The relevance and applicability of oocyst prevalence as a read-out for mosquito feeding assays

Abstract

Mosquito feeding assays are important in evaluations of malaria transmission-reducing interventions. The proportion of mosquitoes with midgut oocysts is commonly used as an outcome measure, but in natural low intensity infections the effect of oocyst non-rupture on mosquito infectivity is unclear. By identifying ruptured as well as intact oocysts, we show that in low intensity P. falciparum infections i) 66.7-96.7% of infected mosquitoes experienced oocyst rupture between 11-21 days post-infection, ii) oocyst rupture led invariably to sporozoite release, iii) oocyst rupture led to salivary gland infections in 97.8% of mosquitoes, and iv) 1250 (IQR 313-2400) salivary gland sporozoites were found per ruptured oocyst. These data show that infectivity can be predicted with reasonable certainty from oocyst prevalence in low intensity infections. High throughput methods for detecting infection in whole mosquitoes showed that 18s PCR but not circumsporozoite ELISA gave a reliable approximation of mosquito infection rates on day 7 post-infection.

Figure 1. Oocyst prevalence and intensity for experimental feeds utilising the NF54 P. falciparum strain conducted in the mosquito infectivity unit at RUMC, Nijmegen between 2011 and 2013.

Each data point represents the mean oocyst intensity and oocyst prevalence in a single group of mosquitoes. Criteria for inclusion in these figures were sample size (>10 mosquitoes), parasite strain (NF54 strain only) and mosquito species. Data were collected from both experimental and control feeds between 6 and 9 days PI. In total 1323 separate experimental feeds comprising 21240 mosquito dissections are shown (1233 experimental feeds with A. stephensi, 90 experimental feeds with A. gambiae).

Figure 2. Classification of oocyst condition.

Mosquito midguts were dissected in phosphate buffered saline (PBS) and stained using 3SP2-Alexa488 anti-CSP antibodies. After staining, midguts were washed twice with PBS for 10 minutes before being sealed under a glass cover slip with Vaseline petroleum jelly. All oocysts were identified and assigned a condition by two independent microscopists. Intact oocysts were visibly unbroken. Ruptured oocysts and oocyst remnants were visibly broken or degraded. All samples were examined within 8 hours of dissection, but sample condition remained stable for a number of days (stored at 4°C) prior to dissection for re-examination. Three ruptured and three intact oocysts are shown in the figure for demonstrative purposes.

Figure 3. Total oocyst prevalence and prevalence of oocyst rupture.

n/N = oocyst positive mosquitoes/total number of mosquitoes dissected. Prevalence of oocyst rupture is given as the proportion of mosquitoes in which any oocysts were observed to have undergone rupture. Tables indicate the range and frequency of oocyst intensities observed in experiment 1 and 2. In experiment 1, 16/22 (72.7%, 95% CI = 49.8–89.3%) infected mosquitoes had at least one ruptured oocyst at day 14 PI, while 12/18 (66.7%, 95% CI = 41–86.7%) were rupture positive at day 21 PI. In experiment 2, 30/32 (93.8%, 95% CI 79.2–99.2%) infected mosquitoes were rupture positive at day 14 PI, while 27/30 (90%, 95% CI = 73.5–97.9%) were rupture positive at day 21 PI.

Figure 4. The number of sporozoites in the mosquito body or salivary glands, and ruptured oocyst intensity in the same mosquitoes.

The detection limit for sporozoites detected in a single mosquito sample was 78 sporozoites (1 sporozoite observed in 64 Bürker chamber fields/0.256 μl homogenate). The Y-axis is split from 1–3000 sporozoites and from 3000–50,000 sporozoites because of the heterogeneous distribution of sporozoites within mosquitoes. Data points in red indicate erroneous observations (sporozoite +, rupture – or vice versa). a. Y-axis: Sporozoite total calculated from duplicate samples of whole homogenised mosquito (excluding gut) stained with 3SP2-Alexa488 conjugate antibodies. X-axis: Ruptured oocyst intensity. Median oocyst intensity in infected mosquitoes was 2 (IQR 1–4, range 1–10). The data point in red is a single mosquito for which 1500 sporozoites were found associated with no ruptured oocysts. b. Y-axis: Sporozoite total calculated from duplicate samples of homogenised 3SP2-Alexa488 stained salivary glands only. X-axis: Ruptured oocyst intensity. Median oocyst intensity in infected mosquitoes was 2 (IQR 1–3, range 1–9). The data points in red are a mosquito in which 320 sporozoites were found associated with no ruptured oocysts, and a mosquito in which a single ruptured oocyst was found associated with no sporozoites.