Real-time measurement of Plasmodium falciparum-infected erythrocyte cytoadhesion with a quartz crystal microbalance
- 1.7k Downloads
An important virulence mechanism of the malaria parasite Plasmodium falciparum is cytoadhesion, the binding of infected erythrocytes to endothelial cells in the second half of asexual blood stage development. Conventional methods to investigate adhesion of infected erythrocytes are mostly performed under static conditions, many are based on manual or semi-automated read-outs and are, therefore, difficult to standardize. Quartz crystal microbalances (QCM) are sensitive to nanogram-scale changes in mass and biomechanical properties and are increasingly used in biomedical research. Here, the ability of QCM is explored to measure binding of P. falciparum-infected erythrocytes to two receptors: CD36 and chondroitin sulfate A (CSA) under flow conditions.
Binding of late stage P. falciparum parasites is measured in comparison to uninfected erythrocytes to CD36- and CSA-coated quartzes by QCM observing frequency shifts. CD36-expressing cell membrane fragments and CSA polysaccharide were coated via poly-l-lysine to the quartz. The method was validated by microscopic counting of attached parasites and of erythrocytes to the coated quartzes.
Frequency shifts indicating binding of infected erythrocytes could be observed for both receptors CD36 and CSA. The frequency shifts seen for infected and uninfected erythrocytes were strongly correlated to the microscopically counted numbers of attached cells.
In this proof-of-concept experiment it is shown that QCM is a promising tool to measure binding kinetics and specificity of ligand-receptor interactions using viable, parasite-infected erythrocytes. The method can improve the understanding of the virulence of P. falciparum and might be used to cross-validate other methods.
KeywordsMalaria PfEMP1 Cytoadhesion Thickness shear mode sensors Biosensor CSA CD36
American type culture collection
bovine serum albumin
chondroitin sulfate A
indirect immunofluorescence test
infected red blood cell
magnetic cell isolation and cells separation
non-essential amino acid solution
Plasmodium falciparum erythrocyte membrane protein 1
quartz crystal microbalance
red blood cell
Quartz crystal microbalances (QCM) are highly sensitive sensors that can reliably weigh material in the nanogram range and are highly sensitive to changes in biomechanical properties of coupled biomaterial. They consist of a thin, usually round slice of crystalline quartz with a gold electrode on each side. Oscillating crystals are composed of α-Quartz (SiO2) that is known to be piezoelectric material, characterized by: (a) appearance of electrical potential when it is subjected to mechanical stress (piezoelectric effect); and, (b) deformation of the material when it is subjected to electrical potential (inverse piezoelectric effect). Due to the piezoelectric character the quartz is stimulated to oscillations when connected to AC voltage . Material attaching to the quartz’s surface reduces the frequency. Thus, an increase in weight leads to an adequate decrease in frequency. In biosciences, mass-sensitive QCM are already well accepted, while applications for clinical questions, in particular with whole blood samples, are novel and have not yet been exploited. Lately, this technology has been applied to the typing of blood groups, detection of bacteria, bacterial toxins, viruses, and measuring affinity of antibodies. Further studies on adherent cell types have been performed but the measurement of whole cell interactions of primarily non-adherent cells has not been addressed thoroughly [1, 2, 3, 4, 5, 6].
Here a novel QCM-based assay is presented to evaluate whole cell interactions by measuring cytoadhesion of erythrocytes infected with the malaria parasite Plasmodium falciparum to endothelial cell receptors. Malaria is the most important parasitic infection and accounted for more than 200 million episodes and 438,000 deaths in 2015 . The disease is caused as a result of the infection by protozoan parasites of the genus Plasmodium. Symptoms only occur during the blood stage of the infection when parasites multiply in red blood cells. Among five different species known to infect humans, P. falciparum causes the most severe form of the disease and is responsible for the vast majority of deaths. In contrast to the other species, P. falciparum-infected red blood cells (iRBCs) adhere to vascular endothelium of postcapillary venules (sequestration) and to non-infected erythrocytes (rosette formation) in the second half of their asexual replication, cycle via P. falciparum erythrocyte membrane protein 1 (PfEMP1), which is expressed on the surface of iRBCs [8, 9]. Through cytoadhesion the parasite avoids clearance by the spleen; this is an immune evasion mechanism that can lead to excessive parasite multiplication and may result in complications such as cerebral and placental malaria [10, 11, 12, 13]. Important receptors involved in this process are CD36, ICAM1 in cerebral malaria and chondroitin sulfate A (CSA) in placental malaria [14, 15, 16]. This work focuses on the binding of iRBC to CD36 and CSA. CD36 is expected to be the main receptor of iRBC adhesion as nearly all laboratory isolates of P. falciparum, as well as clinical isolates from patients attach to this receptor [15, 16, 17]. CSA on the other hand, is well described to be the main receptor of parasite binding in placental malaria and parasites expressing the highly conserved PfEMP1 gene var2csa are known to bind to this polysaccharide [14, 18, 19]. Placental malaria can cause low birth weight of newborns, premature delivery, abortion, stillbirth, maternal anaemia, as well as infant and maternal morbidity and mortality [20, 21]. CSA is also present on other endothelial tissue where parasites might bind to it .
Different methods have been developed to measure PfEMP1-mediated cytoadhesion. Up until now most of the established cytoadhesion assays have been performed under static conditions [23, 24, 25, 26], thus lacking shear stress which is assumed to be important for the parasites to get in contact with the appropriate receptors. However, flow-based assays exist as well and some include static periods at the beginning of the measurement to allow settling of parasites, so that they are able to make contact with the immobilized receptors [27, 28, 29, 30, 31, 32]. The disadvantage of some of these systems is that handling of the constructions is cumbersome; for example, the read-out is microscopy-based and therefore analysis of results is less standardized [33, 34, 35, 36]. Hence, a real-time detection of cytoadhesion is not possible and an adequate evaluation of the experiments performed with these microscope-based assays is time consuming and often reader biased. Another limiting aspect is that measuring of binding times and kinetics as well as discrimination between loosely adsorbed and truly adhered cells are very difficult.
The analysis of this complex interaction would benefit from other highly sensitive and reproducible methods based on a different technology for validation. Therefore, QCM is adapted in this proof-of-concept experiment for the analysis of PfEMP1-mediated iRBC binding to the cell surface of CD36-expressing melanoma cells as well as to CSA. It is the first time that QCM under flow conditions has been used with viable iRBCs and these experiments show the great potential of this approach for future studies in cell–cell as well as cell-receptor studies.
The parasite strains FCR3-CD36 and FCR3-CSA were kindly provided by Artur Scherf, Institute Pasteur Paris, and were cultivated as previously described  in fresh group O+ erythrocytes at 5 % haematocrit in RPMI-1640 supplemented with 2 mM l-glutamine, 50 µg/ml gentamycin, 25 mM HEPES, 0.5 % Albumax II, and 5 % AB+ human serum (complete medium). Culture flasks were aerated with 5 % carbon dioxide, 5 % oxygen and 90 % nitrogen. Specific binding phenotype of parasites was maintained by regular selection. For CD36 this was performed by monthly panning of late-stage, infected erythrocytes on C32-melanoma cells, at pH 6.8, as described elsewhere [38, 39]. To select parasites for specific binding to CSA, late blood stages were panned monthly on polystyrene six-well plates, coated with CSA from bovine trachea (Sigma) as previously described . No negative selection of parasites was done and parasites were used for experiments after at least one round of selection. Prior to adhesion assays, cultures were synchronized by treatment with 5 % sorbitol and late stages were purified/enriched by magnetic cell sorting (MACS) as described elsewhere .
C32-melanoma cell culture
C32-melanoma cells expressing large amounts of CD36 were obtained from the American Type Culture Collection (ATCC) and cultured at 5 % carbon dioxide and 37 °C using DMEM supplemented with 50 µg/ml gentamycin, 10 % FCS, 2 mM l-glutamine, and 1 % MEM non-essential amino acid solution (NEAA). Splitting of cells was performed when cultures reached 80–90 % confluence.
The QCM sensors used in this work were AT-cut, 10 MHz quartz crystals (8 mm in diameter, thickness of 166 μm (KVG Quartz Crystal Technology, Neckarbischofsheim, Germany) with a special gold electrode design developed by FKG.
Base coating of quartz surface
To successfully immobilize receptors to the quartz, a base coating with poly-l-lysine (PLL, Fluka 70–150 kDa) to untreated sensors was done by dropping 60 µl of 0.5 mg/ml PLL in H2O to the quartz and subsequent drying under a stream of nitrogen . For each experiment a new quartz was used as a complete cleaning of used quartzes was not possible.
Coating of CD36 to the quartz surface
CD36 belongs to the class B scavenger receptor proteins and is characterized by a special structure that is necessary to ideally present the specific binding site to parasitized erythrocytes. Thus, to preserve this structure for CD36 coating, membranes or membrane fragments of C32 cells were prepared by homogenization of C32 cells with a potter on ice using a fractioning buffer consisting of Tris/HCl (50 mM), NaCl (150 mM) and EDTA (1 mM). To separate the membrane fragments from nuclei and intact cells, the suspension was centrifuged at 540×g (10 min, 4 °C). In a next step the supernatant was centrifuged at 2500×g (15 min, 4 °C) to remove further organelles. The last centrifugation step using an ultracentrifuge at 100,000×g (1 h, 4 °C) served to concentrate the membrane fragments. After removal of the supernatant the pellet was re-suspended in 400 µl fractioning buffer supplemented with 800 µl of a proteinase-inhibitor solution (Roche). Aliquots were stored at −20 °C. To achieve a coating as homogeneous as possible, membrane suspensions were incubated with PBS-Tween-20 (0.1 %) for 5 min. Immobilization on PLL-coated quartzes was done by dropping 60 µl of suspension to the quartz, and subsequent incubation for 10 min before unbound membranes were washed away with PBS. This was followed by fixation with 4 % paraformaldehyde in PBS on ice for 7 min and blocking with 2 % BSA in PBS (1 h, RT).
Coating of CSA to the quartz surface
To coat CSA to the quartz surface CSA (Sigma, 1 mg/ml in PBS) was immobilized ‘online’ to PLL. Following the insertion of the PLL-coated quartz to the QCM platform, the CSA solution was allowed to run over the quartz at a flow rate of 50 µl/min for 10 min to enable CSA to bind to the PLL-coated quartz. This was followed by a washing step with flow medium for 5 min at 100 µl/min.
To verify binding specificity for PLL basis coating, control human serum was added to the quartz outside of the platform before adding iRBCs (FCR3-CSA) without prior coating of CSA.
Verification of coating by immunofluorescence
Distribution of the receptors CD36 and CSA on the quartz were verified by an indirect immunofluorescence test (IFT). A separate sample was treated with secondary antibody only, to control for unspecific binding of the antibody. After coating, quartzes were fixed with 4 % paraformaldehyde in PBS on ice for 7 min, followed by blocking with 2 % BSA in PBS (1 h, RT).
The distribution of CD36 on the membrane of C32 cells was visualized by staining with a monoclonal mouse-anti CD36 antibody (Abcam, 1:10 in PBS, 1 h, RT); and a sheep anti-mouse IgG F(ab′)2 fragment–FITC antibody (Sigma, 1:30 in PBS, 1 h, RT). For verification of CSA-coating, quartzes were stained with a monoclonal mouse anti-CSA antibody (Sigma, 1:100 in PBS, 1 h, RT) and a FITC-labelled goat anti-mouse IgM antibody (Sigma, 1:128, 1 h, RT).
Pictures of stained quartzes were taken by the digital camera Casio QV 5700 at the Axioskop 2MAT microscope with 100× or 500× magnification. Overlay of pictures was made with Adobe Photoshop 7.0.
Sensor platform and adhesion assay using QCM-sensors
To ensure a sterile environment for measurements, the fluidic system was first disinfected by incubation with 1 % sodium hypochlorite for 10 min. After washing with sterile distilled water and a de-gassed flow medium (complete medium without sodium hydrogen carbonate and without human serum), sensors were inserted into the pre-heated platform (37 °C). Cell suspensions were diluted in flow medium and were transferred to the sensor surface at a continuous flow rate (50 µl/min).
The optimal parameters for detection of adhesion processes to the receptors by iRBCs were achieved by systematically adapting measurement periods, cell concentrations and flow rates. Final experiments with CD36 and CSA coated to PLL were performed with the optimized closed fluidics system. The flow rate was adjusted to 50 µl/min to fill the fluidic system at the onset of the measurement with flow medium, and to pump the cell suspension over the sensor during the period of the whole measurement of cell adhesion. For the assays, 1.5 × 105 iRBCs/ml or RBCs/ml were diluted in flow medium and allowed to circulate and flow over the quartz at 50 µl/min for 2.5 h in a closed system. The inner diameter of the tubes used was 0.38 mm, resulting in a velocity of 73 mm/s for the used flow rate.
Verification of parasite adhesion by microscopy after QCM
After binding experiments, quartzes were removed, rinsed with flow medium and fixed with 4 % paraformaldehyde in PBS on ice for 7 min. To determine the stage and viability of iRBCs adherent to the sensor’s surface, samples were stained with the DNA-stain DAPI (Sigma, 0.2 µg/ml, 5 min, RT). The mean number of bound iRBCs and RBCs of three visual fields, corresponding to a surface of 22.1 × 103 µm2, were determined by counting under a microscope (AxioImager, Zeiss, magnification 500×).
Frequency shifts were analysed using qCell T software. Median differences, the interquartile range as well as the single measurements are shown for frequency shifts and the subsequent microscopic count data of attached parasites to the quartz. Microscopic counts of attached infected and uninfected RBCs are correlated to frequency shifts by Spearmans rho for paired samples.
Verification of CD36 coating of the quartz
Images obtained by fluorescence microscopy show that an almost homogenous coating of the quartzes with CD36 could be achieved and that the CD36 antibody epitope was not destroyed by the processes of cell homogenization and the following differential centrifugation (Fig. 2b). Minor differences in the dissemination of membranes to the sensors are caused by the nature of the biological medium in which CD36 is embedded. To preserve the complex structure of the CD36 protein, which belongs to class B scavenger receptor proteins, whole membranes of C32 melanoma cells are used. Whole cells could not be used as the layer on the sensor would be too thick and not rigid enough to determine interactions of the parasites with the coated sensors.
Verification of CSA coating via fluorescence microscopy
Coating of the quartz sensor with the receptor was confirmed by fluorescence microscopy (Fig. 2c). CSA immobilized via PLL resulted in a homogenous and dense coating, illustrated by a strong, specific fluorescence signal. Coating of the sensor with CSA via PLL was additionally verified by on-line measurement in the platform. Injection of CSA led to a strong binding signal, which remained stable even after rinsing with high flow rates. Further incubation steps with different concentrations of bovine serum albumin (BSA) showed no additional binding events (Additional file 1). Thus, potential unspecific binding sites appear to be non-accessible.
Measurement of binding of Plasmodium falciparum iRBCs to receptors (CD36 and CSA) via QCM
The number of attached cells was correlated with the observed frequency shift (rho = −0.55, p = 0.017). Example images of one microscopic field are given in Fig. 4b, c for CD36 and iRBCs and RBCs, respectively, and in Fig. 5b, c for CSA and iRBCs and RBCs, respectively. They show that iRBCs with FCR3-CD36 parasites attached to CD36-coated quartzes and that iRBCS with FCR3-CSA parasites attached to CSA-coated quartzes (Figs. 4b, 5b), in contrast to uninfected RBC (Figs. 4c, 5c), where only a very few cells attached.
PLL by itself is very sticky for any kind of cell and can therefore not be used without a receptor/analyte. PLL-coated quartzes alone bound added cells very strongly and quickly (within minutes) (see Additional file 2 for RBCs). As a separate control experiment, human serum proteins were immobilized to PLL and as expected, frequency shifts showed that iRBCs (FCR3-CSA) did not bind to human serum proteins [median frequency shift: 46 Hz (n = 3), Additional file 3].
Measuring with QCM allows real-time observation of cellular interactions by evaluating adhesion processes of intact P. falciparum-infected RBCs and their respective receptor. Here a novel method is presented to measure cell adhesion processes of iRBCs to two different receptors under flow conditions. Binding is analysed of iRBCs to CD36, a receptor most P. falciparum isolates adhere to, and binding to CSA, the main receptor involved in placental malaria. Dampening of the signal showed that iRBCs bound with higher affinity to the receptor when compared to uninfected RBCs, corresponding with the results obtained by microscopic counting of attached cells. Control experiments with human serum as unspecific analyte showed that iRBCs did not bind to serum proteins coated to PLL. The observed increase in frequency similar to the experiments of CD36 and RBCs is probably caused by a change of the visco-elasticity of the sensors surface called the missing mass effect . Only by adding cells and medium to the sensor, could a conformation change of PLL be achieved so that a more rigid layer is present, influencing the sensor’s behaviour. However, a control experiment with PLL and human serum shows that frequency shifts obtained with the iRBCs using the PLL/CD36- or PLL/CSA-coated sensors are due to specific interactions between the iRBCs and the receptors, and not to unspecific interactions with PLL. Parasite binding to CSA is a relevant model for placental malaria, and evaluation of inhibiting antibodies and vaccine candidates under flow conditions by QCM could complement existing methods. Further experiments with the QCM should be performed to explore the potential of this methodology in studying malaria parasites in more depth. Future experiments should include different parasite strains, including freshly isolated clinical isolates from the relevant organ (e.g., placenta), in addition experiments with inhibiting antibodies and possible blocking proteins should be performed to evaluate the potential of the methodology. One weakness of the system is that coating of a sugar as done with CSA might not represent the three-dimensional structure found in nature, a problem circumvented by coating of membranes displaying CD36 on their surface. Despite this however, this methodology has the weakness that one cannot fully control correct positioning of the membranes, and some might even be attached upside down. So far QCM technology has been applied only to analyse molecules in malaria research but not for studying the whole parasite. One group established the differentiation of Plasmodium species by analysing DNA fragments . Another group evaluated the kinetics of protein interactions on the example of binding kinetics of the PfEMP1 protein var2CSA to different antibodies . Here for the first time whole cell interactions of live parasites are investigated. This opens possibilities for malaria research not restricted to cytoadhesion studies. In addition, parasite invasion and egress or intracellular growth could be investigated. QCM technology is a promising tool for research in basic science on parasite biology and parasite interactions, and can also be applied for research on malaria vaccine and drug development.
With the presented experiments QCM technology should be made accessible for malaria research. This technology enables the real-time detection of cytoadhesion of iRBCs under flow conditions without affecting viability of parasites. Head-to-head comparison to the standard assays in relevant experimental settings, for example in clinical settings within vaccine trials, are an obvious and important next step that should be taken.
The study was conceived and experiments designed by DK, JH, HRD, SCT, HN, BM and FKG. DK and SFM performed the experiments and DK, JH, SFM, HRD, SCT, BM and FKG analysed the data. DK, JH, SFM, HRD, SCT, HN, BM and FKG contributed reagents/materials/analysis tools and wrote the paper. All authors read and approved the final manuscript.
Serum for parasite culture was kindly provided by Torsten J. Schulze, Blood Donation Centre, Institute of Transfusion Medicine and Immunology, Mannheim, Germany. We acknowledge support by Deutsche Forschungsgemeinschaft and Open Access Publishing Fund of University of Tübingen.
HRD and FKG work at 3t-analytic, Germany.
- 7.WHO. World malaria report. Geneva: World Heath Organization; 2015.Google Scholar
- 8.Baruch DI, Gormely JA, Ma C, Howard RJ, Pasloske BL. Plasmodium falciparum erythrocyte membrane protein 1 is a parasitized erythrocyte receptor for adherence to CD36, thrombospondin, and intercellular adhesion molecule 1. Proc Natl Acad Sci USA. 1996;93:3497–502.CrossRefPubMedPubMedCentralGoogle Scholar
- 16.Barnwell JW, Asch AS, Nachman RL, Yamaya M, Aikawa M, Ingravallo P. A human 88-kD membrane glycoprotein (CD36) functions in vitro as a receptor for a cytoadherence ligand on Plasmodium falciparum-infected erythrocytes. J Clin Invest. 1989;84:765–72. doi: 10.1172/JCI114234.CrossRefPubMedPubMedCentralGoogle Scholar
- 39.Ljungström I, Perlmann H, Schlichtherle M, Scherf A, Wahlgren M. Methods in Malaria Research. 4th ed. Manassas: MR4/ATCC; 2004.Google Scholar
- 40.Beeson JG, Brown GV, Molyneux ME, Mhango C, Dzinjalamala F, Rogerson SJ. Plasmodium falciparum isolates from infected pregnant women and children are associated with distinct adhesive and antigenic properties. J Infect Dis. 1999;180:464–72. doi: 10.1086/314899.CrossRefPubMedPubMedCentralGoogle Scholar
- 44.Gehring FK. http://www.3t-analytik.de/. Accessed 31 July 2015.
- 46.Wangmaung N, Chomean S, Promptmas C, Mas-Oodi S, Tanyong D, Ittarat W. Silver quartz crystal microbalance for differential diagnosis of Plasmodium falciparum and Plasmodium vivax in single and mixed infection. Biosens Bioelectron. 2014;62:295–301. doi: 10.1016/j.bios.2014.06.052.CrossRefPubMedGoogle Scholar
- 47.Joergensen LM, Salanti A, Dobrilovic T, Barfod L, Hassenkam T, Theander TG, et al. The kinetics of antibody binding to Plasmodium falciparum VAR2CSA PfEMP1 antigen and modelling of PfEMP1 antigen packing on the membrane knobs. Malar J. 2010;9:100. doi: 10.1186/1475-2875-9-100.CrossRefPubMedPubMedCentralGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.