this suggestion very relevant but beyond the scope of the current work.
Relevantly, during the time the manuscript was under review at eLife, we have been able to examine the multiplicity of infection in our field experiments. This was, as written in the original manuscript, a key reason to also perform experiments in the field where there is a greater diversity of parasite lines. We have successfully performed AMA-1 amplicon deep sequencing on infected mosquito salivary glands and infected skins. Although this does not change the key messages of the manuscript and is secondary to our main hypothesis, we do consider it a relevant addition since we were able to demonstrate that for some infected mosquitoes from the Burkina Faso study, multiple clones were expelled by mosquitoes during probing on a single piece of artificial skin. We have added a short paragraph to our revised manuscript and updated the acknowledgement section to include the supporting researcher who conducted those experiments.
Reviewer #1 (Public Review):
Summary: There is a long-believed dogma in the malaria field; a mosquito infected with a single oocyst is equally infectious to humans as another mosquito with many oocysts. This belief has been used for goal setting (and modelling) of malaria transmission-blocking interventions. While recent studies using rodent malaria suggest that the dogma may not be true, there was no such study with human P. falciparum parasites. In this study, the numbers of oocysts and sporozoite in the mosquitoes and the number of expelled sporozoites into artificial skin from the infected mosquito was quantified individually. There was a significant correlation between sporozoite burden in the mosquitoes and expelled sporozoites. In addition, this study showed that highly infected mosquitoes expelled sporozoites sooner.
Strengths:
The study was conducted using two different parasite-mosquito combinations; one was lab-adapted parasites with Anopheles stephensi and the other was parasites, which were circulated in infected patients, with An. coluzzii. Both combinations showed statistically significant correlations between sporozoite burden in mosquitoes and the number of expelled sporozoites.
Usually, this type of study has been done in group bases (e.g., count oocysts and sporozoites at different time points using different mosquitoes from the same group). However, this study determined the numbers in individual bases after multiple optimization and validation of the approach. This individual approach significantly increases the power of correlation analysis.
Weaknesses:
In a natural setting, most mosquitoes have less than 5 oocysts. Thus, the conclusion is more convincing if the authors perform additional analysis for the key correlations (Fig 3C and 4D) excluding mosquitoes with very high total sporozoite load (e.g., more than 5-oocyst equivalent load).
In the revised manuscript, we have also performed our analysis including only the subset of mosquitoes with low oocyst burden. In our Burkina Faso experiments, where we could not control oocyst density, 48% (15/31) of skins were from mosquitoes with <5 oocyst sheets. Whilst low oocyst densities were thus not very uncommon, we acknowledge that this may have rendered some comparisons underpowered. At the same time, we observe a strong positive trend between oocyst density and sporozoite density and between salivary gland sporozoite density and mosquito inoculum. This makes it very likely that this trend is also present at lower oocyst densities, an association where sporozoite inoculation saturates at high densities is plausible and has been observed before for rodent malaria (DOI: 10.1371/journal.ppat.1008181) whilst we consider it less likely that sporozoite expelling would be more efficient at low (unmeasured) sporozoite densities.
As written as the second limitation of the study, this study did not investigate whether all expelled sporozoites were equally infectious. For example, Day 9 expelled sporozoites may be less infectious than Day 11 sporozoites, or expelled sporozoites from high-burden mosquitoes may be less infectious because they experience low nutrient conditions in a mosquito. Ideally, it is nice to test the infectivity by ex vivo assays, such as hepatocyte invasion assay, and gliding assay at least for salivary sporozoites. But are there any preceding studies where the infectivity of sporozoites from different conditions was evaluated? Citing such studies would strengthen the argument.
We appreciate this thought and can see the value of these experiments. We are not aware of any studies that examined sporozoite viability in relation to the day of salivary gland colonization or sporozoite density.
One previous study assessed the NF54 sporozoite infectivity on different days post infection (days 12-13-14-15-16-18) and observed no clear differences in ‘per sporozoite hepatocyte invasion capacity’ over this period (DOI: 10.1111/cmi.12745). We nevertheless agree that it is conceivable that sporozoites require maturation in the salivary glands and might not all be equally infectious. While hepatocyte invasion experiments are conducted with bulk harvesting of all the sporozoites that are present in the salivary glands, it would even be more interesting to assess the invasion capacity of the smaller population of sporozoites that migrate to the proboscis to be expelled. This would, as the reviewer will appreciate, be a major endeavour. To do this well the expelled sporozoites would need to be harvested from the salivary glands/proboscis and used in the best and most natural environment for invasion. The suggested work would thus depend on the availability of primary hepatocytes since conventional cell-lines like HC-04 are likely to underestimate sporozoite invasion. Importantly, there are currently no opportunities to include the barrier of the skin environment in invasion assays whilst this may be highly important in determining the likelihood that sporozoites manage to achieve invasion and give rise to secondary infections. In short, we agree with the reviewer that these experiments are of interest but consider these well beyond the scope of the current work. We have added a section to the Discussion section to highlight these future avenues for research. ‘Of note, our assessments of EIP and of sporozoite expelling did not confirm the viability of sporozoites. Whilst the infectivity of sporozoites at different time-points post infection has been examine previously (https://doi.org/10.1111/cmi.12745), these experiments have never been conducted with individual mosquito salivary glands. To add to this complexity, such experiments would ideally retain the skin barrier that may be a relevant determinant for invasion capacity and primary hepatocytes.’
Since correlation analyses are the main points of this paper, it is important to show 95% CI of Spearman rank coefficient (not only p-value). By doing so, readers will understand the strengths/weaknesses of the correlations. The p-value only shows whether the observed correlation is significantly different from no correlation or not. In other words, if there are many data points, the p-value could be very small even if the correlation is weak.
We appreciate this comment and agree that this is indeed insightful. We have added the 95% confidence intervals to all figure legends and main text. We also provide them below.
Fig 3b: 95% CI: 0.74, 0.85
Fig 3c: 95% CI: 0.17, 0.50
Fig 4c: 95% CI: 0.80, 0.95
Fig 4d: 95% CI: 0.52, 0.82
Supp Fig 5a: 95% CI: 0.74, 0.85
Supp Fig 5b: 95% CI: 0.73, 0.93
Supp Fig 6: 95% CI: 0.11, 0.48
Supp Fig 7: 95% CI: -0.12, 0.16
Reviewer #2 (Public Review):
Summary: The malaria parasite Plasmodium develops into oocysts and sporozoites inside Anopheles mosquitoes, in a process called sporogony. Sporozoites invade the insect salivary glands in order to be transmitted during a blood meal. An important question regarding malaria transmission is whether all mosquitoes harbouring Plasmodium parasites are equally infectious. In this paper, the authors investigated the progression of P. falciparum sporozoite development in Anopheles mosquitoes, using a sensitive qPCR method to quantify sporozoites and an artificial skin system to probe for parasite expelling. They assessed the association between oocyst burden, salivary gland infection intensity, and sporozoites expelled.
The data show that higher sporozoite loads are associated with earlier colonization of salivary glands and a higher prevalence of sporozoite-positive salivary glands and that higher salivary gland sporozoite burdens are associated with higher numbers of expelled sporozoites. Intriguingly, there is no clear association between salivary gland burdens and the prevalence of expelling, suggesting that most infections reach a sufficient threshold to allow parasite expelling during a mosquito bite. This important observation suggests that low-density gametocyte carriers, although less likely to infect mosquitoes, could nevertheless contribute to malaria transmission.
Strengths: The paper is well written and the work is well conducted. The authors used two experimental models, one using cultured P. falciparum gametocytes and An. stephensi mosquitoes, and the other one using natural gametocyte infections in a field setup with An. coluzzii mosquitoes. Both studies gave similar results, reinforcing the validity of the observations. Parasite quantification relies on a robust and sensitive qPCR method, and parasite expelling was assessed using an innovative experimental setup based on artificial skin.
Weaknesses: There is no clear association between the prevalence of sporozoite expelling and the parasite burden. However, high total sporozoite burdens are associated with earlier and more efficient colonization of the salivary glands, and higher salivary gland burdens are associated with higher numbers of expelled sporozoites. While these observations suggest that highly infected mosquitoes could transmit/expel parasites earlier, this is not directly addressed in the study. In addition, whether all expelled sporozoites are equally infectious is unknown. The central question, i.e. whether all infected mosquitoes are equally infectious, therefore remains open.
We agree that the manuscript provides important steps forward in our understanding of what makes an infectious mosquito but does not conclusively demonstrate that highly infected mosquitoes are more likely to initiate a secondary infection. We consider this to be beyond the scope of the current work although the current work lays the foundation for these important future studies. For human Plasmodium infections the most satisfactory answer on the infectiousness of low versus high infected mosquitoes comes from controlled human infection models. In response to reviewer comments, we have extended our Discussion section to highlight this importance. To accommodate the (very fair) reviewer comments, we have avoided any phrasings that suggest that our findings demonstrate differences in transmission.
Reviewer #3 (Public Review):
Summary: This study uses a state-of-the-art artificial skin assay to determine the quantity of P. falciparum sporozoites expelled during feeding using mosquito infection (by standardised membrane feeding assay SMFA) using both cultured gametocytes and natural infection. Sporozoite densities in salivary glands and expelled into the skin are quantified using a well-validated molecular assay. These studies show clear positive correlations between mosquito infection levels (as determined by oocyst numbers), sporozoite numbers in salivary glands, and sporozoites expelled during feeding. This indicates potentially significant heterogeneity in infectiousness between mosquitoes with different infection loads and thus challenges the often-made assumption that all infected mosquitoes are equally infectious.
Strengths: Very rigorously designed studies using very well validated, state-of-the-art methods for studying malaria infections in the mosquito and quantifying load of expelled sporozoites. This resulted in very high-quality data that was well-analyzed and presented. Both sources of gametocytes (cultures vs. natural infection) show consistent results further strengthening the quality of the results obtained.
Weaknesses: As is generally the case when using SMFAs, the mosquito infections levels are often relatively high compared to wild-caught mosquitoes (e.g. Bombard et al 2020 IJP: median 3-4 ), and the strength of the observed correlations between oocyst sheet and salivary gland sporozoite load even more so between salivary gland sporozoite load and expelled sporozoite number may be dominated by results from mosquitoes with infection levels rarely observed in wild-caught mosquitoes. This could result in an overestimation of the importance of these well-observed positive relationships under natural transmission conditions. The results obtained from these excellently designed and executed studies very well supported their conclusion - with a slight caveat regarding their application to natural transmission scenarios
For efficiency and financial reasons, we have worked with an approach to enhance mosquito infection rates. If we had worked with gametocytes at physiological concentrations and a small number of donors, we probably have had considerably lower mosquito infection rates. Whilst this would indeed result in lower infection burdens in the sparse infected mosquitoes, addressing the reviewer concern, it would have made the experiments highly inefficient and expensive. The skin mimic was initially provided free of charge when the matrix was close to the expiry date but for the experiments in Burkina Faso we had to purchase the product at market value. Whilst we consider the biological question sufficiently important to justify this investment – and think our findings prove us right – it remained important to avoid using skins for uninfected mosquitoes. Since oocyst prevalence and density are strongly correlated (doi: 10.1016/j.ijpara.2012.09.002; doi: 10.7554/eLife.34463), a low oocyst density in natural infections typically coincides with a high proportion of negative mosquitoes.
Of note, our approach did result in the inclusion of 15 skins from infected mosquitoes with 1-4 oocysts. This number may be modest but we did include observations from this low oocyst range which is, we agree, highly important for better understanding malaria epidemiology.
This work very convincingly highlights the potential for significant heterogeneity in the infectiousness between individual P. falciparum-infected mosquitoes. Such heterogeneity needs to be further investigated and if again confirmed taken into account both when modelling malaria transmission and when evaluating the importance of low-density infections in sustaining malaria transmission.
Reviewer #4 (Public Review):
Summary: The study compares the number of sporozoites expelled by mosquitoes with different Plasmodium infection burden. To my knowledge this is the first report comparing the number of expelled P. falciparum sporozoites and their relation to oocyst burden (intact and ruptured) and residual sporozoites in salivary glands. The study provides important evidence on malaria transmission biology although conclusions cannot be drawn on direct impact on transmission.
Strengths: Although there is some evidence from malaria challenge studies that the burden of sporozoites injected into a host is directly correlated with the likelihood of infection, this has been done using experimental infection models which administer sporozoites intravenously. It is unclear whether the same correlation occurs with natural infections and what the actual threshold for infection may be. Host immunity and other host related factors also play a critical role in transmission and need to be taken into consideration; these have not been mentioned by the authors. This is of particular importance as host immunity is decreasing with reduction in transmission intensity.
Weaknesses: The natural infections reported in the study were not natural as the authors described. Gametocyte enrichment was done to attain high oocyst infection numbers. Studying natural infections would have been better without the enrichment step. The infected mosquitoes have much larger infection burden than what occurs in the wild.
Nevertheless, the findings support the same results as in the experiments conducted in the Netherlands and therefore are of interest. I suggest the authors change the wording. Rather than calling these "natural" infections, they could be called, for example, "experimental infections with wild parasite strains".
We have addressed these concerns and, in the process, also changed our manuscript title. The following sentences have been changed:
“It is currently unknown whether all Plasmodium falciparum infected mosquitoes are equally infectious. We assessed sporogonic development using cultured gametocytes in the Netherlands and natural infections in Burkina Faso”.
Now reads:“It is currently unknown whether all Plasmodium falciparum infected mosquitoes are equally infectious. We assessed sporogonic development using cultured gametocytes in the Netherlands and experimental infections with naturally circulating parasite strains in Burkina Faso”.226-228 “Experimental infections with naturally circulating parasite strains show comparable correlation between oocyst density, salivary gland density and sporozoite inoculum”.
Has now replaced the original phrasing:“Natural infected mosquitoes by gametocyte carriers in Burkina Faso show comparable correlation between oocyst density, salivary gland density and sporozoite inoculum”.
I do not believe the study results generate sufficient evidence to conclude that lower infection burden in mosquitoes is likely to result in changes to transmission potential in the field. In study limitations section, the authors say "In addition, our quantification of sporozoite inoculum size is informative for comparisons between groups of high and low-infected mosquitoes but does not provide conclusive evidence on the likelihood of achieving secondary infections. Given striking differences in sporozoite burden between different Plasmodium species - low sporozoite densities appear considerably more common in mosquitoes infected with P. yoelii and P. berghei the association between sporozoite inoculum and the likelihood of achieving secondary infections may be best examined in controlled human infection studies. However, in the abstract conclusion the authors state "Whilst sporozoite expelling was regularly observed from mosquitoes with low infection burdens, our findings indicate that mosquito infection burden is associated with the number of expelled sporozoites and may need to be considered in estimations of transmission potential." Kindly consider ending the sentence at "expelled sporozoites." Future studies on CHMI can be recommended as a conclusion if authors feel fit.
We agree that we need to be very cautious with conclusions on the impact of our findings for the infectious reservoir. We have rephrased parts of our abstract and have updated the Discussion section following the reviewer suggestions. We agree with the reviewer that CHMI studies are recommended and have expanded the Discussion section to make this clearer. The sentence in the abstract now ends as:
"Whilst sporozoite expelling was regularly observed from mosquitoes with low infection burdens, our findings indicate that mosquito infection burden is associated with the number of expelled sporozoites. Future work is required to determine the direct implications of these findings for transmission potential."
Reviewer #1 (Recommendations For The Authors):
• Prevalence data shown in Fig 2A and Table S1 are different. For example, >50K at Day 11, Fig 2A shows ~85% prevalence, but Table S1 says 100%. If the prevalence in Table S1 shows a proportion of observations with positive expelled sporozoites (instead of a proportion of positive mosquitoes shown in Fig 2A), then the prevalence for <1K at Day 11 cannot be 6.7% (either 0 or 20% as there were a total of 5 observations). So in either case, it is not clear why the numbers shown in Fig 2A and Table S1 are different.
Figure 2A and Table S2 are estimated prevalence and odds ratios from an additive logistic regression model (i.e. excluding the interaction between day and sporozoite categories). Table S1 includes this interaction when estimating prevalence and odds ratios and as we can see some categories in the interaction were extremely small resulting in blown up confidence intervals especially in day 11. So Table S1 and Fig 2A are the results from two different models. Whilst our results are thus correct, we can understand the confusion and have added a sentence to explain the model used in the figure/table legends.
Figure. 2 Extrinsic Incubation Period in high versus low infected mosquitoes. A. Total sporozoites (SPZ) per mosquito in body plus salivary glands (x-axis) were binned by infection load <1k; 1k-10k; 10k-50k; >50k and plotted against the proportion of mosquitoes (%) that were sporozoite positive (y-axis) as estimated from an additive logistic regression model with factors day and SPZ categories.Supplementary Table S1. The extrinsic incubation period of P. falciparum in An. stephensi estimated by quantification of sporozoites on day 9, 10, 11 by qPCR. Based on infection intensity mosquitoes were binned into four categories (<1k, 1k-10k, 10k-50k, >50) that was assessed by combining sporozoite densities in the mosquito body and salivary gland. Prevalences and odds ratios were estimated from a logistic regression model with factors day, SPZ category and their interaction.
There are 3 typos in the paper. Please fix them.
Line 464; ...were counted using a using an incident....
Line 473; Supplementary Figure 7 should be Fig S8.
Line 508: ...between days 9 and 10 using a (t=-2.0467)....
We appreciate the rigour in reviewing our text and have corrected all typos.
Reviewer #2 (Recommendations For The Authors):
High infection burdens may result in earlier expelling capacity in mosquitoes, which would reflect more accurately the EIP. The fact that earlier colonization of SG and correlation between SG burden and numbers expelled suggest it could be the case, but it would be interesting to directly measure the prevalence of expelling over time to directly assess the effect of the sporozoite burden (not just at day 15 but before). This could reveal how the parasite burden in mosquitoes is a determinant of transmission.
We appreciate this suggestion and will consider this for future experiments. It adds another variable that is highly relevant but will also complicate comparisons where sporozoite expelling is related to both time since infectious blood meal and salivary gland sporozoite density (that is also dependent on time since infectious bloodmeal). Moreover, we then consider it important to measure this over the entire duration of sporozoite expelling, including late time-points post infectious bloodmeal. This may form part of a follow-up study.
Another question is whether all sporozoites (among expelled parasites) are equally infective, i.e. susceptible to induce secondary infection. If not, this could reconcile the data of this study and previous results in the rodent model where high burdens were associated with an increased probability to transmit.
As also indicated above, we are aware of a single study that assessed NF54 sporozoite infectivity on different days post infection (days 12-13-14-15-16-18) and observed no clear differences in ‘per sporozoite hepatocyte invasion capacity’ over this period (DOI: 10.1111/cmi.12745). We nevertheless agree that it is conceivable that sporozoites require maturation in the salivary glands and might not all be equally infectious. While hepatocyte invasion experiments are conducted with bulk harvesting of all the sporozoites that are present in the salivary glands, it would even be more interesting to assess the invasion capacity of the smaller population of sporozoites that migrate to the proboscis to be expelled. This would, as the reviewer will appreciate, be a major endeavour. To do this well the expelled sporozoites would need to be harvested from the salivary glands/proboscis and used in the best and most natural environment for invasion. The suggested work would thus depend on the availability of primary hepatocytes since conventional cell-lines like HC-04 are likely to underestimate sporozoite invasion. Importantly, there are currently no opportunities to include the barrier of the skin environment in invasion assays whilst this may be highly important in determining the likelihood that sporozoites manage to achieve invasion and give rise to secondary infections. In short, we agree with the reviewer that these experiments are of interest but consider these well beyond the scope of the current work. We have added a section to the Discussion section to highlight these future avenues for research. ‘Of note, our assessments of EIP and of sporozoite expelling did not confirm the viability of sporozoites. Whilst the infectivity of sporozoites at different time-points post infection has been examine previously (ref), these experiments have never been conducted with individual mosquito salivary glands. To add to this complexity, such experiments would ideally retain the skin barrier that may be a relevant determinant for invasion capacity and primary hepatocytes.’
The authors evaluated oocyst rupture at day 18, i.e. 3 days after feeding experiments (performed at day 15). Did they check in control experiments that the prevalence of rupture oocysts does not vary between day 15 and day 18?
We did not do this and consider it very unlikely that there is a noticeable increase in the number of ruptured oocysts between days 15 and 18. We observe that salivary gland invasion plateaus around day 12 and the provision of a second bloodmeal that is known to accelerate oocyst maturation and rupture (doi: 10.1371/journal.ppat.1009131) makes it even less likely that a relevant fraction of oocysts ruptures very late. Perhaps most compellingly, the time of oocyst rupture will depend on nutrient availability and rupture could thus occur later for oocysts from a heavily infected gut compared to oocysts from mosquitoes with a low infection burden. We observe a very strong association between salivary gland sporozoite density (day 15) and oocyst density (assessed at day 18) without any evidence for change in the number of sporozoites per oocyst for different oocyst densities. In our revised manuscript we have also assessed correlations for different ranges of oocyst intensities and see highly consistent correlation coefficients and find no evidence for a change in ‘slope’. If oocyst rupture would regularly happen between days 15 and 18 and this late rupture would be more common in heavily infected mosquitoes, we would expect this to affect the associations presented in figures 3B and 4C This is not the case.
The authors report higher sporozoite numbers per oocyst and a higher proportion of SG invasion as compared to previous studies (30-50% rather than 20%). How do they explain these differences? Is it due to the detection method and/or second blood meal? Or parasite species?
We were also intrigued by these findings in light of existing literature. To address potential discrepancies, it is indeed possible that the 2nd bloodmeal made a difference. In addition, NF54 is known to be a highly efficient parasite in terms of gametocyte formation and transmission. And there are marked differences in these performances between NF54 isolates and definitely between NF54 and its clone 3D7 that is regularly used. We also used a molecular assay to detect and quantify sporozoites but consider it less likely that this is a major factor in terms of explaining SG invasion since sporozoite densities were typically within the range that would be detected by microscopy. We can only hypothesize that the 2nd bloodmeal may have contributed to these findings and acknowledge this in the revised Discussion section.
The median numbers of expelled sporozoites seem to be higher in the natural gametocyte infection experiments as compared to the cultures. Is it due to the mosquito species (An. coluzzii versus An. stephensi?).
The added value of our field experiments, a more relevant mosquito species and more relevant parasite isolates, is also a weakness in terms of understanding possible differences between in vitro experiments and field experiments with naturally circulating parasite strains. We only conclude that our in vitro experiments do not over-estimate sporozoite expelling by using a highly receptive mosquito source and artificially high gametocyte densities. We have clarified this in the revised Discussion.
39% of sporozoite-positive mosquitoes failed to expel, irrespective of infection densities. Could the authors discuss possible explanations for this observation?
In paragraph 304-307 we now write that:” This finding broadly aligns with an earlier study of Medica and Sinnis that reported that 22% of P. yoelii infected mosquitoes failed to expel sporozoites. For highly infected mosquitoes, this inefficient expelling has been related to a decrease of apyrase in the mosquito saliva”.
In Figure 3, it would be interesting to zoom in the 0-1k window, below the apparent threshold for successful expelling.
We have generated correlation estimates for different ranges of oocyst and sporozoite densities and added these in Supplementary Table 5. We agree that this helps the reader to appreciate the contribution of different ranges of parasite burden to the observed associations.
In Fig S8. Did they observe intact oocysts with fixed samples? These could be shown as well in the figure.
We have incorporated this comment. An intact oocyst from fixed samples was now added to Fig S10.
Minor points
-line 119: LOD and LOQ could be defined here.
We agree that this should have been defined. We changed line 119 to explain LOD and LOQ to: …“the limit of detection (LOD) and limit of quantification (LOQ)”….
line 126: the title does not reflect the content of this paragraph.
We have changed the title: “Immunolabeling allows quantification of ruptured oocysts ”into: A comparative analysis of oocyst densities using mercurochrome staining and anti-CSP immunostaining.
-line 269: infectivity is not appropriate. The data show colonization of SG.
Line 269: infectivity has been changed with colonization of salivary glands.
There seems to be a problem with Fig S6. The graph seems to be the same as Fig 3C. Please check whether the graph and legends are correct.
Supplementary Figure 6 shows the sporozoite expelling density in relation to infection burden with a threshold set at > 20 sporozoites while Fig 3C shows the total sporozoite density (residual salivary gland sporozoites + sporozoites expelled, X-axis) in relation to the number of expelled sporozoites (Y-axis) by COX-1qPCR without any threshold density. We have explained this in more detail in the revised supplemental figure where we now state
“Of note, this figure differs from Figure 3C in the main text in the following manner. This figure presents sporozoite expelling density in relation to infection burden with a threshold set at > 20 sporozoites to conclude sporozoite positivity while Figure 3C shows the total sporozoite density (residual salivary gland sporozoites + sporozoites expelled, X-axis) in relation to the number of expelled sporozoites (Y-axis) by COX-1 qPCR without any threshold density and thus includes all observations with a qPCR signal”
Reviewer #3 (Recommendations For The Authors):
Congratulations to the authors for the really excellently designed and rigorously conducted studies.
My main concern is in regards to the relatively high oocyst numbers in their experimental mosquitoes (from both sources of gametocytes) compared to what has been reported from wild-caught mosquitoes in previous studies in Burkina Faso.
We have addressed this concern above. For completeness, we include the main points here again. We enriched gametocytes for efficiency reasons, experiments on gametocytes at physiological concentrations would have resulted in a lower oocyst density and thus more ‘natural’ although a minority of individuals achieves very high oocyst densities in all studies that included a broad range of oocyst densities e.g. doi: 10.1016/j.exppara.2014.12.010; doi: 10.1016/S1473-3099(18)30044-6. Of note, we did include 15 skins from low oocyst densities (1-4 oocysts). Whilst low oocyst densities were thus not very uncommon in our sample set, we acknowledge that this may have rendered some comparisons underpowered. At the same time, we observe a strong positive trend between oocyst density and sporozoite density and between salivary gland sporozoite density and mosquito inoculum. This makes it very likely that this trend is also present at lower oocyst densities, an association where sporozoite inoculation saturates at high densities is plausible and has been observed before for rodent malaria (DOI: 10.1371/journal.ppat.1008181) whilst we consider it less likely that sporozoite expelling would be more efficient at low (unmeasured) sporozoite densities. In the revised manuscript we have also performed our analysis including only the subset of mosquitoes with low oocyst burden.
The best way to address this would be to do comparable artificial skin-feeding experiments on such wild-caught mosquitoes, but I appreciate that this is very difficult to do.
This would indeed by difficult to do. Mostly because infection status can only be examined post-hoc and it is likely that >95% of mosquitoes are sporozoite negative at the moment experiments are conducted (in many settings this will even be >99%). Importantly, also in wild-caught mosquitoes very high oocyst burdens are observed in a small but relevant subset of mosquitoes (doi: 10.1016/j.ijpara.2020.05.012).
Instead, I would suggest the authors conduct addition analysis of their data using different cut-offs for maximum oocyst numbers (e.g. <5, <10, <20) to determine if these correlations hold across the entire range of observed oocyst sheets and salivary gland sporozoite load.
We have provided these calculations for the proposed range of oocyst numbers. In addition, we also provided them for a range of sporozoite densities. These findings are now provided in
Entire range of observed oocyst sheets and salivary gland sporozoite load. A minor point on the regression lines in Figures 3 & 4: both variables in these plots have inherent variation (measurement & natural), but regression techniques such as reduced major exit regression (MAR) that allow error in both x and y variables may be preferable to a standard lines regression. Also, as it is implausible that mosquitoes with zero sporozoite in salivary glands expel several hundred sporozoites at feeding, the regression should probably also be constrained to pass through the 0,0 point.
Since the main priority of the analyses is the correlation, and not the fit of the regression line – which is only for indication, and also because of the availability of software, we did not change the type of regression. We have however added a disclaimer to the legend, and we have also forced the intercept to 0 – which does indeed better reflect the biological association. Additionally we added 95% confidence intervals to all Spearman’s correlation coefficients in the legends.