A flow cytometry-based assay for measuring invasion of red blood cells by Plasmodium falciparum

Abstract

Variability in the ability of the malaria parasite Plasmodium falciparum to invade human erythrocytes is postulated to be an important determinant of disease severity. Both the parasite multiplication rate and erythrocyte selectivity are important parameters that underlie such variable invasion. We have established a flow cytometry-based method for simultaneously calculating both the parasitemia and the number of multiply-infected erythrocytes. Staining with the DNA-specific dye SYBR Green I allows quantitation of parasite invasion at the ring stage of parasite development. We discuss in vitro and in vivo applications and limitations of this method in relation to the study of parasite invasion.

INTRODUCTION

Clinical manifestations of malaria are largely associated with the asexual, erythrocytic stage of the Plasmodium life cycle. This stage is characterized by the exponential growth of the parasite in the bloodstream, with cycles of erythrocyte invasion, development, and egress. Efficient erythrocyte invasion and lack of erythrocyte selectivity have been associated with the severity of clinical disease (1).

Underlying efficient invasion are the parameters of parasite multiplication rate (PMR) and the selectivity index (SI) (2). PMR is the fold increase in parasitemia following each round of asexual growth, while the SI is the ratio of observed multiply-infected erythrocytes (i.e. erythrocytes that harbor more than one parasite) to those expected by random events, as predicted by a Poisson distribution. Parasite invasion efficiency and selectivity have been previously measured by light microscopy (3–5), however this method is time consuming and fraught by bias.

Flow cytometry-based methods have been previously described for P. falciparum (6–8), but have not been focused on invasion-related studies. We have established a flow cytometry-based protocol involving SYBR Green I that can simultaneously determine parasitemia as well as the number of multiply-infected erythrocytes, making the assessment of nuanced invasion phenotypes possible.

METHODS

Parasite culture

P. falciparum asexual stages were maintained in vitro in human O⁺ erythrocytes at 4% hematocrit in RPMI-1640 media supplemented with 25 mM HEPES, 0.21% sodium bicarbonate, 50 mg/L hypoxanthine, and 0.5% Albumax II. Invasion assays were performed by mixing enzyme-treated infected donor cells with equivalent cell number of RPMI treated erythrocyte control cells, or enzyme-treated erythrocyte negative control cells. Ring-stage parasitized donor cells from the peripheral blood of Senegalese patients were treated with α-2-3,6,8–Vibrio cholera neuraminidase (66.7 mU/ml, Calbiochem), trypsin (1 mg/ml, Sigma), and chymotrypsin (1 mg/ml, Worthington) to prevent reinvasion. Parasites were plated in duplicate at a final parasitemia of between 0.45%-1% at 2% hematocrit in complete RPMI media. Following reinvasion, 48 hours post-plating, parasitemia was assessed by microscopy and by flow cytometry.

Flow cytometry

200 µl of sorbitol-synchronized, ring-stage parasites were cultured in a 96-well microtiter plate at 1% parasitemia. Samples were plated in triplicate and incubated at 37°C until reinvasion. Cultures were pelleted via centrifugation (1200 rpm, 5 minutes) and washed twice in 100 µl 1× phosphate buffered saline (PBS) + 0.5% bovine serum albumin (BSA) + 0.02% sodium azide. Cells were then incubated with 75 µl of 1:1000 SYBR Green I (Molecular Probes) for 20 minutes at 25°C.

Cells were washed in 1× PBS + 0.5% BSA + 0.02% sodium azide and resuspended in PBS. Flow cytometry data was collected using a FACSCalibur (Becton Dickinson) with an acquisition of 100,000 events per sample. Initial gating was carried out with unstained, uninfected erythrocytes to account for erythrocyte autofluorescence.

Light microscopy

Giemsa-stained thin films were prepared on glass slides. Parasitemia, defined as the percentage of parasite infected-erythrocytes, was determined by counting a minimum of 500 erythrocytes using a Miller graticule (total of 4500 erythrocytes). Multiply-infected cells were counted using an electronic differential counter.

RESULTS

Concordance between microscopy- and flow cytometry-determined parasitemia

The goal of this study was to compare the accuracy of the SYBR Green I flow cytometry method with light microscopy in the determination of parasitemia as well as selectivity. P. falciparum HB3 parasites were serially diluted and resulting parasitemia was measured by the flow cytometry-based method. A high linear correlation (R²=0.9925) was observed in the measurement of parasitemia between both the microscopy- and flow cytometry-based methods with a limit of detection of 0.2% (Figure 1A).

Figure 1. Accuracy of flow cytometry determined assessment of first round invasion parasitemia.

(A) Flow cytometry provides accurate determinations of P. falciparum parasitemia. There is a high correlation between microscopy- and flow cytometry-determined parasitemia. This representative experiment depicts an R-squared value of 0.9925 between the two methods. Individual data points are the average of triplicate samples and error bars represent the standard deviation. (B) Flow cytometry resolves singly and multiply infected erythrocytes. Following staining with SYBR Green I, 100,000 events were counted and three peaks emerge, corresponding to singly, doubly, and triply infected erythrocytes. (C) Parasitemia within singly, doubly, and triply infected erythrocytes was measured by flow cytometry and microscopy. The resulting parasitemia calculated by both methods was highly correlated and highly reproducible between experiments.

Flow cytometry resolves multiply-infected erythrocytes

The high concordance between microscopy- and flow cytometry-determined parasitemia is attributable to the ability of the flow cytometer to resolve multiply-infected erythrocytes. SYBR Green I staining of P. falciparum cultures allows clear resolution of singley-, doubly-, and triply-infected erythrocytes (Figure 1B). As demonstrated by both light microscopy and flow cytometry, multiply-infected erythrocytes are less common than singly-infected red blood cells. Erythrocytes harboring more than three parasites are not well-resolved by flow cytometry. However, microscopic determination of multiply-infected erythrocytes has revealed that in laboratory isolates it is a relatively rare event occurring less than 0.01% of the time. We have counted the number of singly and multiply infected cells in three independent trials. In all cases, flow cytometry accurately measures the parasitemia of infected erythrocytes harboring both singly and multiply infected erythrocytes (Figure 1C).

Utility of flow cytometry-based method for measuring parasitemia in vivo and ex vivo in a field-based setting

To validate our flow cytometry based method, we used it to assess the parasitemia of samples directly from Senegalese patients in vivo and cultured for a single round of reinvasion ex vivo (Figure 2). We found that measuring overall parasitemia by flow was reproducibly accurate and in concordance with parasitemia assessed by microscopy for ex vitro invasion assays (Figure 2A). Additionally, by this method, we were able to detect the differences in in vivo parasitemia compared to in vitro parasitemeia in the number of multiple invasion events (Figure 2B). While most in vivo parasitemia was limited to single invasion events, ex vivo parasitemia revealed many multiple invasion events, and further, more multiple invasion events observed in ex vivo isolates than for in vitro culture adapted lines (Figure 2B, Figure 1B). Further, we show good concordance between the measures of multiple invasion events by microscopy and flow for ex vivo isolates, similar to what was seen for laboratory isolates (Figure 1C). Taken together, these data indicate that our flow cytometry method is robust and accurate in the measure of invasion parasitemia for both culture adapted isolates and wild isolates and that this methodology is applicable to field based research settings.

Figure 2. Accuracy of flow cytometry determined assessment of in vitro parasitemia and first round invasion parasitemia ex vivo from Senegalese patient isolates.

(A) Flow cytometry provides accurate determinations of ex vivo invasion assay P. falciparum parasitemia. Three representative Senegalese samples are shown to demonstrate the similarity between microscopy- and flow cytometry-determined parasitemia. Individual data points represent the average of duplicate samples and error bars represent the range. (B) Following staining with SYBR Green I, 100,000 events were counted for both in vivo (initial) and ex vivo (re-invasion) parasitemia. For in vitro parasitemia, four peaks are observed, corresponding to singly, doubly, and triply, and quadruply infected erythrocytes. (C) Parasitemia within singly, doubly, triply, and quadruply infected erythrocytes was measured by flow cytometry and microscopy. The resulting parasitemia calculated by both methods was highly correlated for a given patient isolate.

DISCUSSION

We have established a flow cytometry-based method to measure determinants of P. falciparum virulence –namely, PMR and SI. This SYBR Green I based method allows the accurate measurement of parasitemia that distinguishes between singly- and multiply-infected erythrocytes, in a manner which is highly comparable with light microscopy, yet avoids both the time-intensive nature and subjectivity.

SYBR Green I is added directly to parasite-infected erythrocytes, with very little handing and avoids the fixation steps required for propidium iodide. SYBR Green I emits on the fluoroscein isothiocyanate (FITC) channel. SYBR Green I displays an 11-fold greater preference for double-stranded DNA than single-stranded DNA (9) and has a low binding affinity for RNA. While reticulocytes harbor RNA, mature erythrocytes are devoid of DNA and RNA (10). As SYBR Green I adheres only to double-stranded DNA, any fluorescence that is detected is attributable to parasite DNA.

The precision with which flow cytometry resolves multiply-infected erythrocyte peaks requires the culture to be at the ring-stage. As the culture transitions from late rings to early trophozoites the parasite begins to replicate its DNA. The flow cytometer is unable to distinguish between the fluorescence emitted by an erythrocyte harboring three, ring stage parasites versus one infected by a single, early trophozoite. The requirement for ring-stage parasites can easily be met for laboratory-adapted parasites cultured in vitro by synchronization by sorbitol lysis (11). However, ex vivo parasites are almost always ring-stage parasites as later stages of parasite development, as late-stage trophozoites and schizonts, sequester in vivo and blood collected from malaria patients consists almost exclusively of ring-stage parasites.

One challenge of adapting this flow-based method to ex vivo assays with uncultured isolates is that some isolates do not grow as synchronously as laboratory-adapted cultures in the first round of invasion. A small proportion of parasites will not complete the cycle to re-invasion as rings and will instead arrest at troph/schizont. These parasites can easily be removed from the analysis by separating the donor parasitized erythrocytes from acceptor erythrocytes by labeling the acceptor erythrocytes with a dye such as FITC, allowing for the distinction of donor from acceptor erythrocytes (12).

Measurement of the total number of successfully invaded merozoites rather than just the total number of parasitized erythrocytes provides more comprehensive data on the nature of parasite invasion. As well as the parasite multiplication rate (PMR), knowledge of the number of multiply infected erythrocytes allows for the assessment of the selectivity index (SI), an index that measures parasite preference for erythrocytes.

Other ex vivo applications of this method include 1) the measurement of the inhibitory effects of antibodies used in parasite neutralization assays and 2) the ability of parasites to invade erythrocytes of different ages by co-staining for age-specific markers such as CD71 and phosphatidyl serine (PS) (13, 14). By measuring newly invaded erthrocytes at the ring-stage, our flow cytometry-based can separate parasite invasion from parasite growth. This is not possible with other high throughput assays that depend on metabolic labeling, such as the standard tritiated hypoxanthine assay.

The rate with which flow cytometry can determine parasitemia and multiply-infected erythrocytes is advantageous, as light microscopy is time-consuming; especially for cultures that exhibit low parasitemia which is often the case with field isolates subjected to first-round in vitro invasion assays. The ease, specificity, and convenience of our flow cytometry-based method makes it suitable not only to the laboratory, but to the field. Further development making it suitable to the 96-well format will further increase the usefulness of the method.