Haemoglobin interference and increased sensitivity of fluorimetric assays for quantification of low-parasitaemia Plasmodium infected erythrocytes

Abstract

Background: Improvements on malarial diagnostic methods are currently needed for the correct detection in low-density Plasmodium falciparum infections. Microfluorimetric DNA-based assays have been previously used for evaluation of anti-malarial drug efficacy on Plasmodium infected erythrocytes. Several factors affecting the sensitivity of these methods have been evaluated, and tested for the detection and quantification of the parasite in low parasitaemia conditions.

Methods: Parasitaemia was assessed by measuring SYBRGreen I (SGI) and PicoGreen (PG) fluorescence of P. falciparum Dd2 cultures on human red blood cells. Different modifications of standard methods were tested to improve the detection sensitivity. Calculation of IC50 for chloroquine was used to validate the method.

Results: Removal of haemoglobin from infected red-blood cells culture (IRBC) increased considerably the fluorescent signal obtained from both SGI and PG. Detergents used for cell lysis also showed to have an effect on the fluorescent signal. Upon depletion of haemoglobin and detergents the fluorescence emission of SGI and PG increased, respectively, 10- and 60-fold, extending notably the dynamic range of the assay. Under these conditions, a 20-fold higher PG vs. SGI fluorescent signal was observed. The estimated limits of detection and quantification for the PG haemoglobin/detergent-depleted method were 0.2% and 0.7% parasitaemia, respectively, which allow the detection of ~10 parasites per microliter. The method was validated on whole blood-infected samples, displaying similar results as those obtained using IRBC. Removal of white-blood cells prior to the assay allowed to increase the accuracy of the measurement, by reducing the relative uncertainty at the limit of detection from 0.5 to 0.1.

Conclusion: The use of PG microassays on detergent-free, haemoglobin-depleted samples appears as the best choice both for the detection of Plasmodium in low-density infections and anti-malarial drugs tests.

Figure 1

DNA titration using SYBR^®Green I and PicoGreen^® dyes. PicoGreen^® fluorescence (A) and SYBR^®Green I fluorescence (B) with bacteriophage λ DNA in absence (black circle, solid line), or presence (dotted lines) of different amounts of detergents: saponin 0.008% + Triton X-100 0.08% (white triangle); saponin 0.008% (white square); Triton X-100 0.08% (white diamonds); Triton X-100 2% (white circle). Background fluorescence, defined as fluorescence detected in the absence of DNA, was subtracted from each data point. Fluorescence is measured as arbitrary units (AU). Data show average from three replicate experiments. Error bars indicate standard deviations. Lines were calculated by linear regression; r² >0.99.

Figure 2

Effect of haemoglobin on fluorescence emission. A) Fluorescence emission of SYBR^®Green I (solid bars) and PicoGreen^® (empty bars) of 1 μg/mL DNA, (485/2.5 nm excitation, 528/2.5 nm emission) in the presence of different amounts of red blood cells. The inset shows the fluorescence signal at 485/15 nm excitation and 528/15 nm emission. Discontinuous bars indicate off-scale fluorescent reading. Fluorescence at 0% haematocrit was obtained from a 1 μg/mL solution of λDNA in culture medium. B) Absorption spectrum of red blood cell lysates.

Figure 3

Chloroquine dose-response curve. Chloroquine responses of P. falciparum in erythrocyte cultures obtained using PG (A) and SGI (B), after removal of haemoglobin/detergents. Sigmoid graphs show the survival of P. falciparum Dd2 at different concentrations of drug. Bars indicate the standard errors of the mean for two independently processed samples. The calculated IC₅₀ values were 136.2 ± 28 nM for PG and 118 ± 15 nM for SGI.

Figure 4

Correlation between parasitaemia and fluorescence emission. PG (A) and SGI (B) fluorescence detected in Plasmodium falciparum infected erythrocytes at different parasitaemias. A serial fold dilution of synchronized infected culture (15% mature ring stage) with non-infected erythrocytes was used. Results are shown for haemoglobin non-depleted samples (black square), and haemoglobin depleted samples in the absence (white square) or presence (white circle) of 2% Triton X-100. Bars indicate the standard deviation of the mean from three independently processed samples.

Figure 5

Relative uncertainty of fluorescent measurements of PG-DNA adducts in low parasitaemia samples. Plot of the relative uncertainty calculated form the fluorescence readings at different parasitaemias shown in the Table 1. Results are displayed for infected red-cell cultures (IRBC, black square, dashed line), infected red-cell cultures diluted with peripheral blood (IRBC+WBC, white circle, solid line), and infected red-cell cultures diluted with peripheral blood and depleted of white-cells (IRBC+WBC/W, white square, dotted line). All samples were processed for removal of haemoglobin and detergents. Curves fitted using inverse first order polynomial equation, r²> 0.97.