Resistance to artemisinin of malaria parasites
(Plasmodium falciparum) infecting
alpha-thalassemic erythrocytes in vitro. Competition
in drug accumulation with uninfected
erythrocytes.
S Kamchonwongpaisan, … , M A Avery, Y Yuthavong
J Clin Invest.
1994;
93(2)
:467-473.
https://doi.org/10.1172/JCI116994
.
Plasmodium falciparum infecting hemoglobin (Hb)H and/or Hb Constant Spring
erythrocytes has higher resistance to artemisinin in vitro than when infecting normal
erythrocytes. This is due to low drug accumulation of infected erythrocytes resulting from
competition with uninfected variant erythrocytes, which have a higher accumulation capacity
than genetically normal cells. Drug accumulation of the parasite was shown to be saturable
and dependent on metabolic energy. The 50% inhibitory concentrations (IC50's) for the
parasite in HbH/Hb Constant Spring erythrocytes were decreased when normal
erythrocytes were added to the infected cells, and correspondingly, the IC50's in normal
erythrocytes were increased when HbH/Hb Constant Spring erythrocytes were added to the
infected cells. The changes of IC50 corresponded to the variation in drug accumulation of
mixtures of normal and variant erythrocytes of different compositions. The IC50's for the
parasite in variant erythrocytes were also greatly decreased when the hematocrit of the
culture was lowered, while the IC50's in normal erythrocytes were independent of the
hematocrit. The increase in IC50 values for the parasites infecting variant erythrocytes was
also related to the decrease in parasite accumulation, indicating that drug accumulation
capacity of the parasite also has a role in determining drug sensitivity. Artemisinin sensitivity
therefore is determined by its accessibility to the parasite, which is decreased in infected
variant erythrocytes.
Research Article
Find the latest version:
Resistance
to
Artemisinin of Malaria
Parasites
(Plasmodium
falciparum)
Infecting
a-Thalassemic
Erythrocytes
In
Vitro
Competition inDrug Accumulation with Uninfected Erythrocytes
SumaleeKamchonwongpaisan,*GanigarChandra-ngam,* Mitchell A.Avery,tandYongyuth Yuthavong *
*Department of Biochemistry, Faculty of Science, Mahidol University, Bangkok 10400, Thailand, and Departmentof Chemistry, University ofNorthDakota, GrandForks,North Dakota 58202
Abstract
Plasmodium
falciparum infecting hemoglobin (Hb)H
and/or HbConstantSpring erythrocytes hashigher
resistancetoarte-misinin in vitrothan wheninfecting normal erythrocytes. This is dueto lowdrugaccumulation of infected erythrocytes result-ing from competition with uninfected variant erythrocytes, which have a higher accumulation capacity than genetically
normal cells.Drug accumulation oftheparasitewasshown to
be saturable anddependentonmetabolicenergy. The50% inhib-itoryconcentrations
(IC5,'s)
fortheparasiteinHbH/Hb
Con-stantSpring erythrocytesweredecreasedwhen normal
erythro-cytes were added to theinfected cells,and
correspondingly,
theIC5R's
innormalerythrocytes wereincreased whenHbH/Hb ConstantSpring
erythrocytes wereaddedtotheinfected cells.The changesof
IC50
correspondedtothevariationindrugaccu-mulationofmixtures of normal and variant erythrocytes of dif-ferent compositions.The
ICFo's
for theparasiteinvarianteryth-rocytes werealsogreatly decreased when the hematocrit of the culture was lowered, while the
IC50's
in normal erythrocyteswereindependentofthehematocrit. The increase in
IC50
valuesfor the parasites infecting variant erythrocyteswasalso related tothe decrease inparasite accumulation, indicatingthatdrug accumulation capacity oftheparasite also hasarolein deter-mining drug sensitivity. Artemisinin sensitivity thereforeis
de-termined by its accessibilitytotheparasite, which is decreased
ininfectedvarianterythrocytes.(J. Clin.Invest.
1994.93:467-473.) Key words: malaria*Plasmodiumfalciparum*
a-thalas-semia*variant erythrocytes-artemisinin
Introduction
Resistance of the malaria parasite Plasmodium falciparumto
chloroquine
and otherdrugs invariouspartsofthe worldne-cessitates the development ofneweffective antimalarials
(1).
Derivatives
of artemisinin (qinghaosu)(2, 3)comprise a newfamily
of such drugs presently undergoing development andclinical trials. Their fast actionandeffectivenessagainst
chloro-quine-resistant
strains oftheparasites make them potentiallyvaluable drugs, especially in such areas of high resistance as
Address correspondence to Dr. Y. Yuthavong, Department of Bio-chemistry, Faculty of Science, Mahidol University, Rama 6 Road,
Bangkok 10400, Thailand.
Receivedforpublication17December1991 and inrevisedform 30
June1993.
Southeast Asia and the Pacific. Theseareasalso have ahigh incidence ofthalassemia and abnormal hemoglobins (4),some
genotypes for which are associated with innate resistance to
falciparum malaria (5, 6). Hence, hemoglobin
(Hb)'H-con-taining erythrocytes from a-thalassemic (genotype
a-thalasse-mia 1 /a-thalassemia 2,
--/
-a)ora-thalassemic/HbConstantSpring individuals (genotype a-thalassemia 1/Hb Constant Spring,
--/acsa)
have been shown to be resistant to infection by P. falciparum in vitro (7, 8). Enhanced immunerecognitionand clearance of parasitized erythrocytes have also been impli-cated ascontributingtothe apparentprotectiveeffect ofthalas-semia against falciparum malaria (9). Theseexperimental re-sults support an earlier observation that prevalence ofthe genes for a-thalassemia parallels the endemicity of falciparum
ma-laria in the Pacific Islands(5).Thehighfrequenciesof genes for various types of thalassemia and abnormalhemoglobinsin Southeast Asia and various other parts of the world make it
relevant to ask whether antimalarial drugs would have the same
efficacy
in malariapatients carrying
these genes as ingeneticallynormalpatients. This question also carries implica-tions fordevelopment of drugresistance, whichoriginated in
SoutheastAsia for a number ofantimalarials, including chloro-quine ( 1). If the drug concentrations that are sufficient to kill theparasitesinnormalpatientsweresuboptimalforgenetically variantpatients, these would impose selectivepressure on the
parasitestobecomedrugresistant.
Wehavereportedearlierthat P.falciparum infectingHbH
and/orHb Constant Spring erythrocytes displaysmore resis-tance to artesunate, aderivative of
artemisinin,
andchloro-quinethan the sameparasitestraininfecting genetically nor-malerythrocytes (10). This novel hosteffectonsensitivityto artesunate isespecially strikingin view of the facts thatthese
genetically variant erythrocytesarealreadysubjecttohigh
oxi-dative stress(11), thatP.
falciparum
issensitive tooxidativestress( 12-14),andthatartemisinin probablyexertsits action
through
an oxidative mode ( 15-17). Two possible explana-tionsofthe observationwereraised:thehigherlevels ofantioxi-dativeenzymes inthevariant erythrocytes could give riseto acompensation
effect lessening the oxidativedamagepotentialofartemisinin, and/or the infected varianterythrocytescould not accumulate as much drug as infected normal erythrocytes. In thispaper, we show that the latter explanation is correct.
Furthermore,we show that the reduction in artemisinin
accu-mulationof infectedvariant erythrocytes is due partly to
com-petition with uninfectederythrocytes forthe drug, and partly to
lowerdrugaccumulation capacities of the parasite.
1. Abbreviations used in thispaper:Hb, hemoglobin;
IC50,
50% inhibi-tory concentration.ArtemisininResistance ofMalariaParasites ina-ThalassemicErythrocytes 467
J.Clin. Invest.
©TheAmerican Society for Clinical Investigation, Inc.
0021-9738/94/02/0467/07 $2.00
Methods
CultureofP.falciparum.Chloroquine-resistantK,strainwasobtained
from Kanchanaburi Province, Thailand,andmaintained in vitro by the candle jar technique ofTrager and Jensen ( 18). The cultureswere
synchronizedatringstageby5%sorbitoltreatment(19) and
concen-tratedatschizontstage byPercoll layering technique(20)toobtain 95%parasitemia.The concentratedschizontsweresubculturedwith
normal and variant erythrocytes foratleast 72 hbeforeantimalarial
testingandfor 90 hfordrugaccumulation study,atwhichtimethe synchronywasstilllargelymaintained.Theculturemedium usedwas
RPMI 1640 supplemented with 25 mM Hepes, 0.2% NaHCO3, 40
Ag/ml gentamicin,
and10% humanserum(bloodgroupAB), pH7.4.50%inhibitoryconcentration(IC50)testing. Theantimalarial activ-ity of artemisinin againstP.falciparum-infectednormal and thalasse-micerythrocyteswasmeasuredbyusingthe [3H]hypoxanthine
incor-poration methodofDesjardins (21).Briefly, artemisininwasdissolved
inDMSOanddilutedwith culture mediumtotheappropriate
concen-tration. Thefinal concentration ofDMSOwas0.001%, which hadno
effectonparasitegrowth.25Alof mediumcontaining drug and 200
Al
of 1.5% cell suspension (or asindicated) with 1-2% parasitemiaat
immature schizont or ring stage (or as indicated) were cultured in
triplicatedmannerfor 24 h. 25 Mlof 0.5 MCi [3H]hypoxanthinewas
added.Afteranadditional18 hofculture,the cellswereharvestedonto
glassfiberfilters. Theradioactivitywasmeasured byaliquid
scintilla-tion counter (LS-1 801; Beckman Instrs., Inc., Fullerton, CA). The
IC50's
(thedrugconcentrations requiredfor 50% reduction of theradio-activity as compared with control withoutthe drug) ofartemisinin
against theseinfectedcellswereobtainedfromdose-responsecurves.
Insomeexperiments,aconstantamountofP.falciparum-infected
erythrocytes (ring stage) wasdiluted with a known composition of
uninfectednormaland variant erythrocytes, with the final hematocrit of 1.5%. In other experiments, the infected erythrocytes (immature schizontstage)werediluted with variousamountsof normal and
vari-antcellstoobtainasetof cells with various hematocrit values. These
cellsweresimilarlytestedforthe
IC50-["'C]Artemisinin accumulation. After a 90-h culture, P.
falci-parum-infected (mostly at late trophozoite and immature schizont
stage) and controluninfectederythrocytes(both normalandvariant)
werewashedwiththemedium withoutserum supplemented.
Dupli-catedsamplesof 70-Mlpackedcellswereincubated with 630 Mlof 130
nM
15-["'Clartemisinin
(dissolved in 0.001% toluene and 0.1% DMSOinincomplete medium without serum),oratvariousconcen-trations for study of concentrationeffect,at37°Cfor2 h inashaking
waterbath.The cellswerepelleted. 600,lofsupernatantwasbleached
with 400,lof 15% hydrogenperoxideat60°C.4 mlof liquid scintilla-tion fluid wasadded and the radioactivity wasmeasured. Thedrug
accumulated by thecells, definedasdrugaccumulationcapacity,was
calculated and expressedastheamountofdrugin70-Ml pelletperunit
concentration of the drug remainingin thesupernatant(nmol/MM).
Insomeexperiments, drug accumulation byconcentrated infected
erythrocyteswas measureddirectly. Concentrated infected cellswith parasites in latetrophozoite-earlyschizontstagewereprepared.A
syn-chronized culture with 30-40% parasitemiawasput in the medium
supplemented with 2.75,g/mlhypoxanthine.The culturewasthen left understandard incubationcondition for 40h,and theparasitized cells,
mostly at late trophozoite-early schizont stage were harvested and washedwith incompletemedium. 15% cell suspensioninincomplete mediumwasoverlaidon anequal volume of 60% Percolland
centri-fugedat600gfor 20 min. Thedark bandatthe interface of thetwo
media,containing infected cellsat - 95% parasitemia, wascollected
and washedthrice with incomplete medium. Duplicated samplesof 10-Ml packedinfected cell pelletswereused fordeterminingthedrug accumulation asabove. In additionto themeasurement in
superna-tant, thedrug accumulation in the cell pellets was also determined
directly.The pelletswerewashedthrice withincomplete medium, in-cubatedwith 700 Mlof2%SDSsolutionat60'Cfor1h,thenbleached
with 400 ,Al of 15% hydrogen peroxide at 60'C overnight. 4 ml of liquid
scintillation fluidwasaddedand theradioactivitywasdetermined. Inthe study ofeffect of iodoacetamideon drug uptake, the cells were preincubated at370Cwith 500,Ad of100MMiodoacetamide in incomplete medium. After 1 h of incubation, 130 Ml of
['4C]-artemisinin was added to the cell suspension to obtain a final
concen-tration of 100nM and a10% cellsuspension.The cells weremixed and incubated further for 2 h, and the drug accumulation was determined asabove.
Statistical analysis.TheMann-WhitneyUtestwasused for com-paringthedatafromnormalandvariant erythrocytes based on inde-pendentrandom samples.
Results
Ashasbeen observed before, the IC50's forartemisininagainst P.falciparum infecting HbH and HbH/Hb Constant Spring erythrocytes are significantlyhigherthanagainsttheparasitein genetically normal erythrocytes (Fig. 1 ). The basis for the dif-ferences in IC50's wasearlier ( 10) suggested asbeing dueto
differencesinoxidativestresses intheinfected variantand
nor-malerythrocytes, ordueto
differences
in drug accumulation.The latterpossibility wasexploredthrough theuseof
radiola-beledartemisinin. Fig.2A shows theaccumulation of [
"C]-artemisininatthe initialconcentration of 130nMby infected
genetically normal erythrocytes (6.6% parasitemia), infected HbHerythrocytes(6.6%parasitemia), and HbH/ Hb Constant
Spring
erythrocytes(5.3%
parasitemia),
whileFig.
2Bshowsthedrugaccumulation bycontrol erythrocytes thatwere not
exposedtotheparasite. Sincethe final medium concentration
wasdecreasedtodifferentextentsbythe
different
amountsof
thedrugaccumulated into thecells,the amounts of
accumula-tionweredivided bythe final mediumconcentrationinorder
to obtain normalizedaccumulation, whichcanbe definedas
accumulationcapacities.Fig. 2,DandE,show the accumula-tion per unit medium concentraaccumula-tion ofwhole infected and
con-trolerythrocytes, andFig. 2,CandF, show the difference be-tween the two for genetically normal, HbH, and HbH/Hb
ConstantSpringerythrocytes. The values in Fig.
2,
C andF,canbe
regarded, respectively,
astheaccumulationandaccu-mulation capacities of the
parasite infecting
the threedifferenti
0
251
20
15
10-5
0
N HbH Hb H/HbCS
p<0.05 p<O.05
Figure 1. TheIC50values with means±SD for artemisinin
against
P.falciparum infecting normal(N), HbH, and
HbH/Hb
ConstantSpring(H/CS)erythrocytes.
(NS) (NS) (p-.004) (p-.001) (p-.004) (p-.001)
1.4
2
w 1.2 0 E
.? 1
Cu
0.
c 0.8
.2
E 0.6
m
8:
c 0.4 E
0
N HbHHbH/CS N HbH HbH/CS (NS) (NS) (p-.004)(p-.001)
N HbH HbH/CS
(p-.004) (p-.001)
Figure 2. The['4C]artemisininaccumulation (A-C) and
accumula-tioncapacity (D-F), with means±SD, by infected erythrocyteswith
6%parasitemia (AandD),uninfectedcontrolerythrocytes (Band
E),and calculated datafor theparasite(C and F). 70 Mul of packed
cellswasincubated with 630Alof medium with initial artemisinin
concentrationof 130 nMfor2 hbefore accumulationmeasurements.
Statisticalanalysis using Mann-Whitney Utestwasshown.N,
nor-mal; HbH/CS, HbH/Hb Constant Spring; NS,notsignificantly dif-ferent.
genetictypesof erythrocytes. While the accumulation and
ac-cumulationcapacities of whole infected(nonparasitizedplus parasitized) cellswerecomparable for the three genetictypes (Fig. 2, A and D), theaccumulationand accumulation capaci-ties for uninfected HbH and, to a greater extent, HbH/Hb ConstantSpring erythrocytesweremuch higher than those of
uninfected normal erythrocytes withvery little accumulation
(Fig. 2, B and E). Bycontrast,theaccumulationand
accumu-lation capacity for the parasite infecting HbH/Hb Constant Spring erythrocytesweresmaller thanthatinfectingHbH
eryth-rocytes,inturnsmaller than that infecting genetically normal
erythrocytes (Fig. 2, C and F).
The accumulation capacities of infected erythrocytes
de-creasedrapidly with increasing initial concentration ofthe drug
upto - 2 MuM for all three genetictypes,indicating saturability
of thebindingsites (Fig. 3). Incontrast,theaccumulation
ca-pacity for uninfected, genetically normal erythrocytes
re-mainedconstantoverthe concentrationrange,whilethe
accu-mulation capacities for uninfected HbH and HbH/Hb
Con-stant
Spring gradually declined
withincreasing
initialdrug
concentration.
The accumulationcapacity
of theparasitesin-fecting
genetically
normalerythrocytes,
as obtainedby
sub-tractingthoseofuninfected from infectederythrocytes (Fig.
3 A), wasabouttwoand three times that of theparasites
infect-ing HbH(Fig.
3 B) andHbH/Hb
ConstantSpring
erythro-cytes
(Fig.
3C), respectively,
attherapeutic
concentrations(<100 nM),
indicating differences
inavailability
ofspecific
sites.
Theaccumulation capacities of
theparasites infecting
three
different
genetic
typesoferythrocytes
declinedtoapproxi-mately thesamelevelat
concentrations
> 2,gM,
indicating
the predominance ofnonspecific binding
atthesehigh
concentra-tions.
Themeasurementof the drug
accumulation
capacity
of theparasites from
thedifference
inaccumulation of infected
eryth-rocytes(6-10%
parasitemia)
anduninfected
erythrocytes
wasvalidatedbydirect measurement
of
drugaccumulation
bycon-centrated
infected
cells(95%
parasitemia).
Fig.
3Dshows the resultsof
suchmeasurement on asuspension of concentrated
infected
genetically normal and HbH/Hb ConstantSpring
erythrocytes, whichare
comparable
tothe resultsobtained by
the
subtraction
method. These resultsindicate further
that theparasites
inparasitized
cellsaccount for themajority
ofdrug
accumulation.
Metabolic
energyrequirement
for thedrug
accumulationof
these cellswasalsostudied
byexamining
the effect of temper-ature orof additionof iodoacetamide.
Fig.
4 Ashows that the decrease inincubation
temperaturemarkedly
decreased theaccumulation
capacity
of both infected normal and infected HbH erythrocytes.Iodoacetamide
hasapartial effect
ondrug
accumulation of
theparasites,
but notofuninfected
erythro-cytes,for both
genetically
normal and HbHerythrocytes (Fig.
4B).
When the data from drug
sensitivity
andaccumulation
ex-periments,
whichwereobtained from
thesamesetof samples,
were combined to explore further the
basis
ofdifferences
in drugsensitivities,
an inverserelationship
wasfound
between the final mediumconcentration of
thedrug incubated
withuninfected erythrocytes
ofdifferent
genetic
typesand theIC50
values for parasites infecting these erythrocytes (Fig. 5 A). This indicates that the increase in observed IC50 values of
parasites
infecting variant erythrocyteswasatleastpartly duetolowered
medium concentration.
Fig. 5 B shows that the IC50 valueswere correlatedwiththeaccumulation capacities of uninfected
erythrocytes,
whichcanaccountfor
the loweredmedium
con-centration
andincreasedIC50
valuesfor infected varianteryth-rocytes. In agreement
with
theresults inFig.
2, theaccumula-tion
capacity of
theparasite,
corrected for loweredmedium
concentration,
wasalsoinfluenced by
thegenetic
type of the erythrocyte that it has infected.Fig.
5C showsaninversecorre-lation
between theIC50
values andnormalized parasite
drug accumulationcapacities, indicating
that theformer is also in-fluenced bythecapacities fordrugaccumulation oftheparasite
insidethevariantcells.
Toprovefurtherthat thedifferences inIC50 valuesfor
para-sitesinfecting various genetictypesoferythrocytes were due to
differentdrugaccessibilities, andtoassess therelative
impor-tanceof
competition
by uninfected cellsandparasitedrugac-cumulation capacities, ring-stage-infected erythrocytes were culturedinthe presence of thedrugtogether with mixtures of
uninfected genetically normaland
variant
cellsofknowncom-positions for 40h.TheIC50values, obtained before theparasite
ArtemisininResistanceofMalaria Parasites in a-Thalassemic Erythrocytes 469
D E F
I.
A
0
_- -u--__ .>_\
0
10-7 10,6
Initial concentration (M)
C
0.
0
---a--0
I .. .. I...1
0.8
i10-7
10-6Initialconcentration (M)
1.4
-i=.
'i
1.2-0
E 1 0. 1
cl
c 0.8
0
0
:3
E0.6
-a) 0 w
0
*a 0.4
-8)
1~0.2
-6
1.4
-a)
1.2-0
E
CL D
1-:
0.8-.2
co
-7 E
0.6-:'
.r-E 0.4
E
0I
o
0.2-10-5
B
*
-a--A
m
---m~~
-,a
D
A
.-- I.I'l ...,7
10.8 10-7 10-6
Initialconcentration (M)
Figure3. Effect ofconcentrationonthe [ '4C]artemisinin accumulation capacity of infected normal (A), HbH(B), HbH/Hb Constant Spring
(C) erythrocytes,and concentratedinfected normaland HbH/Hb Constant Spring (D) erythrocytes, with 9.8, 6.0, 8.9, and 95% parasitemia, respectively (.), uninfected control erythrocytes (i), and calculated data for the parasites(n).70,l (A-C)or 10 Ml(D)ofpacked cellswere
incubated with 630
Al
(A-C)or600AI
(D) of medium with varying artemisinin concentrationsat370Cfor 2hbeforemeasurements.(m,*,and n)Points obtained from concentrationdifferencesinthesupernatants(. inD, those forparasitized normal erythrocytes); (o and a)
ob-tainedfrom directmeasurementinpelletof 95%parasitized normal and HbH/Hb Constant Spring erythrocytes, respectively.
could begina newcycle, should be independent of the
compo-sition of uninfected cells with regard totheir genetic types if theseweredetermined only by the accumulation capacities of
theinfected cells, but would be dependenton the uninfected
cell composition if the accumulation capacitiesarewhollyor
partly determined by competition between infected and unin-fected cells. Data in Fig. 6 show that the IC50 values for both infected genetically normal and HbH/Hb Constant Spring erythrocytes indeed dependonthe composition of uninfected cellswithregardtotheirgenetictypes,sothatthe values of IC50
for both genetictypesconverge asthecompositions of the
un-infected cells approachoneanother. The variation inIC50
val-uesoftheparasiteinfecting bothtypesof cellswasreflected by
the variation in accumulation capacities of uninfected erythro-cytes with different compositions of normal and HbH/Hb Constant Springgenotypes(Fig. 7). It cantherefore be
con-cluded that the IC50 valuesaretoalargeextentinfluenced by drug accumulation of uninfected erythrocytes, thereby lower-ing the medium concentration. This conclusion is also
sup-portedby anothersetofexperiments determining theIC50
val-uesfor infected genetically normal and variant erythrocytesat
differenthematocritvalues in the culture. AsFig. 8 shows,the
IC50 values for infected HbH and HbH/Hb Constant Spring
erythrocytesweredecreasedwithdecreasing hematocrit values, whereas the
IC50
values forinfected genetically normalerythro-cyteswereindependent of the hematocrit. These experiments
show thatthemajorfactorleadingtoapparentdrugresistance
of theparasite infectingthe varianterythrocytesiscompetition indrug accumulation by uninfected variant erythrocytes.
Discussion
The resistancetoartemisinin of P.falciparum infectingHbH
andHbH/HbConstantSpring erythrocytes (Fig. 1) supports
ourpreviousfindingon similarlyincreased resistanceto arte-sunate (10). In this report, wehave shownfurther that such increased resistance ismainlyduetothe fact that uninfected HbH and HbH/Hb Constant Spring erythrocytes have high accumulationcapacity for artemisinin(Figs.2,B andD,3,and 5,AandB),hence, competingforthedrugwithinfected eryth-rocytes.Therefhre,ahigherinitial concentration of thedrugis required so that the infected erythrocytes can accumulate enough drugto achieve the lethal concentration. Uninfected geneticallynormalerythrocytes,onthe otherhand, havelow drugaccumulationcapacity,anddidnotsignificantly compete
for thedrugwith infectednormal erythrocytes. Insupportof
this explanation, itwas found(Fig. 6)thatadecrease in the
470 S.Kamchonwongpaisan, G.Chandra-ngam, M.A.Avery, andY. Yuthavong
1.4
i5
1.2-.D12 EX
08
E 068E 0.64
E
0
.E 0.42
0)
1
lo-,
10-5
lnia10-7 1c0-6
Initial concentration(M)
1.4
i
ie 1.2 -0 E
.5Z
1-CL 0
0.
0
E 0.6
-8
ma I
.r
0.4-E
2
0.2-6
-l 10-5
A Figure 4. (A) Effectof temper-l a b ature on the [
'4C]
artemisinin1.4-1
accumulation capacityofnor-1.2. mal (a) and HbH(b)
erythro-cytes.70 ,ul of
packed
cellswas0_4:
incubatedwith 6301ul
ofme-dium with initial artemisinin
0.6 concentration of130±7nM
041
at370C(o)and40C(m)for2
0.2-
h before accumulationmea-Fl L .
M1_E
_ENJsurements.
I, infected erythro-c P c P cytes; C, uninfected controlB
erythrocytes; P,
calculated data1.6a-
forparasites.
Error barsindi-1.4 caterangesof the results from
threeexperiments. (B)Effect ofiodoacetamide on accumu-lationcapacityof normal(a)
0.8 andHbH
(b) erythrocytes.
700.6
MAI
of packed cells waspre-treated with 500 Ml of 100 iM
0.4l iodoacetamide at370Cfor 1
0.2 h,and 130
Al
of[14CJ-O., _ _ artemisinin was added to a
C P concentration of 100 nM
be-fore further incubation for 2 h(i).(o) Untreated control. I, infected erythrocytes;C, uninfected
controlerythrocytes;P, calculated dataforparasites.Error bars indi-cateranges of the results fromduplicate samples.
25
-20
-
5-257
20
-
10-
5-A 25
-.
20
-15
-.
0
m
10
5-0
0
0
20 40 60 80 100 120 140
Final medium concentration (nM)
C
proportion of
uninfected
HbH/Hb ConstantSpring
erythro-cytesin the culturecaused adecrease in the
IC50 for
P.falci-parum
infecting
HbH/HbConstant Spring
erythrocytes, andanincrease in the
proportion of uninfected
erythrocytes causedacorresponding increase inthe IC50 forthe
parasite infecting
genetically normal erythrocytes. The variation in
IC50
valueswasparalleled by thatindrugaccumulation capacities of eryth-rocytemixtures (Fig. 7).Decreaseof the
hematocrit
value inthe culture with variant erythrocytes, thereby leadingto less
drug accumulation by
uninfected
erythrocytes, also led to acorrespondingdecreasein IC50, whilethehematocrit valuehad
little influenceonIC50of parasites infecting genetically normal erythrocytes (Fig. 8).
Although drug accumulation of
uninfected
HbH andHbH/Hb Constant Spring erythrocytes was far higher than
uninfected genetically
normal erythrocytes, theaccumulation
of allgenetictypes waslinear with externaldrug
concentration
to> 2
,gM.
By contrast,the accumulation by the parasiteswassaturable, consistent withapreviousreportin whichthe
accu-mulation of dihydroartemisinin showed saturation at drug
concentrations of> 3.86
1uM
(22). Furthermore,temperaturedependenceandpartial inhibition by iodoacetamide of
para-sitedrugaccumulation(Fig. 4) indicated at least partial
depen-denceonmetabolic energy.Exceptfor dependenceon
tempera-tureof HbH erythrocytes, therewaslittle effect oftemperature oriodoacetamideon drugaccumulation of uninfected erythro-cytes.
B
.
.
0
0 , 0 0 0.6 0.8 1
0 0.2 0.4 0.6 0.8 1
Erythrocyteaccumulationcapacity (nmole/,M)
0 0
0
0
.1I.
0.2 0.4 0.6 0.8 1
Parasiteaccumulationcapacity(nmole/VM)
1.2
Figure5.Correlationbetween the IC50 values and (A) final['4C]artemisininconcentration inthemedium of uninfectederythrocytes,r=0.83;
(B) accumulation capacities of uninfectederythrocytes, r= 0.88; and (C) calculateddatafortheaccumulationcapacities of intracellular P.
falciparum, r=0.68.(.) Normal erythrocytes, (c ) HbH erythrocytes, and (in) HbH/Hb Constant Springerythrocytes.
ArtemisininResistanceofMalariaParasitesina-ThalassemicErythrocytes 471
E .-i
C) Cad
Ck
LZ
3E
C.
eG
9 e
Fu
k
c
15
0 - 10
-e
eC
1
25 6
.
/
050
40
-3
-930
-0
/
0
0
20 40 60 80
%Hb H/HbCSerythrocytes
20
10
-100
Figure 6.The ICse valuesfor artemisinin against P.falciparum in-fectingnormal(oand*) and HbH/Hb Constant Spring (o and .) erythrocytesincultures with variouscompositionofuninfected
nor-mal andHbH/Hb ConstantSpring erythrocytes. Data fortwo
exper-imentsareshown.
0
0
-1
/
7,
0
2
7,x 0~~ 0
0 0
0 0.3 0.6 0.9 1.2 1.5 1.8
%Hematocnt
Figure 8. TheICsOvalues forartemisinin against P.falciparum in-fectingnormal(oand .),HbH(o),andHbH/Hb ConstantSpring, (m)erythrocytesatvarioushematocrit values. Data fortwo
experi-mentsareshown.
Although the data in Figs. 6 and 8 indicate thatthe
IC50
valuesaremainly determinedby the genetictypesandconcen-trations ofuninfected cells that compete for the drug in the
medium, other data (Fig. 2, Cand F, and 5 C) suggestthat
drug accumulation capacity ofthe parasite infectingthe variant
erythrocytes, i.e., the accumulationcorrected foraltered
me-dium concentration,mayalso be another determining factor,
albeitaless importantone. The low drug accumulation ofthe
parasite infectinggenetically variant erythrocytesmaybe due
tolimitationof transit ofthe drug from erythrocytecytoplasm
totheparasiteor atruedecrease in drug accumulation capacity oftheparasite infectingthe variant cells.
Ourresultsreported herecarryimplications for the
treat-ment of P.falciparum-infected patients with artemisinin or
their derivatives, especially in theareaswith high prevalence of
thalassemic/hemoglobinopathic genes, which are largely the
same asthose withmalaria endemicity (5). Hence, ifthe differ-encesinsensitivity alsoexistatthe clinical level, there will bea
0.4
2
0~~~~~~~~~
E
*' 0.31
0
0. E
C
0 20 40 60 80 100
%Hb H/Hb CSerythrocytes
Figure7. The['4C]artemisinin accumulationcapacitiesof mixtures
ofuninfected normal andHbH/HbConstantSpringerythrocytesas afunctionof the percentage of the latter.
potential problem ofpossibleinduction of resistanceto arte-misinin duetoexposureof the parasite infecting variant
eryth-rocytestosubcurative doses of thedrug.Inaddition, theremay
also be problems related with drug toxicity topatients who carrying the genetically variantgenes. Ithas been shown that hemin formsanadduct withartemisinin, which leadsto oxida-tive damage on membranes (23). In addition, it was also shown that ferrous, ferric, hemoglobin, and hemin enhanced the generation of superoxide and hydrogen peroxide by arte-sunate(24). Variant erythrocytes have increased hemichrome (25), nonheme iron (26). Thesecomponentsand other hemo-globin degradative products containing iron may potentiate
artemisinin-induced oxidative damages to the erythrocytes, leadingtosuch consequencesasdecreased erythrocyte
defor-mability and increased methemoglobin level. Eveninthe ab-senceof these drugs,suchdamageshavealreadybeen foundto be morepronounced inerythrocytes containingunstable Hb
(HbSS,HbEE, and HbAE) thanin normalerythrocytes (16). Therefore, the variationineffective dosage and therapeutic
in-dexof artemisinin fortreatmentof P.falciparuminfection in
thesegroupsofpatients should be investigated.
Acknowledgments
This work received financialsupportfrom United Nations
Develop-mentProgram/World Bank/World Health Organization Special
Pro-grammefor Research andTraininginTropical Diseases,the Rocke-fellerFoundation,and the INSERMNord-SudProgramme.
References
1.Wernsdorfer,W.H. 1991.Thedevelopmentandspreadofdrug-resistant malaria. Parasitol.Today.7:297-303.
2.WorldHealthOrganization.1986.Thedevelopmentof artemisininand its
derivatives.Meetingof the ScientificWorking GroupontheChemotherapyof
Malaria. World HealthOrganization,Geneva.6-7 October 1986.
3.Klayman,D. L. 1985.Qinghaosu(artemisinin),anantimalarialdrugfrom
China. Science(Wash.DC).228:1049-1055.
4.Wasi, P.,P.Pootrakul,S.Fucharoen,P.Winichagoon,P.Wilairat,and A.
Promboon. 1985. ThalassemiainSoutheastAsia: determination ofdifferent de-greesofseverityof anemiainthalassemia. Ann.NYAcad. Sci. 445:119-126.
5.Weatherall,D. J. 1987. Commongeneticdisordersof the red celland the
"malariahypothesis."Ann.Trop.Med. Parasitol. 81:539-548.
6.Nagel,R.L.,and E.F.Roth,Jr. 1989.Malaria and red cellgeneticdefects.
Blood. 74:1213-1221.
7.Yuthavong, Y.,P.Butthep,A.Bunyaratvej,S.Fucharoen,and S.
Khu-472 S.Kamchonwongpaisan, G. Chandra-ngam, M. A.Avery,and Y. Yuthavong
20-
15-C c
0
- 10
-
5-0
60-smith. 1988. Impaired parasite growthand increasedsusceptibilityto phagocyto-sis of P.falciparum-infecteda-thalassemia and/or hemoglobin Constant Spring erythrocytes. Am. J. Clin. Pathol.89:521-525.
8.Ifediba,T.C.,A.Stem,A.Ibrahim,and R. F.Rieder.1985.Plasmodium falciparuminvitro. Diminished growth in hemoglobin H disease erythrocytes.
Blood. 65:452-455.
9.Luzzi, G. A., A. H. Merry, C. I. Newbold, K. Marsh, G. Pasvol, and D. J. Weatherall. 1991. Surface antigen expression on Plasmodium falciparum-in-fected erythrocytes is modified in a- andfl-thalassemia.J. Exp. Med. 173:785-791.
10. Yuthavong, Y., P.Butthep, A. Bunyaratvej, and S. Fucharoen. 1989. Decreasedsensitivityto artesunate andchloroquine of Plasmodiumfalciparum infectinghemoglobin H and/or hemoglobin Constant Spring erythrocytes. J. Clin.Invest.83:502-505.
11.Brunori, M., G. Falcioni, E. Fioretti, B. Giardina, and G. Rotilio. 1975. Formation of superoxide in the autoxidation of the isolated a and 3 chains of humanhemoglobinanditsinvolvementinhemichromeprecipitation.Eur.J. Biochem. 53:99-104.
12. Wozencraft,A.O., H. M. Dockrell, J. Taverne, G. A. T. Targett, and J. H. L.Playfair.1984.Killing of human malaria parasites by macrophage secre-tory products.Infect.Immun.43:664-669.
13. Friedman, M. J. 1979. Oxidant damage mediates red cell resistance to malaria. Nature(Lond.).280:245-247.
14. Clark,I.A., and N. H.Hunt. 1983. Evidence for reactive oxygen interme-diatescausinghemolysisandparasite death in malaria. Infect. Immun. 39:1-6.
15. Levander, 0.A., A. L.Ager, J. V. C. Morris, and R. G. May. 1989. Qinghaosu, dietaryvitamin E, selenium, and cod-liver oil: effect on the suscepti-bility ofmice to the malarial parasite Plasmodiumyoeldi.Am.J. Clin. Nutr. 50:346-352.
16.Scott, M. D., S. R. Meshnick, R. A. Williams, D. T.-Y. Chiu, H. C. Pan, B. H.Lubin, and F. A. Kuypers. 1989.Qinghaosu-mediated oxidation in normal andabnormal erythrocytes.J.Lab. Clin.Med. 114:401-406.
17.Krungkrai, S. R., andY.Yuthavong. 1987. The antimalarial actionon
Plasmodiumfalciparumofqinghaosuandartesunatein combination with agents whichmodulate oxidant stress. Trans. R.Soc.Trop.Med.Hyg.81:710-714.
18. Trager, W., and J. B. Jensen. 1976. Human malaria parasites in continu-ous culture. Science(Wash. DC). 193:673-676.
19. Lambros,C.,and J. P.Vanderberg. 1979.Synchronizationof Plasmo-diumfalciparum erythrocyticstages in culture.J.Parasitol. 65:418-420.
20. Dejkriengkraikhul, P., R.Santiyanont, and P. Wilairat. 1982. Arapid method forconcentrating schizont-infectedred cells ofPlasmodiumfalciparum. SoutheastAsianJ. Trop.Med. PublicHealth. 13:663-665.
21.Desjardins, R. E., C. J.Canfield,J. D.Haynes,and J. D.Chulay. 1979. Quantitative assessment of antimalarialactivityin vitrobyasemiautomated microdilutiontechnique. Antimicrob.AgentsChemother. 16:710-718.
22. Gu, H. M., D. C. Warhurst, andW.Peters. 1984.Uptake of[3H] dihy-droartemisininby erythrocytesinfectedwithPlasmodiumfalciparuminvitro. Trans. R.Soc. Trop.Med.Hyg.78:265-270.
23. Meshnick,S. R., A. Thomas,A.Ranz,C. M.Xu,and H.-Z. Pan. 1991. Artemisinin (qinghaosu):the roleofintracellularheminin its mechanism of antimalarial action. Mol. Biochem. Parasitol. 49:181-190.
24. Pan,H.-Z., F.-B.Lin, and Z.-N. Chang. 1991. Oxidative damage in
ma-laria-infectedhumanerythricytes and the actionofsodium artesunate. In Anti-malarialDrug Development in the People's Republic of China. Report of the MeetingoftheScientificWorking Group on the Chemotherapy of Malaria. 24-26April 1989.World Health Organization, Geneva. J. Y. Shen, editor. 53-61.
25. Rachmilewitz, E. A., J. Peisach, T. B. Bradley, and W. E. Blumbery. 1969. Roleofhaemichromesinthe formationof inclusionbodies in haemoglobin H disease.Nature(Lond.).222:248-250.
26.Vatanavicharn,S., M.Anuwatanakulchai,P.Yenchitsomanus, and C. Siddhikol. 1988.Relationship ofserum vitamin E, erythrocyte non-hemeiron, andmalonyldialdehyde (lipid membraneperoxidationproduct) inthalassemia. InThalassemia: Pathophysiology and Management. Part A. S. Fucharoen, P. T. Rowley, and N. W. Paul, editors. Alan R. Liss, New York. 207-21 1.