Characterising co-infections with Plasmodium spp., Mansonella perstans or Loa loa in asymptomatic children, adults and elderly people living on Bioko Island using nucleic acids extracted from malaria rapid diagnostic tests

Abstract

Background: Regular and comprehensive epidemiological surveys of the filarial nematodes Mansonella perstans and Loa loa in children, adolescents and adults living across Bioko Island, Equatorial Guinea are lacking. We aimed to demonstrate that blood retained on malaria rapid diagnostic tests, commonly deployed for malaria surveys, could be used as a source of nucleic acids for molecular based detection of M. perstans and L. loa. We wanted to determine the positivity rate and distribution of filarial nematodes across different age groups and geographical areas as well as to understand level of co-infections with malaria in an asymptomatic population.

Methodology: M. perstans, L. loa and Plasmodium spp. parasites were monitored by qPCR in a cross-sectional study using DNA extracted from a subset malaria rapid diagnostic tests (mRDTs) collected during the annual malaria indicator survey conducted on Bioko Island in 2018.

Principal findings: We identified DNA specific for the two filarial nematodes investigated among 8.2% (263) of the 3214 RDTs screened. Positivity rates of M. perstans and L. loa were 6.6% and 1.5%, respectively. M. perstans infection were more prominent in male (10.5%) compared to female (3.9%) survey participants. M. perstans parasite density and positivity rate was higher among older people and the population living in rural areas. The socio-economic status of participants strongly influenced the infection rate with people belonging to the lowest socio-economic quintile more than 3 and 5 times more likely to be L. loa and M. perstans infected, respectively. No increased risk of being co-infected with Plasmodium spp. parasites was observed among the different age groups.

Conclusions/significance: We found otherwise asymptomatic individuals were infected with M. perstans and L. loa. Our study demonstrates that employing mRDTs probed with blood for malaria testing represents a promising, future tool to preserve and ship NAs at room temperature to laboratories for molecular, high-throughput diagnosis and genotyping of blood-dwelling nematode filarial infections. Using this approach, asymptomatic populations can be reached and surveyed for infectious diseases beyond malaria.

Fig 1. Positivity rates of M. perstans and L. loa by rural and urban districts.

(A) across all age groups investigated, (B) grouped according to socio-economic status and (C) stratified by gender. Positivity rates and 95% confidence intervals were calculated as the proportion of llmp-qPCR positive mRDTs in all tests carried out in each group and are given on top of each bar. Data supporting the Fig 1 are detailed in S2 Table.

Fig 2. Phylogenetic tree of the ribosomal internal transcribed spacer 1 region of M. perstans and L. loa.

PCR products were amplified from DNA extracted from mRDTs and sequenced. The scale on the x-axis corresponds to the number of substitutions per site (number of changes/sequence length).

Fig 3. Positivity rate of Plasmodium spp, M. perstans and L. loa across Bioko Island.

(A) Prevalence of PfHRP2-positive mRDTs for the entire MIS 2018 population. (B) Plasmodium spp. RT-qPCR positivity rate of mRDTs selected for molecular analysis. (C) M. perstans qPCR based positivity rate. (C) L. loa qPCR based positivity rate. The size of the squares represents the number of people analysed living in the corresponding 1x1 km² grid. The areas marked in green are nature reserves. Greyed out spots on the maps represent settlements from which mRDTs were collected with null positivity rate for Plasmodium spp, M. perstans and L. loa.

Fig 4. Plasmodium spp., M. perstans and L. loa multi-parasitism infections on Bioko Island.

(A) Number of M. perstans, L. loa and Plasmodium spp. co-infections amongst 2775 individuals tested. (B) Positivity rates of Plasmodium spp., M. perstans and L. loa co-infections, stratified by age group. The blue lines and orange lines depict the observed and predicted co-infection rates, respectively. (C) Geographical distribution of M. perstans co-infections with malaria and L. loa. (D) Geographical distribution of L. loa co-infections with malaria and M. perstans. In (C) and (D) only 1x1 km² grids with at least one case of M. perstans or L. loa infection are presented. The areas marked in green are nature reserves.

Fig 5. Comparison of Cq values obtained from M. perstans and L. loa infected individuals aged above and below 40 years.

(A) Scatter plots of Cq values for L. loa, M. perstans qPCR and the corresponding Cq values for the human RNaseP gene qPCR. The geometric mean values for each group are shown. (B) Cumulative frequency of Cq values for M. perstans infected individuals. At a Cq value of 33 (dashed line), the cumulative frequency among individuals aged above 40 years is 33.6% compared to 15.5% among individuals aged below 40 years. (C) Histogram of the distribution of Cq values for M. perstans infected individuals stratified according to age.