DEVELOPMENT OF A RAPID URINE-BASED DIPSTICK TEST FOR DIAGNOSIS OF MALARIA

SUMMARY

Bloodsmear microscopy is currently the gold standard and the principal test for confirmatory diagnosis of malaria. However, microscopy is labour-intensive and is limited by inadequate sensitivity and specificity.

Apart from microscopy, rapid diagnostic tests (RDTs), which include the quantitative buffy coat test (QBC) and the immuno-chromatographic tests (ICTs) are alternative tests for confirmatory diagnosis of malaria. RDTs, particularly, ICTs are fast gaining momentum as alternative tests for routine diagnosis of the disease. These RDTs are fast and have yielded remarkable diagnostic outputs in certain localities, however, they are invasive, and sometimes, have limited sensitivity and/ or specificity.

The work reported in this thesis was performed in order to generate monoclonal antibodies (MAb) towards the development of rapid urine-based assay (RUBDA) for noninvasive diagnosis of malaria, through detection of parasite antigens in the urine of infected individuals. Such an antibody-based test, is hoped to serve as a promising alternative for diagnosis of the disease, especially, for home-based management of malaria, particularly, in the hard-to-reach, low-income communities where the morbidity and mortality are highest.

Urine and blood samples were collected from Kpone-on-Sea (KOS), a malaria endemic coastal fishing village in south-eastern Ghana. KOS is being developed as a model site for malaria intervention studies in a larger study titled the Kpone Malaria Project (KMP). Blood smear microscopy, quantitative buffy coat (QBC) test and commercial rapid diagnostic test (RDT) kits were used to distinguish between malaria infected and uninfected blood and urine samples. This was necessary, partly to select infected urinary antigens for immunizations towards MAb generation. Also, it was necessary to select true negative control urine samples for ascertaining the specificity of the MAb.

Analysis of the results showed a high malaria prevalence of 46.5% among the study population. Also, the results showed that malaria was significantly associated with sex (P = 0.0001), age (P= 0.0001), and fever (P = 0.0001). The results however showed that malaria was not associated with anaemia (P = 0.173).

The observation of significantly higher prevalence of malaria and higher parasite density among children than adults in the studies were important findings that confirmed reports by other writers that children are more susceptible to the disease than adults. In addition, the occurrence of higher prevalence of malaria among females in this community, even though it confirmed what Ayele et al. (2012) reported, was suspected to be linked to behavior more than to sex, as reported by other authors (Steketee et al., 2001). Immunological and biochemical analyses needed to ascertain the link between malaria and sex were not carried out so it was difficult to conclude that malaria was sex-linked.

Immunization of mice towards MAb generation showed that 30 μg of urinary antigen was more immunogenic than 15 μg. Mice immunized with 15 μg of the antigen showed weak immune response, with absorbance values ranging between 0.36 and 0.45, measured at 414nm. A majority (80%) of these mice did not show any observable response at all. In contrast, almost all (93.3%) of the 15 mice immunized with 30 μg of immunogen, showed high immune response (absorbance = 2.75) with only one (6.7%) not showing any response at all. Also, the Plasmodium-infected human urinary antigen extract (PAgHU) was found to be more immunogenic than the cultured parasite antigen extract (CPfAg) in these experiments. In all, 96 MAb clones were successfully generated against urinary and cultured Plasmodium parasite antigens. Determination of the immunoglobulin class of some of the MAb clones produced showed that 57.14% were of the IgM and 42.86% were of the IgG isotype. Characterization of these MAbs by microplate ELISA showed that they were all (100%) reactive to urinary Plasmodium parasite antigen extract (PAgHU). Also, 66 (68.8%) of these clones were reactive to cultured parasite antigens (CPfAg), 30 (31.3%) reacted specifically to only PAgHU, and none reacted with CPfAg alone. One of the MAb clones (UCP4W7), showed strong reactivity in ELISA and immunoglobulin class analysis. This UCP4W7 MAb was consequently selected and evaluated by microplate ELISA for sensitivity and specificity in detecting malaria antigens in the urine of a cohort of 420 randomly selected individuals from the study community. In this experiment, using microscopy as a gold standard test, the UCP4W7 MAb distinguished between infected and uninfected urine with a relative sentivity and specificity of 96.9 and 75.6%, respectively. Compared to the UCP4W7 MAb, the mouse anti-Plasmodium PAb even though it had a relatively higher sensitivity (97.4%) in microplate ELISA the specificity was extremely lower (21.3%).

Also, determination of the relative sensitivities and specificities of the tests for various biomarkers of malaria compared to microscopy showed that tests for ketones, microhematuria, leukocytes, nitrites, glucose, bilirubinuria urobilinogen as well as PAb ELISA, were highly specific but had very low sensitivities. Each of these tests had a combined reactivity (Relative sensitivity + specificity) being less than the 170% threshold required to qualify it as a promising biomarker for diagnosis of malaria. Characterization of proteins in PAgHU, CPfAg and other antigen preparations by SDSPAGE showed that PAgHU contained more protein bands in the profile than the rest of the antigen preparations. The results also showed that the number of bands in the profiles of various urine samples from infected individuals decreased with decreasing parasite density. These findings demonstrated the likelihood of association between proteinuria and parasite density, and therefore, suggested that perhaps resolution of urinary Plasmodium proteins by SDS-PAGE could be used to quantify parasite density in infected individuals.

Further characterization of the urinary malarial antigens specific to UCP4W7, using microplate ELISA, SDS-PAGE, western blotting assay, mass spectrometry sequencing (MSS) and commercial RDT kits, showed that some of these antigens comprised the P. falciparum species-specific histidine-reach protein 2 (HRP2), Plasmodium pan-specific lactate dehydrogenase (LDH) antigen, the anti-malaria human metabolite peptide hormone (hepcidin), as well as other malarial proteins whose identities are yet to be determined. The western blotting assay also revealed that the selected UCP4W7 MAb could distinguish between Plasmodium infected and uninfected human urine. In addition, it was found that the MAb could differentiate between urinary and cultured Plasmodium parasite antigens by western blotting assay. Furthermore, the western blotting analysis demonstrated that UCP4W7 could distinguish specifically between urinary Plasmodium antigens and antigens from Schistosoma parasites as well as poliomyelitis and measles vaccines. These results showed that speciesspecific MAbs against Plasmodium parasite proteins in urine would be promising for diagnosis of malaria and should therefore be explored further for development of accurate non-invasive tests for diagnosis of the disease.

The results, however, showed that the UCP4W7 MAb could not distinguish between malaria antigens in urine and yellow fever vaccine antigens because of non-specific reactivity. On the other hand, cross-reactivity of the MAbs with yellow fever vaccine suggests that the Plasmodium-infected urine samples used in generating the MAbs, also had yellow fever antigens possibly originating from vaccination and/ or active infections with the virus. In this respect, it was thought that ruling out yellow fever vaccination or infections from malaria suspected urine samples, would enable the UCP4W7 MAb to be used in its current state to distinguish between malaria infected and uninfected persons.

Finally, examination of urine samples from malaria infected and uninfected individuals by MSS, demonstrated that urinary hepcidin was associated with malaria parasitaemia in infected individuals, but was not associated with anaemia. Hepcidin is a peptide hormone produced by chronic inflammation in affected individuals. Since malaria causes inflammation in infected persons, the results were thought to suggest that urinary hepcidin would be a promising biomarker for malaria. Ultimately, these results therefore suggested that urinary hepcidin could be used to develop a non-invasive test for diagnosis of malaria.