A Plasmodium falciparum screening assay for anti-gametocyte drugs based on parasite lactate dehydrogenase detection

A Plasmodium falciparum screening assay for anti-gametocyte drugsbased on parasite lactate dehydrogenase detection

Sarah D’Alessandro1, Francesco Silvestrini2, Koen Dechering3, Yolanda Corbett1, Silvia Parapini1,Martijn Timmerman3, Laura Galastri4, Nicoletta Basilico5*, Robert Sauerwein6, Pietro Alano2

and Donatella Taramelli1

1Dipartimento di Scienze Farmacologiche e Biomolecolari, Universita degli Studi di Milano, 20133 Milan, Italy; 2Dipartimento di MalattieInfettive, Parassitarie, Immunomediate, Istituto Superiore di Sanita, 00161 Rome, Italy; 3TropIQ Health Sciences, Nijmegen 6525 GA, TheNetherlands; 4Associazione Italiana Volontari Sangue (AVIS) Comunale di Milano, 20133 Milan, Italy; 5Dipartimento di Scienze Biomediche,Chirurgiche e Odontoiatriche, Universita degli Studi di Milano, 20133 Milan, Italy; 6Department of Medical Microbiology, Radboud University

Nijmegen Medical Center, Nijmegen 6525 GA, The Netherlands

*Corresponding author. Dipartimento di Scienze Biomediche, Chirurgiche e Odontoiatriche, Universita degli Studi di Milano, via Pascal,36, 20133 Milan, Italy. Tel: +39-02-5031-5069; Fax: +39-02-5031-5093; E-mail: [email protected]

Received 19 December 2012; returned 15 February 2013; revised 29 March 2013; accepted 2 April 2013

Objectives: Plasmodium gametocytes, responsible for malaria parasite transmission from humans to mosquitoes,represent a crucial target for new antimalarial drugs to achieve malaria elimination/eradication. We developed anovel colorimetric screening method for anti-gametocyte compounds based on the parasite lactate dehydrogen-ase (pLDH) assay, already standardized for asexual stages, to measure gametocyte viability and drugsusceptibility.

Methods: Gametocytogenesis of 3D7 and NF54 Plasmodium falciparum strains was induced in vitro and asexualparasites were depleted with N-acetylglucosamine. Gametocytes were treated with dihydroartemisinin, epoxomi-cin, methylene blue, primaquine, puromycin or chloroquine in 96-well plates and the pLDH activity was evaluatedusing a modified Makler protocol. Mosquito infectivity was measured by the standard membrane feeding assay(SMFA).

Results: A linear correlation was found between gametocytaemia determined by Giemsa staining and pLDH activ-ity. A concentration-dependent reduction in pLDH activity was observed after 72 h of drug treatment, whereas anadditional 72 h of incubation without drugs was required to obtain complete inhibition of gametocyte viability.SMFA on treated and control gametocytes confirmed that a reduction in pLDH activity translates into reducedoocyst development in the mosquito vector.

Conclusions: The gametocyte pLDH assay is fast, easy to perform, cheap and reproducible and is suitable forscreening novel transmission-blocking compounds, which does not require parasite transgenic lines.

Keywords: malaria, gametocytes, drug screening

IntroductionMalaria is a protozoan parasitic disease endemic in .100 coun-tries, with a profound negative impact on productivity and eco-nomic development. Despite a reduction of incidence by 17%and of mortality rates by 26% since 2000 (due to intensifiedcontrol measures and increased funding), malaria still accountedfor .200 million cases and .600000 deaths in 2010.1

Plasmodium falciparum is responsible for the most severeform of the disease. Parasites are transmitted by female Anophelesmosquitoes, which inject sporozoites into the human host,

causing an asymptomatic hepatic infection. This is followed bythe intraerythrocytic proliferation responsible for the symptomsof malaria and the fatal complications of the disease, such assevere anaemia and cerebral malaria. Inside red blood cells(RBCs) the parasites replicate asexually several times, while somedevelop into sexual forms called gametocytes. Male and femaleP. falciparum gametocytes mature through five stages of develop-ment (from I to V), which last �10 days in total. Mature gameto-cytes persist in the circulation for several days and can be takenup by a mosquito, where they mature into gametes. Sexualreproduction ensues inside the mosquito midgut, followed by the

# The Author 2013. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved.For Permissions, please e-mail: [email protected]

J Antimicrob Chemotherdoi:10.1093/jac/dkt165

1 of 11

Journal of Antimicrobial Chemotherapy Advance Access published May 3, 2013 at Istituto Superiore di Sanita on M

ay 10, 2013http://jac.oxfordjournals.org/

Dow

nloaded from

http://jac.oxfordjournals.org/

generation of infective sporozoites, which migrate to the insect sal-ivary gland.

Over recent decades the international strategy against malariahas been based on disease control measures: prevention of mos-quito bites (insecticide-treated nets and residual indoor spraying)and drug treatment with artemisinin-based combination therapies(ACTs).2 More recently, however, the perspective has changedtowards malaria elimination, and ultimately eradication. Toachieve this goal new drugs inhibiting liver stage reproduction orgametocyte development, the latter directly blocking parasitetransmission, are a necessary element.3 – 6 Drug screeningmethods suitable for the identification of anti-gametocyte com-pounds are therefore urgently needed. These kinds of assays arepresently the object of investigation as standardized gametocytecultivation protocols are not fully developed.7 Moreover, very fewgametocytocidal reference compounds exist to validate newassays.

In the last few years new methods of screening drugs againstgametocytes have been published.8 These methods are laborious:some require one or more gametocyte purification steps by Percollgradient or magnetic bead column separation, and others requirethe use of reporters in transgenic parasites, making them unsuit-able for field isolates.9 – 11

Here, we report the development of a novel, fast and cheapassay for the screening of anti-gametocyte compounds measuringthe parasite lactate dehydrogenase (pLDH) activity of gameto-cytes in 96-well plates. The method was tested by using referenceanti-gametocyte drugs and by comparing the assay readout withmeasures of mosquito infectivity by the standard membranefeeding assay (SMFA).

Materials and methods

P. falciparum culture and gametocytogenesisThe strains 3D7 and NF54 of P. falciparum were cultured in vitro as describedby Trager and Jensen with minor modifications.12 Briefly, parasites weremaintained in human type 0-positive RBCs at 5% haematocrit (HCT) inRPMI 1640 medium containing 24 mM sodium bicarbonate (Euroclone),with the addition of 10% (v/v) naturally clotted heat-inactivated 0+human serum (Interstate Blood Bank, Inc.), 0.37 mM hypoxanthine(Sigma-Aldrich) and 2 mM L-glutamine (Euroclone). The cultures weremaintained at 378C in a standard gas mixture consisting of 1%–3% O2,5% CO2 and 92%–94% N2.

To trigger gametocytogenesis, 5 or 25 mL cultures in 25 or 75 cm2

flasks, respectively, were diluted to 0.5% parasitaemia at 5% HCT.Medium was changed daily for 3 days until .5% parasitaemia wasreached and unhealthy, starved asexual parasites were detectable. Atthis point the HCT was lowered to 2.5% and the culture was treated for48–72 h with 50 mM N-acetylglucosamine (NAG; Sigma-Aldrich) in orderto clear residual asexual parasites and obtain a virtually pure gametocyteculture, typically containing 2–4 gametocytes/100 RBCs at late stage IIto early stage III of maturation. In some experiments, gametocytes wereenriched by Percoll density gradient centrifugation.13

Gametocyte maturation in microwell platesTo analyse gametocyte development in 96-well plates, asexual parasitecultures were synchronized by purifying schizonts with MACS SeparationColumns CS (Miltenyi Biotec) and using them to reinvade erythrocytes.After two rounds of reinvasion the cultures were treated for 72 h with

NAG in order to clear residual asexual parasites and obtain a virtuallypure gametocyte culture.

Aliquots of 200 mL of synchronized gametocyte culture (typically at2%–4% gametocytaemia), diluted to 1% HCT, were seeded in 96-well flat-bottomed plates. To define the optimal protocol for medium replacement,gametocyte viability was evaluated in wells where medium was replaced atdifferent intervals: every 24, 48 or 72 h. Medium replacement was per-formed with a multichannel pipette without plate centrifugation. Smearswith pooled parasitized RBCs from duplicate wells were prepared everyday and stained with Giemsa to determine the total gametocytaemiaand the percentage of each gametocyte maturation stage.

Gametogenesis was measured by treating samples with 50 mMxanthurenic acid (XA; Sigma-Aldrich) in exflagellation buffer (RPMI 1640with 20 mM HEPES, 4 mM sodium bicarbonate, pH 8.0).14 After 5 min of in-cubation at room temperature, Giemsa-stained smears were prepared andthe numbers of macrogametes and non-activated gametocytes werecounted.

Set up of pLDH assay on gametocytesParasites were plated in 96-well plates and serial dilutions to obtain differ-ent percentages of gametocytaemia or different HCTs were performed inplates with uninfected RBCs or mediumalone, respectively.Parasite viabilitywas determined spectrophotometrically by measuring the activity ofthe pLDH, according to a modified version of the method of Makler andHinrichs.15 Briefly, 100 mL of the Malstat reagent [0.11% v/v Triton-100;115.7 mM lithium L-lactate; 30.27 mM Tris; 0.62 mM 3-acetylpyridineadenine dinucleotide (APAD; Sigma-Aldrich); adjusted to pH 9 with 1 MHCl] was mixed with 25 mL of PES/NBT (1.96 mM nitro blue tetrazoliumchloride; 0.24 mM phenazine ethosulphate). Routinely, a fixed volume of20 mL of parasite suspension was mixed with the reagents and opticaldensity (OD) was measured with a Synergy4 reader (BioTek) at the wave-length of 650 nm.

The signal to background ratio (S/B), the signal to noise ratio (S/N) andthe Z′ factor were calculated according to the formulae reported below.16

All the parameters were calculated using means (m) and SD (s) of ODcontrol gametocytes (c+) and OD blank uninfected RBCs (c2) from tripli-cate or quadruplicate wells.

SB= mc+/mc−

SN= (mc+ − mc−)

sc−

Z′ = 1 − (3sc+ + 3sc−)|mc+ − mc−|

Drug treatmentAll the drugs were purchased from Sigma-Aldrich. Dihydroartemisininwas either provided by Professor Richard K. Haynes, (HKUST, Hong Kong)or from Sigma-Aldrich. The compounds were dissolved in either culturemedium (methylene blue, primaquine and chloroquine) or DMSO(dihydroartemisinin, epoxomicin and puromycin) and then diluted withculture medium to achieve the required concentrations (final DMSO con-centration ,0.5%, which is non-toxic to gametocytes). Samples wereplaced in 96-well plates and serial dilutions were made in a final volumeof 100 mL/well. Asynchronous gametocyte cultures (100 mL/well)with gametocytaemiabetween 0.3% and 3%and final HCT 1%were distrib-uted into the plates and incubated at 378C for different times. Total incuba-tion volume was set at 200 mL. Gametocyte viability was evaluated bythe pLDH assay. The results were expressed as the percentageviability compared with untreated controls according to the following

D’Alessandro et al.

2 of 11

at Istituto Superiore di Sanita on May 10, 2013

http://jac.oxfordjournals.org/D

ownloaded from


formula: 100×(ODtreated sample2mc2) / (mc+2mc2). The percentage of via-bility was plotted as a function of drug concentration and curve fittingwas obtained by non-linear regression analysis using a four-parameter lo-gistic method [software Gen5 1.10 provided with the Synergy 4 (BioTek)reader]. The IC50 value was extrapolated as the concentration thatinduced a 50% inhibition of gametocyte viability. In addition, we calculatedcurve-related parameters for maximal inhibitory effect (Emax) and potency(P) of a compound from the fitted curve:

y = (a − d)(1 + (x/c)b) + d

where a is the (theoretical) response at concentration 0, b is the measure ofthe slope of the curve at its inflection point, c is the value of xat the inflectionpoint, d is the (theoretical) response at infinite concentration, x is concen-tration and y is response (% of control).

Starting from the equation parameters, we defined Emax and P asfollows: Emax¼1002d; P¼c.

Standard membrane feeding assay (SMFA)The SMFA was performed as described elsewhere with minor modifica-tions.17 Briefly, 500 mL of gametocyte culture obtained in an automatedtipper system was pre-treated for 24 h with methylene blue, epoxomicinor dihydroartemisinin (100–1000 nM). Fresh drugs at the same concentra-tions were added to the gametocyte-containing blood prior to mosquitofeeding. Six days after feeding, 20 mosquitoes per feeder were dissectedand oocysts were counted using a light microscope after staining of mos-quito stomach with 2% Mercurochrome. The results are expressed asoocyst prevalence, which is the mosquito infection rate, and oocystdensity, which is the mean number of oocysts per mosquito.

Statistical analysisData were expressed as mean+SD or mean+SEM and analysed using atwo-tailed Student’s t-test or by ANOVA with a level of significance ofP,0.05 or P,0.01.

Results

Gametocyte maturation in microwell plates

Gametocytes were seeded at 1% HCT in 96-well plates. Gameto-cyte development was monitored daily at different intervals ofmedium change (24, 48 or 72 h) and compared with that of game-tocytes from the same starting culture maintained in flasks instandard conditions (daily medium change; Figure 1). This analysisshowed that the percentage of gametocytes (Figure 1a) and theprogression of gametocyte morphological maturation from stageII to stage V (Figure 1b) were comparable in the flasks and in thethree regimens of medium replacement in plates.

To verify their viability, stage V gametocytes at day 8 of matur-ation were treated with XA to induce gametogenesis. The numberof spherical macrogametes was evaluated in Giemsa-stainedsmears of culture samples from flasks and microplate wells. Theresults showed a comparable induction of gametogenesis frommature gametocytes grown in the different conditions (Figure 1c).

Thus, gametocyte maturation and ability to form gametes canbe efficiently reproduced in multiwell plates, and medium replace-ment every 72 h is sufficient.

Set up of pLDH assay on gametocytes

Figure 2(a) shows the linear correlation between pLDH enzyme ac-tivity and the number of gametocytes, serially diluted with unin-fected RBCs over the range 3750–60000 (constant 1% HCT).Uninfected RBCs were used as a blank. The quality of the pLDHassay adapted to gametocytes was confirmed by calculating theZ′ factor, the S/B ratio and the S/N ratio at different percentagesof gametocytaemia of both 3D7 and NF54 P. falciparum strains.The best results were obtained using .1% gametocytaemia(Table 1). The results were highly reproducible in the three differentlaboratories that set up this protocol (Table S1; available as Supple-mentary data at JAC Online).

The pLDH activity of gametocytes was also compared with thatof asexual parasites at the same parasitaemia and HCT. It was no-ticeable that gametocytes consistently showed higher pLDH activ-ity compared with the same number of asexual parasites(Figure 2b).

A good S/B ratio was observed at all HCT values tested(Figure 2c).

Colour development of the pLDH reaction was also measured atdifferent timepoints, using parasites at 0.3% and 3% gametocy-taemia or uninfected RBCs. The S/B ratio increased with time,prolonging the length of the colorimetric reaction (data notshown). However, when the reaction was prolonged for .1 h a pre-cipitate appeared in the wells, interfering with the spectrophoto-metric analysis. For this reason, the length of assay developmentwas established at 10 or 30 min according to the proportion ofgametocytaemia (.1% or ,1%, respectively).

Some experiments were conducted using Percoll in order to in-crease the percentage of gametocytes and remove dead asexualparasites and debris from the culture. However, the OD values forthe same percentage of gametocytaemia from normal culture orafter Percoll purification were not significantly different (Table S2;available as Supplementary data at JAC Online). Therefore, thisadditional purification step was not essential.

On the basis of the results obtained in several experiments, theassay was standardized using 1% HCT and 1% gametocytaemia.

Set up of gametocyte pLDH drug assay

To verify the suitability of the pLDH assay for drug screening, a timecourse incubation of gametocytes for 24, 48 and 72 h was per-formed in the presence of different concentrations of dihydroarte-misinin, which is reported to kill young P. falciparum sexual stages invitro.9 The observed decrease in the pLDH activity wasconcentration- and time-dependent and became significant(P,0.01) after 72 h of incubation, a time that was thus chosenfor the subsequent experiments (Figure 3).

The 72 h drug screening assay was then performed using dihy-droartemisinin and three additional drugs known to be effectiveagainst gametocytes: epoxomicin, methylene blue and prima-quine.9,18,19 The results showed that all the drugs inhibitedgametocyte viability in a concentration-dependent manner(Figure 4a). However, the concentration–response curves obtainedwith the pLDH assay readout were not completely superimposableon those obtained by Giemsa staining. In particular, significantpLDH activity was still detectable at the highest concentrationsof the drugs, despite the apparent absence of healthygametocytesby microscopic observation.

A pLDH screening assay for anti-gametocyte drugs

3 of 11

JAC



ownloaded from

http://jac.oxfordjournals.org/lookup/suppl/doi:10.1093/jac/dkt165/-/DC1





At high drug concentrations, most of the gametocytes weremorphologically abnormal, clearly representing dead or unhealthyparasites (Figure 4b). The hypothesis that these unhealthy cellsretained some pLDH activity was partially confirmed when allgametocytes, irrespective of their morphology, were counted inthe Giemsa smears. In this case, the concentration–responsecurves obtained with Giemsa and pLDH were similar (Figure 4a,dashed lines).

A modified assay protocol was then performed. After the first72 h of treatment, the drug-containing medium was replacedwith fresh medium, plates were incubated for an additional 72 h

and pLDH activity was determined again. As shown in Figure 5,the concentration–response curves for dihydroartemisinin, epoxo-micin, methylene blue and primaquine after the 72+72 h protocolwere superimposable on the morphological data, showing almost100% inhibition of gametocyte viability.

IC50 values were calculated for the reference drugs presentedabove (Figures 4 and 5) and for two additional drugs: puromycin,reported to be effective on gametocytes, and chloroquine, usedas negative control drug.20,21 The IC50 values were calculatedonly after the 72+72 h assay, since after the first 72 h the inhib-ition of gametocyte viability did not reach 50% at the highest

2.4

2.2

2.0

1.8

% p

ara

ssit

ae

mia

1.6

1.4

1.2

1.0

day 2 day 3 day 4 day 5 day 6 day 7 day 8

blank

CM 24 h

flask

XA blank

CM 24 h

96 wells

XA blank

CM 48 h

96 wells

XA blank

CM 72 h

96 wells

XA

flask

24 h

48 h

72 h

40

No

. p

ara

site

s

35

30

25

20

15

10

5

0

Macrogametes

GCT V

(a) (c)

100

80

GC

T (

%) 60

40

20

0

24 h 48 h

day 2 day 3 day 4 day 5 day 6 day 7 day 8

72 h 48 h 72 h 48 h 72 h 48 h 72 h 48 h 72 h 48 h 72 h 48 h 72 h24 h 24 h 24 h 24 h 24 h 24 h

fla

sk9

6 w

ell

s9

6 w

ell

s

fla

sk9

6 w

ell

s9

6 w

ell

s

fla

sk9

6 w

ell

s9

6 w

ell

s

fla

sk9

6 w

ell

s9

6 w

ell

s

fla

sk9

6 w

ell

s9

6 w

ell

s

fla

sk9

6 w

ell

s9

6 w

ell

s

fla

sk9

6 w

ell

s9

6 w

ell

s

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

96

we

lls

GCT V

GCT IV

GCT III

GCT II

(b)

Figure 1. Gametocyte development in 96-well plates. Gametocyte culture was dispensed into 96-well plates at day 0 and medium change (CM) wasperformed every 24, 48 or 72 h. Blood smears from duplicate wells and from the control flask were prepared and counted every day. (a) Thepercentage of total gametocytes was counted microscopically by Giemsa staining. Results represent the mean+SD of three independent experimentsin duplicate. (b) The percentage of gametocytes (GCT) in the different stages was calculated by Giemsa staining. Results represent the mean of threeindependent experiments in duplicate. (c) Gametogenesis was induced by adding 50 mM xanthurenic acid (XA) in exflagellation buffer. Giemsa-stainedsmears from induced or control parasites were analysed, and the numbers of macrogametes and gametocytes were counted.


4 of 11



ownloaded from


concentrations tested (Table 2). Dihydroartemisinin, epoxomicin,methylene blue and puromycin were active against gametocytes,with IC50 values ranging from 201 nM for puromycin to 3.9 nM forepoxomicin. Primaquine, in contrast, can be considered inactive,with a high IC50 of 12515 nM. Chloroquine, as expected, was noteffective on gametocytes. The IC50 data were reproduciblebetween the Milan and Nijmegen laboratories (Table S3; availableas Supplementary data at JAC Online).

To compare the compounds at the 72 h timepoint, when theydo not achieve the 50% inhibition of pLDH activity, we calculatedcurve-related parameters for P and Emax (Table 3). The Emax fordihydroartemisinin, epoxomicin and primaquine was �50%, butprimaquine required concentrations .1000-fold higher thanthose of epoxomicin and dihydroartemisinin to induce the sameeffect (P¼30300 nM versus 7.2 nM for epoxomicin and 25.3 nMfor dihydroartemisinin). Methylene blue was less active at 72 h,

0.50(a)

(b)

(c)

R2 = 0.9759

R2 = 0.98940.40

OD

pL

DH

(6

50

nm

)

0.30

0.20

0.10

0.00

0.20Gametocytes

Asex

**

**

**

0.15

0.10

OD

(6

50

nm

)O

D (

65

0 n

m)

OD

(6

50

nm

)0.05

0.00

0.50

0.40

0.30

0.20

0.10

0.00

0 00.5 0.2 0.4 0.61

% HCT % HCT

1.5

0.20

0.15

0.10

OD

(6

50

nm

)0.05

0.00

0 10000

Infected RBC number

20000 0 10000

Infected RBC number

20000

0.50

0.40

OD

pL

DH

(6

50

nm

)

0.30

0.20

0.10

0.00

0 035000 35000

gametocytes number gametocytes number

70000 70000

0.50

0.40

0.30

0.20

0.10

0.00

Gametocytes

Blank RBC

*

Figure 2. pLDH assay on gametocytes. (a) Correlation between pLDH activity and numbers of gametocytes (gametocytaemia ranged between 0.3% and3.00%) obtained by serial dilution with uninfected RBCs (fixed HCT 1%). The figure shows the mean+SD of OD values (y-axis) for different gametocytenumbers (x-axis, mean+SD) from four experiments performed with 3D7 (left) and three with NF54 (right). (b) Comparison of pLDH activity ofgametocytes and asexual parasites (Asex) (gametocytaemia ranged between 0.3% and 1.00%). The figure shows the mean OD+SD from fiveexperiments with 3D7 (left) and three with NF54 (right) (**P,0.01; *P,0.05, gametocytes versus asexual parasites, t-test). (c) pLDH activity wasmeasured in gametocyte cultures (0.4% and 2% gametocytaemia for 3D7 and NF54, respectively) serially diluted with complete medium to obtaindifferent HCT values. The figure shows a representative experiment out of three with 3D7 (left) and two with NF54 (right). The pLDH activity is reportedas OD (650 nm) and each point represents the mean+SD value of three replicates.


5 of 11

JAC



ownloaded from




with 24% Emax. Puromycin and chloroquine were completely in-active at 72 h.

In summary, we propose a dual-checkpoint pLDH assay inwhich gametocyte viability is first measured at 72 h post-treatment and, in the absence of IC50, Emax and potency are calcu-lated for each compound. After removal of the drugs from themedium and a further 72 h of incubation, a second pLDH measure-ment is performed in the same plate and the IC50 is calculated(72+72 h assay).

Mosquito infectivity

To address whether the reduced gametocyte viability measured bythe pLDH assay translated into reduced transmission, we per-formed SMFA with stage V gametocytes untreated or treatedwith epoxomicin, methylene blue or dihydroartemisinin. Theseexperiments showed that epoxomicin reduced oocyst densityand prevalence almost completely at both 100 and 1000 nM,which is consistent with its potent activity measured in the pLDHassay. On the other hand, in agreement with the pLDH assayresults, dihydroartemisinin and methylene blue at 100 nM wereless potent than epoxomicin in the SMFA (Figure 6).

DiscussionWe established an efficient and reliable screening assay for anti-gametocyte compounds based on the detection of pLDH as anindicator of gametocyte viability.

The pLDH assay has already been used for antimalarial drugscreening on asexual stages in high-throughput screening plat-forms.22 Importantly, it does not require the use of parasite trans-genic lines and thus can be applied to different parasite laboratorystrains and to field isolates. Lactate dehydrogenase is a reliablemarker of gametocyte viability since the enzyme has beenshown to be present in parasites in the blood of malaria patients.23

In fact, gametocytes can be detected using OptiMAL, a rapidimmunochromatographic assay based on the detection ofPlasmodium-specific pLDH.24

For a robust screening assay, good production rates and fullviability of gametocytes are an absolute requirement. Our resultsindicated that gametocyte maturation was efficiently and repro-ducibly obtained in 96-well plates with medium replacementevery 72 h without significantly affecting progression through thefive stages of gametocyte maturation and the ability of maturegametocytes to form gametes. Importantly, gametocytes grownin 96-well plates fed to mosquitoes were able to efficientlyundergo fertilization and sporogony in the insect vector.

Using the 96-well plate format, with 200 mL of working volumeper well, we found that the best conditions for the pLDH assay are1% HCT and 1%–2% gametocytaemia, resulting in an average Z′

factor of 0.86, which indicates that the inter-plate and intra-platevariation is low and the assay quality is high. The minimum thresh-old of gametocyte density for reliable pLDH detection is 0.5%gametocytaemia (i.e. 104 gametocytes) with an acceptable S/Bratio of 1.5. In terms of qualityand sensitivity, these results indicatethat the pLDH assay compares well with previously reportedassays.10,25 Being specific for the LDH of parasites, the assay hasthe advantage of not requiring laborious purification steps suchas those needed when the ATP of gametocytes is measured.Since ATP is present also in uninfected erythrocytes, almost100% pure gametocyte cultures are required.10,26

Of interest is the finding that, on a per cell basis, the pLDH activ-ity was higher in gametocytes than in asexual parasites. To ourknowledge this has not been reported previously. However, sincethe cultures were not synchronized, this difference may reflect astage-related variation of the enzymatic activity rather than atrue metabolic difference between gametocytes and asexualparasites. It is known that the levels of pLDH activity gradually

0.5

OD

(6

50

nm

)

0.4

0.3

0.2

0.1

0.0

0.01 0.10 1

DHA (nM)

10 100 1000 10000

24 h

48 h

72 h

Figure 3. Time kinetic of dihydroartemisinin (DHA) activity againstgametocytes. NF54 P. falciparum gametocytes were treated withincreasing concentrations of dihydroartemisinin (0.01–1000 nM) andpLDH activity (measured as OD 650 nm) was evaluated after 24, 48 or72 h. Data show the mean+SD of three independent experiments(P,0.01 24 h versus 72 h, ANOVA).

Table 1. Quality and sensitivity of the pLDH assay using gametocytes at different parasitaemia percentages

Gametocyte percentage

0.2%–0.5% 0.5%–1.0% 1.0%–3.0%

3D7 NF54 3D7 NF54 3D7 NF54

Z′ 0.60+0.21 0.29+0.17 0.59+ 0.20 0.57+ 0.12 0.86+ 0.06 0.80+0.07S/N 20.10+11.53 8.83+4.66 23.10+10.37 23.37+7.88 51.74+7.35 60.63+ 29.05S/B 1.40+0.18 1.40+0.15 1.61+0.20 1.66+0.30 2.71+0.51 2.33+0.58

Data are the mean+SD from five independent experiments on 3D7 and four experiments on NF54.


6 of 11



ownloaded from


increase from young to late trophozoites, and then decline in schi-zonts.27 However, a direct comparison of the pLDH activity of syn-chronized cultures of either asexual or sexual parasites in thedifferent stages has not been reported yet. Experiments are in pro-gress to investigate this aspect further.

The total absence of asexual parasites from the gametocytecultures is essential for the interpretation of the results of ourassay, since the pLDH activity of contaminating asexual, which is

not distinguishable from that of gametocytes, may be a confound-ing element. We found that the 48–72 h NAG treatment was ad-equate to clear asexual stages from the culture, as checked byexamining Giemsa-stained smears during gametocyte maturationin 96-well plates (,0.001% of asexual forms), and no additionalpurification steps were required.

To confirm the gametocyte pLDH assay as a novel toolfor the screening of gametocytocidal compounds, the effects of

120

100

80

60

40

20

00 100

DHA (nM)

Pe

rce

nta

ge

of

co

ntr

ol

Pe

rce

nta

ge

of

co

ntr

ol

1000

120

100

80

60

40

20

00.10 10.00 100.001.00

Epoxomicin (nM)

Pe

rce

nta

ge

of

co

ntr

ol

1000.00

120

100

80

60

40

20

01000 10000 100000

Primaquine (nM)

Pe

rce

nta

ge

of

co

ntr

ol

1000000

140

120

100

80

60

40

20

0

1.00 10.00 100.00

Methylene blue (nM)

1000.00

pLDH assay

Microscopy (healthy)

Microscopy (total)

(b)

(a)

Healthy GCT Unhealthy GCT

Figure 4. Activity of dihydroartemisinin, epoxomicin, methylene blue and primaquine against gametocytes. 3D7 P. falciparum gametocytes(gametocytaemia range 0.5%–1.5%) were plated at 1% final HCT and treated with dihydroartemisinin (DHA; 25–800 nM), epoxomicin (0.8–200 nM),methylene blue (6.3–400 nM) or primaquine (8200–13180 nM) for 72 h. (a) Concentration–response curves obtained with the pLDH assay by Giemsastaining counting only healthy gametocytes or by Giemsa staining counting total gametocytes, both healthy and unhealthy. Results are mean+SD ofthree experiments for the pLDH assay and a representative experiment out of three for Giemsa. (b) Giemsa-stained smears showing healthy andunhealthy gametocytes (magnification×1000). This figure appears in colour in the online version of JAC and in black and white in the print version of JAC.


7 of 11

JAC



ownloaded from


120

Pe

rce

nta

ge

of

co

ntr

ol

100

80

60

40

20

0

1 10 100

DHA (nM)

1000

120

Pe

rce

nta

ge

of

co

ntr

ol

100

80

60

40

20

0

100 1000 10000

Primaquine (nM)

100000

140

Pe

rce

nta

ge

of

co

ntr

ol

100

120

80

60

40

20

0

1 10

Methylene blue (nM)

100

120

Pe

rce

nta

ge

of

co

ntr

ol

100

80

60

40

20

0

0 101 100

Epoxomicin (nM)

1000

72h

72+72h

Microscopy (healthy)

Figure 5. Gametocytocidal activity of drugs measured with the modified pLDH extended protocol. Comparison of the concentration–response curves ofdihydroartemisinin (DHA; 3.1–400 nM), epoxomicin (0.8–200 nM), methylene blue (3.1–100 nM) and primaquine (500–35000 nM) obtained by the pLDHassay after 72 h or after 72+72 h incubation in drug-free medium. The microscopic results of Giemsa staining of smears after 72 h incubation are alsoshown. Results are the mean+SD of three experiments for the pLDH assay and the mean+SEM of two experiments for Giemsa.

Table 2. In vitro activity of different compounds against 3D7 P falciparumgametocytes assessed at two different incubation times with the pLDHassay

Compounds IC50 (nM), 72 ha IC50 (nM), 72+72 ha

Dihydroartemisinin .400 17.0+7.3Epoxomicin .400 3.9+0.8Methylene blue .400 29.5+3.4Puromycin .2000 201.5+93.3Primaquine .20000 12515+6541Chloroquine .2000 .2000

aIC50 was calculated by measuring the pLDH activity after 72 h ofincubation with drugs or after 72 h of incubation with drugs followed by afurther 72 h with drug-free medium.Data are the mean+SD of at least three independent experiments intriplicate.

Table 3. Maximum effect and potency of drugs after 72 h treatment

Compound Emaxa(%), 72 h Pb (nM), 72 h

Dihydroartemisinin 57.0+12.7 25.3+14.6Epoxomicin 44.7+2.7 7.2+2.0Methylene blue 24.0+7.8 82.5+29.7Puromycin NA NAPrimaquine 57.2+29.1 30300+14200Chloroquine NA NA

NA, compound not active; data are the mean+SD of at least threeindependent experiments in triplicate.aEmax is the maximal inhibition of gametocytes viability measured bypLDH assay.bP is the compound concentration that induces 50% of Emax (see Materialsand methods for details).


8 of 11



ownloaded from


dihydroartemisinin, epoxomicin, methylene blue, primaquine,puromycin and chloroquine were measured and the IC50 valueswere calculated. Dihydroartemisinin was chosen because it isthe active metabolite of artemisinins, the mainstay class of anti-malarial drugs, and is itself an antimalarial. Artemisinins areknown to be effective in vitro against stage I–III gametocytesbut not against stages IVand V.28 Methylene blue is the oldest syn-thetic antimalarial drug registered for clinical use, with a potentgametocytocidal effect in vivo and in vitro on all gametocytestages.9,29,30Primaquine is an 8-aminoquinoline, officially licensedas an antimalarial drug by the US FDA in 1952, able to clear intra-hepatic Plasmodium vivax parasites and the only commerciallyavailable drug that in vivo eliminates also mature P. falciparumgametocytes. For this reason, primaquine has been proposedas a single-dose treatment after ACT of acute malaria attacks,to shorten the infectivity period of patients and reduce

transmission.2,31 However, due to the absence of the active meta-bolites, which are generated by liver enzymes, primaquine remainspractically inactive in vitro, with IC50 .10 mM on gameto-cytes.17,26,32 Epoxomicin is a proteasome inhibitor, which hasbeen recently described as a good gametocytocidal agent, killingalso stage V gametocytes in the nanomolar range.33 However,the potencies of epoxomicin reported in the literature are quitevariable, with differences in IC50 up to 10-fold depending on thetime of incubation (24–96 h) and the assay used.19,25 Puromycin,an inhibitor of protein synthesis, was described in 1979 by Sindenand Smalley20 to have activity against sexual and asexual Plasmo-dium stages, although mature gametocytes were less susceptible.Chloroquine has been widely used as an antimalarial for itsmarked and rapid schizonticidal activity but has little or no game-tocytocidal activity in the range of doses used against asexualparasites.28

25(a) (b)

(c) (d)

Oo

cyst

s /

mo

squ

ito

Oo

cyst

s /

mo

squ

ito

20

15

10

5

0

4

3

2

1

0

0.1%

DM

SO

0.1%

DM

SO

0.1 mM

DHA

1 mM D

HA

0.1 mM

epoxo

mic

in

1 mM e

poxom

icin

0.1 mM

meth

ylene b

lue

1 mM m

ethyl

ene blu

e

100

% in

fec

ted

mo

squ

ito

es 80

60

40

20

0

0.1%

DM

SO

0.1 mM

epoxo

mic

in

1 mM e

poxom

icin

0.1 mM

meth

ylene b

lue

1 mM m

ethyl

ene blu

e

% in

fec

ted

mo

squ

ito

es

100

60

40

20

0

0.1%

DM

SO

0.1 mM

DHA

1 mM D

HA

80

Figure 6. Effect of epoxomicin, methylene blue and dihydroartemisinin (DHA) on mosquito infectivity. Gametocytes were pre-treated for 24 h with twoconcentrations (100 or 1000 nM) of epoxomicin, methylene blue or dihydroartemisinin. Fresh drugs were added to gametocytes before themosquitoes fed. The histograms show the mean+SEM from duplicate measurements of oocyst number from 20 mosquitoes per sample. Oocystdensity (a and c) and prevalence (b and d) are shown.


9 of 11

JAC



ownloaded from


In our hands and in agreement with previous reports, dihydroar-temisinin, epoxomicin, methylene blue and primaquine exerted aconcentration-dependent toxicity on gametocytes measured bymicroscopy after the 72 h assay. However, a discrepancybetween the pLDH readouts and the gametocyte microscopiccounts, particularly at high drug concentrations, was observed.This result was interpreted as due to residual pLDH activity asso-ciated with morphologically abnormal, probably dead gameto-cytes. Similar unhealthy forms can be observed in theGiemsa-stained pictures of epoxomicin-treated gametocytes pub-lished by Czesny et al.19

To determine whether the morphologically unhealthy gameto-cytes were committed to die or already dead or were able torecover from the drug treatment, the incubation was prolongedfor an additional 72 h in drug-free medium. The pLDH activity mea-sured after 72+72 h dropped to zero with all four compounds, in-dicating that the aberrant gametocytes did not recover.

By measuring the pLDH activity after 72+72 h of incubation, itwas possible to calculate the IC50 values for methylene blue, epox-omicin, dihydroartemisinin, puromycin and primaquine, whichwere comparable to those obtained by other methods. In particu-lar, we confirmed that epoxomicin is the most active anti-gametocyte compound, with an IC50 in the low nanomolar range[from 0.42 nM (Lelievre et al.26) to 3.5 nM (Tanaka and William-son25) and 3.7 nM (ourselves)]. Similarly, and in accordance withthe literature, primaquine had a detectable effect onlyat micromo-lar concentrations (.5 mM).26,34 The IC50 of methylene blue in thepLDH test (IC50 29.5 nM) was similar to that reported using trans-genic parasite lines expressing GFP–luciferase.9 However, methy-lene blue was several times more active in our assay than in theluminescence ATP assay described by Lelievre et al. (IC50

490 nM).26 A possible explanation for this discrepancy could bethe prolonged length of the pLDH assay compared with the 48 hATP test. Due to the stage-dependent effect of dihydroartemisinin,it is not straightforward to compare the IC50 obtained with thepLDH assaywith the data in the literature. It is recognized thatdihy-droartemisinin has low activity in vitro against stage IV–Vgameto-cytes (IC50 3.56 mM in Lelievre et al.26 and 25%–50% inhibition ofstage V gametocyte viability at 120 nM in Adjalley et al.9).However, significant inhibition of the viability of stage I–II gameto-cytes at 12 nM has also been reported.9 This is consistent with ourresults, in which dihydroartemisinin was used primarily on young,stage II–III gametocytes at the onset of the experiments, andthe resulting IC50 was 17 nM. Puromycin has not been used in therecently published gametocyte screening assays. The IC50 of201.5 nM in the pLDH assay is consistent with the results reportedby Sinden and Smalley20 in 1979, which showed an effect on stageIV–V gametocytes only at high concentrations (≥580 nM), buttotal inhibition of viability of young stages already at 120 nM.Finally, chloroquine was not effective at the concentrations usedin this study (IC50 .2000 nM) and this is consistent with data byBenoit-Vical et al.21 (IC50 300 nM on stage II–III and 10700 nMon stage IV–V).

Importantly, the results of the SMFA confirmed that the72+72 h pLDH measurements on gametocytes treated withmethylene blue, dihydroartemisinin or epoxomicin truly reflectedthe fact that these parasites were dead. Both the prevalenceand the intensity of mosquito infection with drug-treated game-tocytes were dramatically reduced compared with untreatedcontrol gametocytes, and in fact were abolished at the highest

concentrations. In the SMFA, gametocytes were pre-exposed todrug for 24 h prior to mosquito feeding. The strong reduction intransmission indicates that the gametocytes lose viability quiterapidly after the initial drug exposure. This was confirmed in the72+72 h pLDH assay, which showed that gametocytes do notrecover after the initial drug exposure.

The above results indicate that the determination of gameto-cyte pLDH activity at both the 72 h and the 72+72 h timepointsmay provide information on how compounds interfere with themetabolism of gametocytes. Importantly, gametocytocidal com-pounds for which an IC50/IC90 can be obtained at the first timepointwill be readily identified as fast killers and prioritized, while thosewith a slower killing effect will be identified by the pLDH measure-ments at the second timepoint. The pLDH assay may also help toidentify compounds with a ‘static’ effect, in which pLDH activitymay resume during the prolonged incubation. Finally, the doublecheckpoint does not impose a significant burden on the gameto-cyte pLDH assay as the second timepoint is performed using ma-terial from the same plate.

In conclusion, our work shows that the gametocyte pLDH assayis suitable to screen the anti-gametocyte activity of compound li-braries in order to identify new gametocytocidal drugs. Ourresults indicate that the assay is reliable, cheap and highly reprodu-cible, and represents a novel, much-needed addition to improvethe current tools to identify agents able to inhibit malaria transmis-sion.

AcknowledgementsWe thank Marga van der Vegte-Bolmer and Geert-Jan van Gemert for theircontribution to the mosquito infectivity assay; Helmi E. Pett for discussion;Paola Verducci and Tiziana Bianchi from AVIS Comunale Milano forcollecting blood samples for parasite culture; Prof. Richard K. Haynes forproviding dihydroartemisinin; and Paola Misiano for helpful discussionand critical reviewing of the manuscript. Part of this work was presentedat the meeting ‘Challenges in Malaria Research: Progress towardselimination’ held in Basel on 10–12 October 2012.

FundingThis work was funded by grant no. OPP1040394 of the Global HealthProgram of the ‘Bill & Melinda Gates Foundation’.

Transparency declarationsNone to declare.

Supplementary dataTables S1 to S3 are available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/).

References1 World Health Organization. World Malaria Report 2011. ISBN:9789241564403. http://www.who.int/malaria/publications/atoz/9789241564403/en/index.html (04 April 2013, date last accessed).

2 World Health Organization. 2010. Guidelines for the Treatment of Malaria(Second Edition). ISBN: 9789241547925. http://www.who.int/malaria/


10 of 11



ownloaded from




http://www.who.int/malaria/publications/atoz/9789241564403/en/index.html








publications/atoz/9789241547925/en/index.html (04 April 2013, date lastaccessed).

3 Roll Back Malaria Partnership 2008. The Global Malaria Action Plan—For aMalaria Free World. http://www.rbm.who.int/gmap/gmap.pdf (04 April2013, date last accessed).

4 Roll Back Malaria Partnership 2011. Refined/Updated GMAP Objectives,Targets, Milestones and Priorities Beyond 2011. http://www.rbm.who.int/gmap/gmap2011update.pdf (04 April 2013, date last accessed).

5 Alonso PL, Brown G, Arevalo-Herrera M et al. A research agenda tounderpin malaria eradication. PLoS Med 2011; 8: e1000406.

6 White NJ. The role of anti-malarial drugs in eliminating malaria. Malar J2008; 7 Suppl 1: S8.

7 Roncales M, Vidal-Mas J, Leroy D et al. Comparison and optimization ofdifferent methods for the in vitro production of Plasmodium falciparumgametocytes. J Parasitol Res 2012; 2012: 927148.

8 Lucantoni L, Avery V. Whole-cell in vitro screening for gametocytocidalcompounds. Future Med Chem 2012; 4: 2337–60.

9 Adjalley SH, Johnston GL, Li T et al. Quantitative assessment ofPlasmodium falciparum sexual development reveals potenttransmission-blocking activity by methylene blue. Proc Natl Acad Sci USA2011; 108: E1214–23.

10 Peatey CL, Spicer TP, Hodder PS et al. A high-throughput assay for theidentification of drugs against late-stage Plasmodium falciparumgametocytes. Mol Biochem Parasitol 2011; 180: 127–31.

11 Buchholz K, Burke TA, Williamson KC et al. A high-throughput screentargeting malaria transmission stages opens new avenues for drugdevelopment. J Infect Dis 2011; 203: 1445–53.

12 Trager W, Jensen JB. Human malaria parasites in continuous culture.1976. J Parasitol 2005; 91: 484–6.

13 Kariuki MM, Kiaira JK, Mulaa FK et al. Plasmodium falciparum: purificationof the various gametocyte developmental stages from in vitro-cultivatedparasites. Am J Trop Med Hyg 1998; 59: 505–8.

14 Garcia GE, Wirtz RA, Barr JR et al. Xanthurenic acid induces gametogenesisin Plasmodium, the malaria parasite. J Biol Chem 1998; 273: 12003–5.

15 Makler MT, Hinrichs DJ. Measurement of the lactate dehydrogenaseactivity of Plasmodium falciparum as an assessment of parasitemia. Am JTrop Med Hyg 1993; 48: 205–10.

16 Zhang JH, Chung TD, Oldenburg KR. A simple statistical parameter foruse in evaluation and validation of high throughput screening assays.J Biomol Screen 1999; 4: 67–73.

17 van Pelt-Koops JC, Pett HE, Graumans W et al. The spiroindolone drugcandidate NITD609 potently inhibits gametocytogenesis and blocksPlasmodium falciparum transmission to anopheles mosquito vector.Antimicrob Agents Chemother 2012; 56: 3544–8.

18 Pukrittayakamee S, Chotivanich K, Chantra A et al. Activities ofartesunate and primaquine against asexual- and sexual-stage parasitesin falciparum malaria. Antimicrob Agents Chemother 2004; 48: 1329–34.

19 Czesny B, Goshu S, Cook JL et al. The proteasome inhibitor epoxomicinhas potent Plasmodium falciparum gametocytocidal activity. AntimicrobAgents Chemother 2009; 53: 4080–5.

20 Sinden RE, Smalley ME. Gametocytogenesis of Plasmodium falciparumin vitro: the cell-cycle. Parasitology 1979; 79: 277–96.

21 Benoit-Vical F, Lelievre J, Berry A et al. Trioxaquines are new antimalarialagents active on all erythrocytic forms, including gametocytes. AntimicrobAgents Chemother 2007; 51: 1463–72.

22 Gamo FJ, Sanz LM, Vidal J et al. Thousandsof chemical starting points forantimalarial lead identification. Nature 2010; 465: 305–10.

23 Oduola AM, Omitowoju GO, Sowunmi A et al. Plasmodiumfalciparum: evaluation of lactate dehydrogenase in monitoringtherapeutic responses to standard antimalarial drugs in Nigeria. ExpParasitol 1997; 87: 283–9.

24 Mueller I, Betuela I, Ginny M et al. The sensitivity of the OptiMAL rapiddiagnostic test to the presence of Plasmodium falciparum gametocytescompromises its ability to monitor treatment outcomes in an area ofPapua New Guinea in which malaria is endemic. J Clin Microbiol 2007; 45:627–30.

25 Tanaka TQ, Williamson KC. A malaria gametocytocidal assay usingoxidoreduction indicator, alamarBlue. Mol Biochem Parasitol 2011; 177:160–3.

26 Lelievre J, Almela MJ, Lozano S et al. Activity of clinically relevantantimalarial drugs on Plasmodium falciparum mature gametocytes in anATP bioluminescence ‘transmission blocking’ assay. PLoS One 2012; 7:e35019.

27 Vivas L, Easton A, Kendrick H et al. Plasmodium falciparum: stage specificeffects of a selective inhibitor of lactate dehydrogenase. Exp Parasitol 2005;111: 105–14.

28 Dechy-Cabaret O, Benoit-Vical F. Effects of antimalarial molecules onthe gametocyte stage of Plasmodium falciparum: the debate. J MedChem 2012; 55: 10328–44.

29 Zoungrana A, Coulibaly B, Sie A et al. Safety and efficacy of methyleneblue combined with artesunate or amodiaquine for uncomplicatedfalciparum malaria: a randomized controlled trial from Burkina Faso. PLoSOne 2008; 3: e1630.

30 Coulibaly B, Zoungrana A, Mockenhaupt FP et al. Stronggametocytocidal effect of methylene blue-based combination therapyagainst falciparum malaria: a randomised controlled trial. PLoS One 2009;4: e5318.

31 Sutanto I, Suprijanto S, Kosasih A et al. The effect of primaquine ongametocyte development and clearance in the treatment ofuncomplicated falciparum malaria with dihydroartemisinin-piperaquinein South Sumatra, Western Indonesia: an open label randomizedcontrolled trial. Clin Infect Dis 2012; 56: 685–93.

32 Vale N, Moreira R, Gomes P. Primaquine revisited six decades after itsdiscovery. Eur J Med Chem 2009; 44: 937–53.

33 Kiszewski AE. Blocking Plasmodium falciparum malaria transmissionwith drugs: the gametocytocidal and sporontocidal properties of currentand prospective antimalarials. Pharmaceuticals 2010; 4: 44–68.

34 Chevalley S, Coste A, Lopez A et al. Flow cytometry for the evaluation ofanti-plasmodial activity of drugs on Plasmodium falciparum gametocytes.Malar J 2010; 9: 49.


11 of 11

JAC



ownloaded from


http://www.rbm.who.int/gmap/gmap.pdf



http://www.rbm.who.int/gmap/gmap2011update.pdf





A Plasmodium falciparum screening assay for anti-gametocyte drugs based on parasite lactate dehydrogenase detection

Documents