Genetic modification of Plasmodium falciparum has provided an invaluable molecular tool to dissect the biology and pathogenesis of this important human pathogen [1–3]. Introduction of exogenous DNA, however, remains a relatively inefficient and costly enterprise, both in terms of reagents and time, and represents a major limiting step in any investigation. A range of approaches have been used to introduce exogenous DNA into P. falciparum infected erythrocytes (IE), including; chemical agents, biolistic delivery and electroporation, with electroporation proving the only consistently reliable approach [2, 4]. Electroporation of ring stage IE was first described in 1995 by Wu et al., who utilized a high voltage/low capacitance electric pulse [5]. Subsequently, a low voltage/high capacitance electric pulse was shown to be more efficient and remains in wide use throughout the community [6]. A final evolution, pioneered by Deitsch et al. was the pre-loading of erythrocytes with exogenous DNA, which were then mixed with late stage IE, where passive spontaneous uptake of DNA by parasites presumably occurs following invasion into the pre-loaded erythrocytes during the subsequent cycle of asexual intra-erythrocytic development [7].

To date, only a comparison of the relative efficiency of these different transfection approaches has been carried out; although these reveal a clear preference for the use of pre-loaded erythrocytes [4]. Of note, however, from these studies is that direct electroporation of IE is cited as having a much lower overall success rate of only 25% [4]. This contrasts with the experience of a number of laboratories, where overall success rates of 70-80% are routinely achieved. Moreover, determining the relative efficiency of these approaches relies on single, or at best two, subjective endpoints; typically either days post-transfection when recovering parasites are first observed or when the cultures reach a 1% parasitaemia [4, 8].

Here a quantitative investigation of the absolute and relative efficiency of direct electroporation of IE and pre-loaded erythrocytes is reported. Using a luciferase reporter construct as the source of exogenous DNA, and an improved single-step lysis protocol for the qualitative measurement of luciferase levels from small samples of IE [9], a time-course for recovery of the transfected parasites post-electroporation can be carried out. In addition, a combination of these two electroporation approaches is investigated here. Ring stage IE are directly electroporated, allowed to mature for 24 hrs and then mixed with pre-loaded erythrocytes; termed here the “double-tap” technique. Using two concentrations of DNA in a “double-tap” also allows the transfection efficiency per unit concentration of DNA to be monitored.

The aim of this study was to compare the absolute and relative efficiencies of three electroporation-based transfection techniques. The standard techniques for direct electroporation of ring stage IE (transfections 1–3, Figure 1A, Protocol 1) and use of preloaded erythrocytes mixed with mature stage IE (transfections 4–6, Protocol 2) were compared with a novel “double-tap” technique that uses two “taps” of either 40 μg (transfections 7–9, Protocol 3) or 20 μg (transfections 10–12, Protocol 4) of plasmid. Using a plasmid bearing a luciferase reporter cassette, and an improved single-step lysis protocol to measure luciferase expression in small samples of IE, a quantitative monitoring of parasite growth can be readily employed to monitor the logarithmic growth of parasites that have successfully been transfected. Care was taken throughout the experiment to ensure as near as possible matched conditions for the transfection and subsequent outgrowth of cultures to facilitate a meaningful comparative analysis.

Examination of the timecourse of luciferase activity for these 12 experiments initially shows a background signal of some 30–50 relative light units (RLU), with an exponential increase in the luciferase signal from varying timepoints as the population of successfully transfected parasites grows, crosses the luciferase detection threshold and then continue to proliferate (Figure 1B). To allow the RLU measured to be correlated with parasite numbers, a standard curve was generated from a sequential dilution of the parasite clone Pfluc. Pfluc is a derivative of the AHEI clone where the pΔ1 plasmid has been integrated into the cg 6 locus on chromosome 7. This standard curve (data not shown) plots log₁₀ RLU against parasite number and shows a strong linear relationship (R² 0.98) from a detection threshold of some 200 IE/40 μl sample to greater than 10⁵ parasites/40 μl sample. A non-linear exponential regression analysis of parasite numbers over time indicates the fold-increase in parasite number per cycle ranged from between 2.25-2.75, showing no correlation with the electroporation protocol employed. For this reason, the median value of a 2.5 fold increase in parasitaemia per cycle was adopted in all further calculations. Whilst resulting in all subsequent calculations being an approximation, the demonstration that growth rates across all the transfection experiments was similar, did provide important confirmation that efforts to ensure identical growth conditions post-transfection were effective.

Using all parameters measured (luciferase expression and microscopic confirmation of at least 1% parasitaemia), preloaded erythrocytes alone (transfections 4–6, Protocol 2) was the most efficient (mean efficiency of 9.59x10^-6) protocol. All transfections recovered to at least a 1% parasitaemia by Day 17, which suggests that between 1,800-2,100 parasites were successfully transfected at the onset of the experiment. Whilst some parasites may have been transfected on subsequent reinvasion of a preloaded erythrocyte, the application of drug selection with 24 hours on Day 2 makes this unlikely. After preloaded erythrocytes, the “double-tap” and direct electroporation of ring IE alone show relative decreases in efficiencies of 5.6 and 184-fold, respectively. This comparison was made against transfections 1–3 (mean efficiency of 5.2x10^-8, Protocol 1) and 10–12 (mean efficiency of 1.71x10^-6, Protocol 4) which both use the same total amount (40 μg) of plasmid. Transfections 7–9 (“double-tap” of 40 μg plasmid) are almost twice as efficient (mean efficiency of 3.17x10^-6, Protocol 3) as when a “double-tap” of 20 μg plasmid was used (mean efficiency of 1.71x10^-6), which indicates under the parameters explored here a linear response in dose-dependent efficiency. Interestingly, the absolute efficiency of both “double-tap” protocols was much closer to that of the preloaded erythrocyte, which suggests that their success is largely attributed to the use of preloaded erythrocytes. It is likely that the measured decrease in efficiency is the result of a kill induced by the electrical current applied directly to the ring-stage parasites on Day 0.

Using the technique for routine quantitative monitoring of growth established here, other questions relating to the optimal conditions for electroporation-based transfections would appear to be simple to address in the future. For example, what concentration of preloaded plasmid DNA is optimal? Or, what is the optimal ratio of mature IE and preloaded erythrocyte? The most efficient transfection reported here of 1.08x10^-5 needs to be significantly improved to overcome this “transfection bottleneck” should high throughput phenotypic studies of genetically-modified parasites become a routine tool. That said, recent improvements in transfection techniques, such as use of 96-well electroporation [14] and suspension cultures [15] go some way to address these limitations. In addition, there has been some success in increased transfection efficiency reported as a result of the sequential addition of preloaded erythrocytes over several days at the start of an experiment [7]. Whilst this was not explored here as the study was attempting a side-by-side comparison of techniques under the same starting conditions, evidence here from the “double-tap” transfections would suggest that this modification should be adopted more widely.

This study provides a procedure to quantitatively assess both the relative and absolute transfection efficiencies of three electroporation protocols that clearly supports the use of preloaded erythrocytes. An exploration of the efficiency of a novel “double-tap” technique, combining previous standard protocols, shows that this technique is some six-fold less efficient than preloaded erythrocytes alone. The almost 200-fold increase in efficiency when preloaded erythrocytes are compared to direct electroporation of IE suggests that any transfection procedure that applies an electrical current directly to the IE parasite should be avoided and that this observation may have wider implications for the electroporation of any other intracellular organism.