Label-free electrochemical detection of malaria-infected red blood cells

Abstract

The precise and rapid diagnosis of malaria is key to prevent indiscriminate use of antimalarial drugs and help in timely treatment and management of the disease. This paper reports a label-free detection of P. falciparum infected red blood cells using a gold nanoparticle (GNP) enhanced platform. The GNPs were electrodeposited on screen-printed electrodes to form a well-controlled matrix that served the dual role of antibody immobilization and signal enhancement. The detection of infected red blood cells was carried out by measuring changes in electrical parameters as a result of its binding to cell reactive antibodies immobilized on the electrode. The assay showed good sensitivity and a linear response between the electron transfer resistance and the logarithm of the number of infected red blood cells which was observed over a concentration range of 10² cells per mL to 10⁸ cells per mL. This is the first report where an antibody-functionalized electrochemical biosensing platform has been employed for the quantitative detection of P. falciparum infected whole red blood cells.

---

Introduction

Malaria is an infectious disease caused by the parasite of the genus Plasmodium. It is a major killer and affects an estimated 3.3 billion people across the globe. An estimated 584 000 deaths occurred due to the disease in 2013.¹ The symptoms of malaria include fever, chills, diaphoresis and headache.² The majority of the patients suffer from fever, a common symptom in other infectious diseases too. Health workers often misdiagnose and wrongly prescribe an antimalarial in the case of malaria negative febrile patients.³ Therefore, the WHO Guideline discourages presumptive treatment and recommends confirmed diagnosis of suspected malaria case by rapid diagnostics tests (RDT) or Giemsa staining before administration of antimalarial drugs. An accurate diagnosis will prevent the misuse of antimalarial drugs, and help in proper treatment.^4,5 Observation of Giemsa stained specimen under the light microscope is simple, cheap, accurate, and considered as a gold standard for parasite detection with a sensitive detection of parasitemia as low as 0.001%.⁶ However, this method is time taking, labour intensive and requires an expert particularly for the detection of a sample with low parasitemia.⁷ Currently, various RDTs are available in cassette, card and dipstick format for the point of care diagnosis^8,9 which are user-friendly and cost effective. These RDTs primarily utilize targets like histidine rich protein-2 (HRP-2), lactate dehydrogenase (LDH) and aldolase.^9,10 HRP-2 is the most widely used antigen but suffers from several drawbacks like sequence variation, deletion, cross-reactivity with rheumatoid factor and long persistence time in the blood resulting false detection.^10–14 Also RDTs, particularly those based on LDH, fail at low parasitemia and non specific reactivity is observed because of no washing step involved during the test.^9,10 Other target candidates like haem-detoxification protein, heat shock protein 70, dihydrofolate reductase, thymidylate synthase, were speculated as potential biomarkers, but none of those have found commercial utility.^14,15 For obtaining parasite density polymerase chain reaction (PCR) based techniques can attain sensitivity of 0.0001% (fewer than 5 parasites per μL), however, the process is costly and time taking.^6,13,16

Electrochemical biosensors based on antibody have the potential to provide a portable, sensitive and reproducible system for detection of clinical analytes.¹⁷ Sharma et al. reported the antibody-based detection of malaria HRP-2 antigen using alkaline phosphatase tagged secondary antibody and 1-naphthyl phosphate as its electrochemical substrate.¹⁸ Lee et al. used aptamer against LDH on gold nanoparticle (GNP) modified screen printed electrode (SPE) and measured the response due to binding of the antigen by electrochemical impedance spectroscopy.¹⁹ However, these and other methods have the inherent problems related to the selected target. Therefore, other strategies with alternate targets are being explored. For instance, several groups used hemozoin for the detection of malaria infection.^6,20,21

The proposed work is based on the fact that during malaria infection, the numbers of parasites/infected cells are large even at a low parasitemia. For example, during normal symptomatic infection with a parasitemia of 0.2%, a single microliter of blood has about 10 000 cells. In severe P. falciparum (Pf) infection, the parasitemia may reach up to 5 to 10 percent (250 000–500 000 infected cell per mL).²² However, the currently available RDTs are not able to provide a quantitative estimation of the parasite load.⁹ Further, the infected red blood cells (IRBC) are different from the normal red blood cells (NRBC) as the parasite carries out extensive modifications in the host red blood cells. Thus, the present study envisages a new label-free detection of P. falciparum infected whole red blood cells using a monoclonal antibody (MAb) reactive with malaria IRBC using gold nanoparticle modified disposable SPE (screen printed electrode). The SPE was modified by electrodeposition of citrate capped GNP followed by adsorption of infected cell reactive antibody on the sensor surface. The resulting change in electrical parameters due to binding of the IRBC and NRBC to the modified electrode surface was recorded. The system was able to distinguish the infected versus the non-infected cells as changes in EIS signal were obtained only with IRBCs binding to its specific antibody on the surface (Fig. 1). Thus, the main aim of the work was to establish an electrical impedance spectroscopy (EIS) based method for the quantitative detection of malaria-infected cells.

Fig. 1 Schematic showing the detection of P. falciparum infected red blood cells by electrochemical impedance spectroscopy.

Materials

Screen printed electrodes (TE-100) were obtained from CH Instruments, USA. The reagents used in the study, hypoxanthine, sodium bicarbonate (NaHCO₃), phosphate buffered saline (PBS) tablets, 2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid (HEPES), potassium ferrocyanide, potassium ferricyanide, glutaraldehyde, sodium citrate, tetrachloroauric acid, 1-ethyl-3-(3-dimethylaminopropy) carbodiimide (EDC) and N-hydroxysuccinimide (NHS) were purchased from Sigma. Other reagents used were: RPMI 1640 (Invitrogen), Albumax II (Gibco), percoll (GE Healthcare), sodium hydroxide, methanol (Merck), H₂O₂ (Calbiochem), ECL kit (Pierce), TMB/H₂O₂ (GeNei). All the solutions and buffers were prepared in ultrapure MilliQ water.

Plasmodium falciparum culture

P. falciparum strain 3D7 was maintained in complete Pf-culture medium (RPMI-1640 10.4 g L⁻¹, hypoxanthine 100 μM, HEPES 25 mM, Albumax II 0.5% (w/v), sodium bicarbonate 1.77 mM and gentamycin 12.5 μg mL⁻¹ using published protocol²³ with slight modifications. Briefly, the culture was done in a 150 cm² vented neck flask in CO₂ incubator at 37 °C. The media was changed after every 24 hours and fresh blood was regularly added to maintain the hematocrit at 1.5%. The culture was monitored by Giemsa staining for the assessment of parasitemia.

Purification of infected cells

P. falciparum infected red blood cells (IRBCs) were purified following the method discussed by Miao and Cui using percoll density centrifugation.²⁴ Pf culture at the schizont stage was centrifuged at 250g for 5 minutes and the pellet was diluted in PBS to obtain a 10% (v/v) suspension. Two layers of percoll (35% on top of 65%) were prepared in a 15 mL centrifugation tube. The resulting cell suspension was layered on top of the 35% percoll layer and centrifugation was done at 1500g for 15 minutes. The dead cells and other debris formed on the 35% percoll were discarded. The cells at the interface of 35% and 65% percoll layer were slowly pulled into a separate tube and washed thrice with PBS.

Whole cell ELISA

The monoclonal antibody MAbD2 used in the study was raised against P. berghei infected cell membrane.²⁵ The reactivity of MAbD2 with the P. falciparum infected red blood cells was checked by whole cell ELISA using the protocol as described earlier²⁶ with slight modifications. Briefly, 2 × 10⁵ cells were coated to each well of 96 well ELISA plate for overnight at 4 °C (MaxiSorp, NUNC). The cells were then fixed with 0.2% glutaraldehyde in PBS for 20 minutes then washed twice with PBS. Blocking was then carried with 10% skimmed milk for 3 hours at room temperature (RT). After 2 washes with PBS, 50 μL primary antibody was added and incubated at RT with mild shaking. Endogenous peroxidase activity was blocked with H₂O₂ in methanol (166 μL/50 mL) for 15 minutes at RT. After four washes with PBS, HRP labeled secondary antibody (1 : 4000) was added and incubation was carried out for 1 hour at RT. After washing, 100 μL of TMB/H₂O₂ substrate solution was added and the plate was incubated for 30 min at 37 °C for colour development. The reaction was finally stopped with 1 N H₂SO₄ and absorbance at 450 nm was measured.

Electrode preparation

Citrate capped GNPs were synthesized following Frens method²⁷ with slight variation.²⁸ Briefly, 100 mL of 0.01% solution of tetrachloroauric was heated in a 250 mL flask on a hot plate stirrer. Once the boiling started, a solution of 4 mL of 1% sodium citrate was added rapidly and the boiling was continued until the colour changed to bright wine red. The heating was carried out for another 10 minutes after which the solution was cooled to RT and shifted to 4 °C for further storage. A solution of 20% (v/v) GNP was prepared in phosphate buffer. For electrodeposition, 100 μL of the GNP solution was spread over the screen-printed electrode in the form of a droplet and voltage of +0.7 V was applied for 10 minutes. To remove excess of GNP, the surface of the electrode was rinsed with phosphate buffer and then dried in a hot air oven for 30 minutes.

Antibody immobilization

Monoclonal antibodies (MAbD2 raised against the malaria infected cells and antibody control) was diluted to a concentration of 25 μg mL⁻¹ in PBS buffer (pH = 7.4) and spread over the surface of the electrode and incubated at RT for 1 hour followed by overnight incubation at 4 °C. After antibody immobilization, the surface of the screen printed electrode (SPE) was rinsed thoroughly with PBS, to remove the loosely bound antibodies. The binding of the antibodies was monitored by recording the cyclic voltammetry and impedance scans.

Electrical impedance spectroscopy

The electrical measurements were made on carbon screen-printed electrodes with 3 mm diameter disk, graphitic carbon powder working and auxiliary electrodes, Ag/AgCl pellet reference (TE-100, Zensor, Taiwan) using electrochemical workstation from CH Instruments, USA. All the impedance measurements were performed in a droplet configuration of the biosensor by application of 100 μL of 5 mM ferricyanide/ferrocyanide (1 : 1) mixture in PBS. The droplet was spread onto the working area of the sensor so as to cover all the 3 electrodes of the SPE. The faradic impedance spectroscopy was recorded at the formal potential of the redox couple (ferricyanide/ferrocyanide) i.e. +0.172 V (versus screen-printed Ag/AgCl) at an amplitude of 5 mV, with the frequency range from 1 Hz up to 100 kHz. For monitoring changes in the spectrum as a result of IRBC binding, a suspension of varying concentrations of IRBCs (10² cells per mL to 10⁸ cells per mL), made in 5 mM ferricyanide/ferrocyanide (1 : 1) mixture in PBS by serial dilution, was spread on the surface of the electrode and incubated for 10 minutes. The surface was then briefly washed with 100 μL of ferricyanide/ferrocyanide solution twice to remove the unbound cells. Impedance records were obtained at RT under ambient conditions. The data analysis was performed using the standard system software available with an analyzer (CH-600) from CH Instruments, USA.

Results and discussion

Selection of the monoclonal antibody

The Plasmodium parasite enters the red blood cells and induces several modifications in it to make the environment suitable for its survival.²⁹ It modifies the cell surface of the IRBC by exporting several proteins on the surface, thus differentiating it from NRBC.^30,31 To distinguish the Pf-infected red blood cells from the normal red blood cells, we used a monoclonal antibody MAbD2 that specifically recognized the infected cells. MAbD2, used in the study, was raised in the previous study in our laboratory against the cell membrane of the rodent parasite, P. berghei infected erythrocytes. This antibody reacted with the rodent IRBCs with greater affinity. However, mild cross-reactivity was also observed with the mouse NRBCs.²⁵ To use the antibody for the development of a biosensor to distinguish the infected versus non-infected cells, the reactivity of MAbD2 was first checked with P. falciparum infected cells harbouring schizont stage of parasite (IRBC) in whole cell ELISA. The IRBCs coated on ELISA plate were probed with different concentrations of test or control antibody. MAbD2 reacted with the infected cells in a dose-dependent manner (Fig. 2a) while no reactivity could be seen with NRBC (Fig. 2b). MAbD2 was also checked for its binding with the ring, trophozoite and schizont stages of infected red blood cells by flow cytometry analysis. Results showed that the antibody reacted with the trophozoite and schizont-stage infected cells while no binding could be seen with the ring stage-infected cells (Fig. S5†).

Fig. 2 Whole cell ELISA showing the binding of MAbD2 to (a) P. falciparum infected human RBC and (b) normal RBC. Wells were coated with 2 × 10⁵ cells as described in method.

Electrochemical surface characterization

Screen printed electrodes (SPEs) are convenient replacements of bulky electrochemical cells in modern electrochemical analysis.³² SPEs are easy to fabricate, cheap, require small sample volume and their sensitivity and selectivity can be enhanced easily by immobilizing suitable substances on the surface.³³ To prepare the electrode, citrate capped GNPs were deposited on the surface of the SPE by bulk electrodeposition procedures by applying a static potential of +0.7 V to a solution of negatively charged GNPs. The relative current response decreased with time indicating deposition of GNPs on the surface of electrodes (Fig. S1†). After electrodeposition, the antibody was immobilized on the surface by the process of physical adsorption by overnight incubation of the antibody over it. To characterize the modified surface, the changes occurring at each stage of the electrode preparation for example (a) bare, (b) after GNP electrodeposition (SPE/GNP) and (c) after antibody immobilization (SPE/GNP/Ab) was monitored by cyclic voltammetry. The CV response scans recorded on SPE/GNP, for the redox of small ion (Fe²⁺), showed an increase in the intensity of current signal for oxidative and reductive peaks due to increase in surface area of the electrodes indicating proper deposition of the GNPs (Fig. 3). The increase in the voltammetric current indicates the increase in conductivity due to electrodeposition of the citrate capped GNPs on the surface of the SPE. As expected, a decrease in voltammetric current was observed after immobilization of the bulky MAbD2 on the GNP/SPE surface. This is due to the increase in charge transfer resistance between the electrode and solution ions due to the deposition of the insulating antibody.

Fig. 3 Cyclic voltammetry showing different surface modifications of the electrodes and binding of the P. falciparum infected cells to the modified surface.

To test the capacity of the sensor to detect the infected cells, infected RBC solution from the dilution of 10² cells per mL to 10⁷ cells per mL was applied to the electrode and allowed to bind to the Ab on the surface. After removing the unbound cells, the CV measurement was recorded in ferro/ferri couple. The current change observed for the peak current from one dilution to another was low or negligible showing that the sensitivity of the platform may not provide high accuracy of quantitative information about the number of cells bound (Fig. S2a†). However, only a small change in the peak current signal was observed at a high cell dilution of 10⁷ cells per mL indicating cell binding onto the electrode surface. Nevertheless, the CV data enabled the monitoring of surface changes on the electrodes and demonstrated the high stability of GNP deposited sensors that showed a slight variation from one scan to another.

Since, the resistance change due to the binding of cells observed in CV was less between one dilution to another, we switched to frequency scanning using impedance spectroscopy to probe deeper into the changes occurring on the surface of the electrodes. Electrochemical impedance spectroscopy (EIS) is a powerful tool to measure the interfacial resistance or capacitance occurring due to the binding of analytes.³⁴ Certain previous reports have also highlighted the importance of impedance biosensors for the detection of bacterial cells using antibody and aptamer biosensors where functionalized poly[pyrrole-co-3-carboxyl-pyrrole] connecting polymer and diazonium grafting based supporting layer on SPE were used for receptor immobilization by previous researchers.^35–37 We have employed gold nanoparticles for functionalization and transducer element for localized field strength enhancement as also elucidated in our previous reports.^28,38,39 They act as fringes protruding from the surface of the electrodes and allow the antibody to establish on its surface mainly by metal binding to primary amine groups present on the antibody during prolonged overnight antibody incubation.³⁹ The other complementing condition is the high amount of chloride ion concentration of 150 mM and the redox couple 5 mM each of the ferri/ferro was used to obtain a good change in impedance due to cells binding. To characterize the surface of the electrodes at every step of modification the impedance spectra was recorded. The resulting impedance spectra were recorded at each of the steps of the sensor preparation that is bare, SPE/GNP and SPE/GNP/Ab. The impedance measurement after electrodeposition of the GNP on the surface of SPE showed a decrease in semi-circular half of the Nyquist plot (Fig. 4) indicative of an increase in conductance of the electrodes. The resistance due to charge transfer dropped from ∼9.8 kΩ for bare electrodes to ∼6.2 kΩ for GNP modified surface as calculated by equivalent circuit modeling of EIS data. The immobilization of the antibody resulted in an increase in the semicircle showing an increase in resistance to ∼9.9 kΩ. The binding of the cells to cells per mL resulted in a larger increase in R_ct indicated by the significant increase in the diameter of the semicircle and the resistance to charge transfer increased to 12.36 kΩ. Only the Nyquist plot at the maximum concentration of 10⁷ cells per mL has been shown in Fig. 4 for simplicity. Impedance measurements at a concentration range of 10² cells per mL to 10⁷ cells per mL resulted in a well-resolved Nyquist plots (Fig. S2b†) thus indicating that the developed platform may be useful for quantitation of infected cells.

Fig. 4 EIS spectrum scans showing different surface modifications of the electrodes and binding of the P. falciparum infected cells to the modified surface.

Monitoring of immunochemical reactions

To derive a relationship between the changes in electrical parameters due to the binding of the cells, impedance measurement data was recorded for the cells at different dilutions from 10² cells per mL to 10⁸ cells per mL. The interfacial changes at the electrode surface were monitored by impedance spectroscopy. The data was recorded using faradic impedance spectroscopy in ferricyanide redox couple. The Nyquist plots obtained after various surface modifications are shown in Fig. 5a. The data was analyzed using the modified Randle's equivalent circuit^40,41 shown in Fig. 5a. The circuit includes the following five elements: (i) solution resistance (R_sol), (ii) Warburg impedance (Z_W), (iii) double layer capacitance (C_dl), (iv) charge transfer resistance (R_ct),⁴² and (v) constant phase element (CPE).⁴³ The C_dl is associated with the electrical double-layer reflecting the interface between the GNP/MAbD2 platform and the solution ions. The circuit completed using CPE element in series, showing reactance and resistance components, provided a good fit to the data. The use of CPE element is logical since we did not make any specific attempt to block the ion pathways to the electrodes as these were still able to access and electrically discharge onto the electrode surface. It is noteworthy that ethylene glycol based polymers are commonly used to protect the electrodes from ions in the solution.⁴⁴ The impedance data was fitted using software by CH Instruments, USA (version 10.06). The obtained results, using Randle's equivalent circuit modeling, showed good agreement between the circuit model and the measurements obtained (Table 1). Fig. 5a shows the EIS response of the SPE/GNP/Ab surface to the different number of the IRBCs. The origin of the plots is representative of the solution resistance (R_sol) and remained the same for all the plots indicating the obvious no change in solution resistance. The diameter of the semicircle, representative of electron transfer resistance R_ct, increased upon increasing the cell number. The total observed window change in R_ct for the tested number of IRBC (10² cells per mL to 10⁸ cells per mL) was roughly equivalent to 2000 Ohms. The R_ct is determined by the changes in surface conductivity of the electrodes as a result of cells binding while C_dl, which is related to double-layering phenomena, shows a minor change of 5 nF for the entire range of tested cells concentration (Table 1). This is obvious because red blood cells are large and substantially affect charge transfer resistance but do not drastically change capacitance. The Z_W that represents the properties of the electrolyte solution and diffusion of the redox probe showed negligible changes up to 10⁷ cells and after that decreased at 10⁸ cells per mL showing that it mainly gets affected at high IRBC concentration. Thus, inferences for the IRBC binding can be elucidated from R_ct that increases with increase in the concentration of the IRBC in an almost linear fashion (Fig. 5b). To rule out any non-specific binding of IRBCs to the sensor surface, the data was obtained with the control antibody (MAbA3E1) that does not recognize IRBC. No change in impedance was observed showing the specific nature of MAbD2 binding to IRBC (Fig. S3†). Further, the ability of the sensor to specifically detect the infected cells, the changes in response due to binding of IRBCs and NRBCs to SPE/GNP/MAbD2 were monitored. A linear increase in EIS response was observed in the case of the IRBCs (Fig. 6a & S4†) with increase in the concentration of cells from 10² cells per mL to 10⁶ cells per mL whereas no such binding response was observed in the case of the NRBCs (Fig. 6b). These experiments conclusively showed that the response obtained in Fig. 5 was solely due to the specific binding of IRBCs to the MAbD2 rendering a good possibility of quantitative cell detection. Experiments were also done to check the change in electrical parameters upon binding to IRBC carrying other two stages of the parasite life cycle i.e. ring and the trophozoite stage. Like schizont stage-IRBC, a linear increase in R_ct, was observed with trophozoite stage-IRBC at increasing cell concentration i.e. from 10² cells per mL to 10⁷ cells per mL. However, no change could be seen in case of the ring stage-IRBC (Fig. S6†).

Fig. 5 (a) Nyquist plots showing the change in EIS spectra with different number of P. falciparum infected cells on modified electrode surface (SPE/GNP/MAbD2). The measurements were recorded in 5 mM ferricyanide/ferrocyanide couple at pH 7.4 in PBS buffer in the frequency range from 1 Hz to 100 kHz. Inset shows the Randles equivalent circuit used for fitting the data. (b) Plot representing linear fit of R_ct versus log[number of infected cells] with R squared value of 0.988 showing linear relationship.

Table 1 Changes in the electrical parameter at the SPE/GNP/Ab modified surface due to P. falciparum infected cell binding. The scans were obtained in 5 mM K₃[Fe(CN)₆]/K₄[Fe(CN)₆] (1 : 1) mixture in PBS buffer pH = 7.4

No of cells per mL	Rsol (Ω)	Cdl (nF)	Rct (Ω)	W	CPE (10^−4)	Error
10^2	150.8	221	5566	1000	2.48	0.07487
10^3	150.5	221	5678	1000	2.48	0.07539
10^4	148.9	222	6093	901.3	2.21	0.07401
10^5	148.5	222	6430	1000	2.09	0.07271
10^6	147.6	223	6750	1000	1.99	0.07151
10^7	146.4	224	7189	1000	1.87	0.0704
10^8	130.2	225	7618	699.2	1.78	0.06402

---

Fig. 6 Nyquist plot showing the change in EIS spectra obtained due to binding of IRBC (a) or NRBC (b) to SPE/GNP/MAbD2.

We have thus shown that the interaction of IRBC on the modified electrode results in a well resolved Nyquist plots and linear relationship between electron transfer resistance and logarithm of the number of infected cells is obtained. In this study, the experiments were done to test the proof of concept using Pf-IRBC. As the P. falciparum parasites mature within the red blood cells to trophozoite and schizont stages, they disappear from the peripheral circulation and localize specifically in the vascular beds of organs such as the brain, through a process named sequestration.^45,46 Thus, the most accessible form in the blood is the ring stage and occasionally a few mature forms.^47,48 Results presented here show that MAbD2 does not react with the ring-stage IRBC. Therefore, applicability of this method/platform in diagnosis of P. falciparum infection will require ring stage IRBC specific antibody for immobilization on the electrode. Udomsangpetch et al., reported increased export of PfEMP1 protein on the surface of the ring stage RBC such that an antibody B6mAb, non reactive to PfEMP1 ring, reacted upon subjecting the culture for one hour at 40 °C due to increased export of PfEMP1 to the surface.⁴⁹ This observation, coupled with the use of ring-stage IRBC antibody/ligand, could be useful in exploiting this platform in diagnosis of falciparum malaria. On the other hand, this method may hold a promise in detection of malaria caused by P. vivax, P. ovale and P. malariae where sequestration is absent except for a few emerging evidence in P. vivax.^47,50 Thus, this work provides a proof of concept for the development of an immunosensor based detection of malaria infected whole red blood cells.

Conclusions

In the present study, we have described for the first time a cell based label-free detection of P. falciparum infected red blood cells using GNP enhanced sensing platform using a monoclonal antibody. The platform establishes the usefulness of impedance spectroscopy and equivalent circuit modelling based techniques for cell based detection of malaria infected red blood cells. The use of such an approach employing whole cells to detect malaria infection may help in overcoming the paucity of known biomarkers in malaria. The main advantage of this strategy is the one-step preparation of GNP enhanced surface that enables label-free impediometric detection of P. falciparum infected cells. This approach may find application in the development of disposable sensors for rapid onsite detection of malaria.

Ethical statement

All the experiments and the work with human blood were done following protocol and guidelines approved by the Institutional Biosafety Committee (IMTECH/IBSC/2013/25), constituted by the Department of Biotechnology, Govt. of India. Fortis Hospital/Max Hospital, Mohali, Punjab provided packed red blood cells of O+ healthy donors, for in vitro culture of parasite, following hospital guidelines with informed consent.

Acknowledgements

This work was supported by funds provided by the Council of Scientific and Industrial Research projects (SPLenDID BSC0104 and GENESIS BSC0121), Govt. of India. This is IMTECH communication number; 0138/2015.

References

1. World Health Organization, World Malaria Report 2014, World Health Organization, 2014.
2. A. Trampuz, M. Jereb, I. Muzlovic and R. M. Prabhu, Crit. Care, 2003, 7, 315–323.
3. B. Cundill, H. Mbakilwa, C. I. Chandler, G. Mtove, F. Mtei, A. Willetts, E. Foster, F. Muro, R. Mwinyishehe, R. Mandike, R. Olomi, C. J. Whitty and H. Reyburn, BMC Med., 2015, 13, 118.
4. H. Reyburn, Br. Med. J., 2010, 340, c2637.
5. WHO, Guidelines for the treatment of malaria, 2015.
6. W. K. Peng, T. F. Kong, C. S. Ng, L. Chen, Y. Huang, A. A. S. Bhagat, N.-T. Nguyen, P. R. Preiser and J. Han, Nat. Med., 2014, 20, 1069–1073.
7. C. W. Pirnstill and G. L. Coté, Sci. Rep., 2015, 5, 1–13.
8. World Health Organization, Malaria rapid diagnostic test performance, Results of WHO product testing of malaria RDTs: Round 5 (2013), World Health Organization, 2014.
9. J. Maltha, P. Gillet and J. Jacobs, Clin. Microbiol. Infect., 2013, 19, 408–415.
10. J. Maltha, P. Gillet and J. Jacobs, Clin. Microbiol. Infect., 2013, 19, 399–407.
11. N. Kumar, V. Pande, R. M. Bhatt, N. K. Shah, N. Mishra, B. Srivastava, N. Valecha and A. R. Anvikar, Acta Trop., 2013, 125, 119–121.
12. B. Mishra, J. C. Samantaray, A. Kumar and B. R. Mirdha, J. Clin. Microbiol., 1999, 37, 1233.
13. A. Moody, Clin. Microbiol. Rev., 2002, 15, 66–78.
14. J. C. Mouatcho and J. P. D. Goldring, J. Med. Microbiol., 2013, 62, 1491–1505.
15. J. H. Kattenberg, I. Versteeg, S. J. Migchelsen, I. J. González, M. D. Perkins, P. F. Mens and H. D. F. H. Schallig, mAbs, 2012, 4, 120–126.
16. F. Perandin, N. Manca, A. Calderaro, G. Piccolo, L. Galati, L. Ricci, M. C. Medici, M. C. Arcangeletti, G. Snounou, G. Dettori and C. Chezzi, J. Clin. Microbiol., 2004, 42, 1214–1219.
17. A. Chopra, S. Tuteja, N. Sachdeva, K. K. Bhasin, V. Bhalla and C. R. Suri, Biosens. Bioelectron., 2013, 44, 132–135.
18. M. K. Sharma, V. K. Rao, G. S. Agarwal, G. P. Rai, N. Gopalan, S. Prakash, S. K. Sharma and R. Vijayaraghavan, J. Clin. Microbiol., 2008, 46, 3759–3765.
19. S. Lee, K.-M. M. Song, W. Jeon, H. Jo, Y.-B. B. Shim and C. Ban, Biosens. Bioelectron., 2012, 35, 291–296.
20. T. F. Kong, W. Ye, W. K. Peng, H. W. Hou, P. R. Preiser, N.-T. Nguyen and J. Han, et al., Sci. Rep., 2015, 5, 11425.
21. E. Y. Lukianova-Hleb, K. M. Campbell, P. E. Constantinou, J. Braam, J. S. Olson, R. E. Ware, D. J. Sullivan and D. O. Lapotko, Proc. Natl. Acad. Sci. U. S. A., 2014, 111, 900–905.
22. L. S. Garcia, Diagnostic medical parasitology, ASM Press, Washington, DC, 2007.
23. A. Radfar, D. Méndez, C. Moneriz, M. Linares, P. Marín-García, A. Puyet, A. Diez and J. M. Bautista, Nat. Protoc., 2009, 4, 1899–1915.
24. J. Miao and L. Cui, Nat. Protoc., 2011, 6, 140–146.
25. M. Owais, G. C. Varshney, A. Choudhury, S. Chandra and C. M. Gupta, Antimicrob. Agents Chemother., 1995, 39, 180–184.
26. J. M. Reul, I. Stec, G. J. Wiegers, M. S. Labeur, A. C. Linthorst, E. Arzt and F. Holsboer, J. Clin. Invest., 1994, 93, 2600.
27. G. Frens, Nature, 1973, 241, 20–22.
28. V. Bhalla, P. Sharma, S. K. Pandey and C. R. Suri, Sens. Actuators, B, 2012, 171–172, 1231–1237.
29. J. A. Boddey and A. F. Cowman, Annu. Rev. Microbiol., 2013, 67, 243–269.
30. M. Foley and L. Tilley, Int. J. Parasitol., 1998, 28, 1671–1680.
31. A. G. Maier, B. M. Cooke, A. F. Cowman and L. Tilley, Nat. Rev. Microbiol., 2009, 7, 341–354.
32. A. Hayat and J. L. Marty, Sensors, 2014, 14, 10432–10453.
33. K. K. Mistry, T. Sagarika Deepthy, C. R. Chaudhuri and H. Saha, Curr. Sci., 2015, 109, 1427–1436.
34. F. Lisdat and D. Schäfer, Anal. Bioanal. Chem., 2008, 391, 1555–1567.
35. Z. Bagheryan, J.-B. Raoof, M. Golabi, A. P. F. Turner and V. Beni, Biosens. Bioelectron., 2016, 80, 566–573.
36. E. Sheikhzadeh, M. Chamsaz, A. P. F. Turner, E. W. H. Jager and V. Beni, Biosens. Bioelectron., 2016, 80, 194–200.
37. S. Ghosh Dastider, S. Barizuddin, N. S. Yuksek, M. Dweik, M. F. Almasri, S. Ghosh Dastider, S. Barizuddin, N. S. Yuksek, M. Dweik and M. F. Almasri, J. Sens., 2015, 2015(2015), e293461.
38. R. Mutreja, M. Jariyal, P. Pathania, A. Sharma, D. K. Sahoo and C. R. Suri, Biosens. Bioelectron., 2016, 85, 707–713.
39. V. Bhalla, S. Carrara, P. Sharma, Y. Nangia and C. Raman Suri, Sens. Actuators, B, 2012, 161, 761–768.
40. F. Patolsky, M. Zayats, E. Katz and I. Willner, Anal. Chem., 1999, 71, 3171–3180.
41. J. E. B. Randles, Discuss. Faraday Soc., 1947, 1, 11–19.
42. A. Bard and L. Faulkner, Electrochemical methods: fundamentals and applications, Wiley, 2001.
43. S. Carrara, V. Bavastrello, D. Ricci, E. Stura and C. Nicolini, Sens. Actuators, B, 2005, 109, 221–226.
44. S. Carrara, L. Benini, V. Bhalla, C. Stagni, A. Ferretti, A. Cavallini, B. Riccò and B. Samorì, Biosens. Bioelectron., 2009, 24, 3425–3429.
45. L. H. Miller, D. I. Baruch, K. Marsh and O. K. Doumbo, Nature, 2002, 415, 673–679.
46. K. R. Hughes, G. A. Biagini and A. G. Craig, Mol. Biochem. Parasitol., 2010, 169, 71–78.
47. B. O. Carvalho, S. C. P. Lopes, P. A. Nogueira, P. P. Orlandi, D. Y. Bargieri, Y. C. Blanco, R. Mamoni, J. A. Leite, M. M. Rodrigues, I. S. Soares, T. R. Oliveira, G. Wunderlich, M. V. G. Lacerda, H. A. del Portillo, M. O. G. Araújo, B. Russell, R. Suwanarusk, G. Snounou, L. Rénia and F. T. M. Costa, J. Infect. Dis., 2010, 202, 638–647.
48. K. Silamut, N. H. Phu, C. Whitty, G. D. Turner, K. Louwrier, N. T. Mai, J. A. Simpson, T. T. Hien and N. J. White, Am. J. Pathol., 1999, 155, 395–410.
49. R. Udomsangpetch, B. Pipitaporn, K. Silamut, R. Pinches, S. Kyes, S. Looareesuwan, C. Newbold and N. J. White, Proc. Natl. Acad. Sci. U. S. A., 2002, 99, 11825–11829.
50. K.-A. Strydom, F. Ismail and J. Frean, Malar. J., 2014, 13, 85.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra07665c

‡ Equal contribution.

---