Cutaneous volatile and semi-volatile organic compounds as markers of malaria-infection by wearable samplers and two-dimensional gas chromatography—time-of-flight-mass spectrometry

Abstract

Malaria has been found to alter normal cutaneous volatile organic compound (VOC) profiles, suggesting their potential application as markers of Plasmodium infection. The cutaneous VOCs and semi-VOCs (SVOCs) of malaria-negative and -positive individuals, who visited two local clinics in the Vhembe district of Limpopo Province, South Africa, were extracted into wearable silicone rubber (polydimethyl siloxane [PDMS]) sampling bands adhered to the surface of the epidermis. After sampling of epidermal VOCs from participants the samplers were analysed by thermal desorption-comprehensive two-dimensional gas chromatography-time-of-flight-mass spectrometry (TD-GC × GC-TOFMS). Individual cutaneous VOCs and SVOCs profiles were constructed from these complex chromatographic profiles in order to identify potential signatures of Plasmodium infection. Fatty acid compounds associated with rancid malodour, and previously reported as mosquito attractants, were found at an overall greater abundance in chemical profiles of malaria-positive cases. A targeted analysis was performed for compounds previously reported to be associated with Plasmodium infection, viz., heptanal, (E)-2-octenal, 2-octanone, octanal, nonanal and (E)-2-decenal. The linearity (R2) range was 0.93–0.99 for a matrix matched (simulated cutaneous sampling) calibration range of 2.5–60 ng. Limits of detection (LOD) ranged from 0.4 pg (2-octanone) to 6.3 pg ((E)-2-octenal), whilst limits of quantification (LOQ) ranged from 1.4 pg to 21.1 pg. The mean percentage recoveries (n = 2) ranged from 77.8 % ((E)-2-decenal) to 118.9 % (2-octanone). The percentage relative standard deviations ( %RSDs; n = 2) ranged from < 1 % for 2-octanone, octanal and nonanal to 27.1 % for (E)-2-octenal. We found that this particular suite of compounds, previously reported as indicators of malaria, was in fact non-specific for Plasmodium infection when compared to control subjects with comorbidities. A previously unreported (in a malaria-infection context) compound, (E)-2-octen-1-ol, correlated with malaria-positive participants, but was also observed for two malaria-negative participants, which could indicate latent malaria. In chronic cases, Plasmodium vivax can occur in reservoirs outside of the bloodstream, and thus blood-based diagnostic tests can miss latent infection. A key advantage of the epidermal sampler over blood tests is that the former collects whole-body organic compounds, and is therefore not limited to blood-borne markers of infection. As such it appears to be feasible for future investigations.