Available online at www.sciencedirect.com

www.elsevier.com/locate/ijpara

International Journal for Parasitology 37 (2007) 1551–1557

Production of erythropoietic cells in vitro for continuousculture of Plasmodium vivax

Tasanee Panichakul a, Jetsumon Sattabongkot b, Kesinee Chotivanich a,Jeeraphat Sirichaisinthop c, Liwang Cui d, Rachanee Udomsangpetch e,*

a Department of Clinical Tropical Medicine, Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailandb Department of Entomology, AFRIMS, Bangkok, Thailand

c Center of Malaria Research and Training, Ministry of Public Health, Saraburi, Thailandd Department of Entomology, Pennsylvania State University, USA

e Department of Pathobiology, Faculty of Science, Mahidol University, Rama VI Road, Bangkok 10400, Thailand

Received 27 February 2007; received in revised form 11 April 2007; accepted 9 May 2007

Abstract

Plasmodium vivax cannot be maintained in a continuous culture. To overcome this major obstacle to P. vivax research, we have devel-oped an in vitro method to produce susceptible red blood cell (RBC) precursors from freshly isolated human cord hematopoietic stemcells (HSCs), which were activated with erythropoietin to differentiate into erythroid cells. Differentiation and maturation of erythroidcells were monitored using cell surface markers (CD71, CD36, GPA and Fy6). Duffy+ reticulocytes appeared after 10 days of erythroidcell culture and exponentially increased to high numbers on days 14–16. Beginning on day 10 these erythroid cells, referred to as growingRBCs (gRBCs), were co-cultured with P. vivax-infected blood directly isolated from patients. Parasite-infected gRBCs were detected byGiemsa staining and a P. vivax-specific immunofluorescence assay in 11 out of 14 P. vivax isolates. These P. vivax cultures were contin-uously maintained for more than 2 weeks by supplying fresh gRBCs; one was maintained for 85 days before discontinuing the culture.Our results demonstrate that gRBCs derived in vitro from HSCs can provide susceptible Duffy+ reticulocytes for continuous culture ofP. vivax. Of particular interest, we discovered that parasites were able to invade nucleated erythroid cells or erythroblasts that arenormally in the bone marrow. The possibility that P. vivax causes erythroblast destruction and hence inflammation in the bone marrowneeds to be addressed.� 2007 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved.

Keywords: Malaria; Plasmodium vivax; Reticulocyte; Erythroblast; Growing red blood cells; Hematopoietic stem cells; Cord blood

1. Introduction

Plasmodium vivax malaria accounts for over half ofmalaria cases outside Africa and is a major public healthproblem in many regions of Asia and Latin America (Men-dis et al., 2001). As an important factor of morbidity andan obstacle for economic development, the re-emergenceand resurgence of P. vivax in many endemic areas and

0020-7519/$30.00 � 2007 Australian Society for Parasitology Inc. Published b

doi:10.1016/j.ijpara.2007.05.009

* Corresponding author. Tel.: +66 2 201 5576; fax: +66 2 354 7158.E-mail address: scrudmu@yahoo.com (R. Udomsangpetch).

the appearance of drug-resistant parasite strains are a greatconcern. Despite this obvious significance, research onP. vivax has lagged far behind that on Plasmodium falcipa-

rum, largely due to the lack of a continuous culture tech-nique for this parasite. The development of an in vitroculture technique was central to the advances in P. falcipa-

rum research (Trager and Jensen, 1976). However, cultur-ing P. vivax has been difficult because of the stringentrequirement for reticulocytes as the target cells.

Compared with P. falciparum, P. vivax has many dis-tinct biological characteristics, including a hypnozoitestage that is responsible for relapses of the disease.

y Elsevier Ltd. All rights reserved.

1552 T. Panichakul et al. / International Journal for Parasitology 37 (2007) 1551–1557

Whereas P. falciparum has evolved multiple invasion path-ways to infect red blood cells (RBCs) (Cowman and Crabb,2006), invasion of P. vivax is dependent on interaction withthe Duffy antigen/receptor for chemokines (DARC)(Miller et al., 1975; Horuk et al., 1993). Furthermore,P. falciparum can infect both normocytes and reticulocytes(Pasvol et al., 1980; Mitchell et al., 1986), but P. vivax canonly infect reticulocytes, which are often below 1% in nor-mal blood (Kitchen, 1938; Mitchell et al., 1986; Monset al., 1988). Consequently, efforts for establishing short-and long-term P. vivax cultures have been undertaken toutilize blood sources enriched with reticulocytes. In anearly attempt to culture P. vivax, Golenda et al. (1997)developed a sophisticated protocol that uses reticulocyte-enriched blood from patients treated for hemochromatosis.However, this technique received very limited application,probably due to the unavailability of this blood source inmost P. vivax-endemic areas. In an earlier study, we devel-oped a technique for short-term culturing of P. vivax freshfield isolates and demonstrated that the parasites can bestudied ex vivo for one week without any requirement foradditional reticulocytes (Chotivanich et al., 2001). Thisshort-term culture system is suitable for studying drugresistance in field parasite isolates (Russell et al., 2003).To extend this short-term culture system, we have soughta more convenient source of reticulocytes, umbilical cordblood, and maintained P. vivax culture for over 1 month(Udomsangpetch et al., 2007).

Hematopoietic stem cells (HSCs) are of interest in bothclinical medicine and basic developmental biology. Allmature cells in the blood are derived from HSCs and allblood cell types can be generated from a single HSC (Osa-wa et al., 1996). HSCs can be obtained from bone marrow,peripheral blood and umbilical cord blood, and HSCs fromall these sources can be cultured in vitro to generate fullymature RBCs (Giarratana et al., 2005). In recognizingHSCs as a potential reticulocyte source, we have developeda technique to culture RBCs from human cord blood stemcells and demonstrated that these growing RBCs (gRBCs)can be used for continuous in vitro culture of P. vivax.

2. Materials and methods

2.1. Collection and isolation of HSCs

Forty-seven umbilical cord blood samples were collectedfrom normal, full-term deliveries for HSC isolation. TheEthical Committee of Research on Human Beings fromRamathibodi Hospital, Faculty of Medicine, Mahidol Uni-versity (ID 04-45-16) approved this study. Mononuclearcells (MNCs), including HSCs, were separated from cordblood by using an IsoPrep solution (Robbins ScientificCorporation, CA, USA). HSCs were then isolated fromthe MNC fraction by using a CD133 isolation kit withmagnetic microbead selection and Mini-MACS columns(Miltenyi Biotech, Germany) as described in the manufac-turer’s protocol. Viability of the cells was determined by

trypan blue staining. The isolated HSCs/CD133+ cells werecultured and used for producing gRBCs.

2.2. Production of RBCs

RBCs were produced from HSCs/CD133+ cells using amodified procedure (Giarratana et al., 2005). HSCs/CD133+ cells, 2 · 105 cells/ml, were cultured in StemlineIImedium (Sigma–Aldrich Corporation, MO, USA) supple-mented with 100 ng/ml stem cell factor (SCF) (PeproTech,Rocky Hill, NJ, USA), 5 ng/ml IL-3 (R&D Systems, Inc.,MN, USA), 10 lM hydrocortisone (Sigma–Aldrich),100 lg/ml transferrin (Sigma–Aldrich), 100 lg/ml Humu-lin� N (Lilly Pharma Fertigung UND Distribution, Gies-sen, Germany), 0.18 mg/ml ferrous sulfate (Sigma–Aldrich), 0.16 M monothioglycerol (Sigma–Aldrich) and4 IU/ml erythropoietin (EPO; Cilag AG International,Zug, Switzerland). The first step of culture was cell expan-sion for 8 days and on day 4, cells were diluted in 4 volumeof fresh medium. At the second step, cells were cultured for4 days in StemlineII medium without SCF, IL-3 and hydro-cortisone. At the last step (8 days), cells were cultured inStemlineII medium without cytokines and maintained inmedium supplemented with 10% human AB serum (Pro-moCell�, Heidelberg, Germany). All cultures were incu-bated at 37 �C with 5% CO2. One-week-old erythroidcells were kept in freezing medium at �80 �C or liquidnitrogen until use.

2.3. Determination of maturation of erythroid cells

Surface membrane markers were analyzed to confirmcell types of HSCs and derived erythroid cells. Cell markerswere detected using mouse antibodies against CD133 (Mil-tonic Biotech), CD34 (Becton Dickinson Biosciences, CA,USA), CD38, CD71, CD45 (Celtic Laboratories, CA,USA), CD36 (Serotec Inc., NC, USA), glycophorin A(Beckman Coulter�, Inc., FL, USA) and Fy6 (kindly pro-vided by Dr. C. King, Center for Global Health and Dis-ease, Case Western Reserve University School ofMedicine, Cleveland, OH, USA) that were conjugated withfluorescein isothiocyanate or phycoerythrin and analyzedby flow cytometry (FACScan, Becton Dickinson). Cellswere also stained with Giemsa and Brilliant cresyl blue toexamine the morphology of RBCs and reticulocytes,respectively. Enucleated cells were monitored for standardhematological variables, including the MCV (fl), MCHC(g/dl) and MCH (pg/cell) using a Technicon H3 RTX/RTC� automat (Bayer Corporation, NY, USA).

Heterogeneous gRBC populations were separated byPercoll using 30–60% discontinuous gradients (Sigma–Aldrich). The gRBCs in suspension were layered on topof the Percoll gradients and centrifuged at 1200g for20 min at 20 �C. Each cell fraction was separately collectedfor analysis of the cell markers and morphology. Separatedcells from two fractions of 50% and 60% Percoll were usedfor parasite cultivation.

Fig. 1. Production of red blood cells from hematopoietic stem cells(HSCs). (a) HSCs (arrow) in the mononuclear cell fraction were enrichedby using a CD133 isolation kit with magnetic microbead selection. (b)Giemsa staining showed morphology of erythroblasts with nucleoli (day 8)and mature red blood cells on day 20 of the culture.

T. Panichakul et al. / International Journal for Parasitology 37 (2007) 1551–1557 1553

2.4. Co-cultivation of P. vivax-infected RBCs with gRBCs

Plasmodium vivax parasites were obtained from malarialpatients attending the malaria clinic in Mae Sot, Tak prov-ince, Thailand. Patients who had parasitemias rangingbetween 0.01% and 0.1% were selected for study. The Eth-ical Committee of Mahidol University, Bangkok, Thailandapproved this study (MU2006-007). Whole blood withasexual stage vivax parasites, confirmed by microscopicexamination, was collected and then filtrated by Plasmod-ipur filter (Euro-Diagnostic B.V., The Netherlands) toremove white blood cells. To obtain asexual parasites,packed infected RBCs from 5 ml of patient blood werediluted 1:20 with RPMI1640 (Invitrogen�, CA, USA), lay-ered on 60% Percoll, and centrifuged at 1200g for 20 min at20 �C (Pasvol et al., 1978). Collected parasites were co-cul-tured with 10–20-day-old gRBCs and the ratios of para-sites: gRBCs were between 1:10 and 1:20. Co-cultures ofparasites and gRBCs were maintained in a 12-well tissueculture plate (Corning Incorporated Costar�, NY, USA)with 1 ml McCoy medium (Invitrogen�) supplementedwith 25% human serum from normal group AB donors.All cultures were incubated at 37 �C in 5% CO2 and exam-ined three times per week for parasitic infection.

2.5. Determination of P. vivax infection in gRBCs

Infected RBCs were determined by Giemsa staining andimmunofluorescence assay (IFA). Cells from the parasite-gRBC co-cultivation were collected on glass slides by cyto-spin, fixed with 95% ethanol and cold acetone, and stainedwith Giemsa or IFA. For IFA, a mouse mAb 3F9 againstPlasmodium yoelii (Gao et al., 1995) cross-reacted withhuman malaria was used to detect parasites. Goat anti-mouse IgG-conjugated with fluorescence (Dako Cytoma-tion, Denmark) was used as a secondary antibody.

2.6. Continuous culture of P. vivax with gRBCs

Co-cultures of vivax parasites and gRBCs were main-tained in McCoy medium with 25% human serum at37 �C in 5% CO2. One hundred micro-liters of cells fromco-culture were harvested and 100 ll of fresh gRBCs werethen replaced. Parasite-infected gRBCs were determined byGiemsa staining and IFA using mAb 3F9. Co-cultureswere examined three times per week for determination ofparasitic infection. The period of continuous culture wasmonitored for 3 months.

3. Results

3.1. Establishment of RBC cultures from cord blood HSCs

HSCs were isolated from the mononuclear cell fractionof cord blood using the CD133+ marker (Fig. 1a). Flowcytometry analysis of the isolated HSCs showed that95% were CD133+, CD34+ and CD45+dimly. From 40

to 100 ml of cord blood, 2 · 105 to 1 · 106 HSCs were iso-lated with 100% cell viability. HSCs were cultured for 20days in a serum-free medium, StemlineII, supplementedwith cytokines and other factors that are required for cellproliferation and differentiation. Specifically, cell prolifera-tion and erythroid differentiation were induced with IL-3,SCF and EPO. Erythrocytes were generated from HSCs inthree cultivation steps and erythroid maturation was mon-itored by cell morphology and surface markers [CD34,CD38, CD71, CD36, glycophorin A (GPA) and Fy6(DARC)]. The first step resulted in up to 100-fold expan-sion of the cell population with 80% of cells becoming ery-throblasts on days 8 and 10 (Fig. 1b). Subsequentproliferation and differentiation of the cell populationswere less synchronous. By day 14, cell proliferationreached 1500-fold, resulting in a heterogeneous cell popu-lation of mixed erythroblasts, nucleated RBCs, matureRBCs and some granulocytes (Fig. 1b). Subsequently, cellsbecame more mature and a large number of mature RBCswere detected on day 20 (Fig. 1b). These mature RBCsresembled native RBCs with a mean corpuscular volume(MCV) of 96 fl, a mean corpuscular hemoglobin (MCH)of 38.6 pg/cell and a mean corpuscular hemoglobin

Fig. 3. Reticulocytes were identified by Brilliant cresyl blue staining.Number of reticulocytes in 1000 cells that appeared during 20 days ofgrowing red blood cell (gRBC) cultivation. Data represent mean ± SDfrom three independent experiments.

Table 1In vitro Plasmodium vivax cultures from clinical isolates

Isolateno.

Number ofparasites · 106

Number oferythroidcells · 106

%Parasitemiaat day 0

Period ofcontinuousculture (days)

1 0.77a 9.7a 7.3 852 1.72 17.5 8.9 603 1.14 13.44 7.8 464 1.15 16.8 6.4 435 0.33 2.8 10.5 396 1.2 18.48 6 287 0.57 4.37 11.6 25

1554 T. Panichakul et al. / International Journal for Parasitology 37 (2007) 1551–1557

concentration (MCHC) of 40.1 g/dl. Cell growth, differen-tiation and maturation were associated with changes in cellsurface markers as determined by flow cytometry (Fig. 2).During the first week of culture, CD34+ cells decreasedrapidly, which was accompanied by the increase ofCD71+, CD38+ and CD36+ cells. As the cells becamemore mature, cell populations expressing these markersalso decreased. Two weeks after initiation of the culture,erythroid and mature RBC populations graduallyincreased, correlated with the increase of DARC- andGPA-positive cells. Reticulocytes, determined by Brilliantcresyl blue staining, were present throughout the entireculture period with a peak of 0.5% observed on day 14(Fig. 3). Our results demonstrated that under these cultureconditions, cord blood HSCs could proliferate, differenti-ate and produce Duffy+ erythroid cells, includingreticulocytes.

3.2. In vitro culture of P. vivax in gRBCs and detection of

parasite-infected gRBCs

To determine whether the gRBC culture was suitablefor P. vivax culture, 14 parasite isolates from patientsand erythroid cells were used for culture (described inTable 1). Erythroid cells, 2.8–26 · 106 from 10 to20-day-old gRBC culture were co-cultivated with 0.33–1.72 · 106 asexual stages of P. vivax-infected RBCs iso-lated from the blood of acute P. vivax-infected volunteers.Parasitemias of co-cultures at day 0 were 3.5–11.6%. Par-asites were detected in cultures using Giemsa staining andIFA with a pan-specific anti-Plasmodium mAb 3F9 that

Fig. 2. Surface markers of growing red blood cells (gRBCs) characterizedby flow cytometry. Cells collected from the cultures at 2-day-intervals werestained with a specific mAb to CD34, CD38, CD71, CD36, Duffy bloodgroup (Fy6) and gylcophorin A. Graph represents mean percentage ofpositive cells ± SD from three independent experiments. Under the X-axis,morphology of erythroid cells from gRBC cultured over 20 days areillustrated to match with the surface marker expression.

8 1.47 19.25 7 219 0.58 15.95 3.5 17

10 1.63 26.13 5.8 1711 0.39 6.56 5.6 1412 0.55 8.61 6 813 1.24 15.4 7.4 4b

14 0.47 12.6 3.6 4b

a The number of parasites (asexual stages) and erythroid cells per well.b The parasites in the culture developed into a large number of

gametocytes.

can detect all erythrocytic stages of P. vivax. This mAbcould react weakly with sexual stage (gametocyte)P. vivax and also with asexual and sexual stages of P. fal-

ciparum (data not shown). Giemsa staining and IFAshowed that vivax merozoites could invade gRBCs anddeveloped complete asexual stages, rings, trophozoitesand schizonts (Fig. 4a and b) and interestingly, parasiteswere able to invade erythroblasts (Fig. 4a, 1–3).

3.3. Continuous culture of P. vivax with gRBCs

To establish cultures from P. vivax field isolates, freshgRBCs were periodically added to provide a continuoussupply of susceptible erythroid cells for the merozoitesto invade. From 14 P. vivax field isolates, five could becultured beyond 1 month (39–85 days), including two

Fig. 4. Determination of Plasmodium vivax-infected growing red blood cells (gRBCs). (a) Plasmodium vivax in gRBC culture stained with Giemsa showsring stage parasites in erythroblasts (1, 2) and developmental stages of the parasites, trophozoite and schizont (3–6). (b) Immunofluorescence assay (IFA)using mAb 3F9 to determine P. vivax-infected gRBCs. Ring stage parasitized cell (set 1, arrows indicate the parasites), schizont-stage parasite (set 2) andIFA control without mAb 3F9 are shown. Parasite nuclei were stained with DAPI.

T. Panichakul et al. / International Journal for Parasitology 37 (2007) 1551–1557 1555

parasite isolates that were continuously cultured for 60and 85 days (Fig. 5, and Table 1) at which time the exper-iment was terminated. The remaining seven parasite iso-lates could be cultured continuously for less than 1month (8–2 days) as shown in Table 1. The number ofparasite-infected cells, determined by IFA, fluctuated dur-ing the culture period, in part reflecting the periodic addi-tion of fresh gRBCs. Parasitemias of these continuousgRBC-parasite cultures, excluding gametocytes, werebetween 0.0001 and 0.0013% as shown in Fig. 5. In addi-tion, two P. vivax isolates grew well initially, but were lostwithin a week. In these short growing cultures, parasitesdeveloped to a large percentage of gametocytes. Despitelow parasitemia, the erythroid cells growing in vitro couldserve as new host cells for continuous culture of P. vivax.

4. Discussion

Continuous culture of P. vivax is limited by the scarcityof reticulocytes in normal blood, which are the target cellsof P. vivax for infection (Kitchen, 1938; Mons et al., 1988).Attempts to solve this problem have concentrated on using

blood sources that are enriched in reticulocytes (Golendaet al., 1997; Udomsangpetch et al., 2007). One reportshowed that reticulocyte-enriched monkey blood obtainedafter treatment with a hemolytic drug could sustainP. vivax growth in vitro for six cycles (Mons et al., 1988).A more sophisticated method for continuous P. vivax cul-ture was developed with reticulocyte-enriched humanblood cells from hemochromatosis patients (Golendaet al., 1997). However, lack of such blood sources in ende-mic areas hampered wide application of this technique inP. vivax culture. To seek a more convenient reticulocyte-enriched blood source, we have tried to culture P. vivaxfield isolates using umbilical cord blood (Udomsangpetchet al., 2007); but the parasites could not be maintainedbeyond 40 days, perhaps due to the inhibitory effect ofhemoglobin F in the cord blood for parasite growth. Thegeneration of RBCs from human HSCs is an alternativeapproach for in vitro RBC production (Giarratana et al.,2005). Here, we have extended earlier research and appliedan in vitro method to differentiate erythroid cells to reticu-locytes from HSCs and demonstrated that they are able tomaintain continuous P. vivax cultures.

Fig. 5. Continuous cultures of Plasmodium vivax in growing red blood cells (gRBCs.) Number of parasitized cells from two vivax parasite isolates co-cultured with gRBCs (from Table 1, isolate 1 and 2) were determined by IFA using mAb 3F9. Parasitemia of the cultures maintained continuously for 85days (–d–), and 60 days are shown. Arrow heads (. and ,) indicate addition of fresh gRBCs into the 85 and 60-day-old cultures, respectively.(*) Parasitemia was not determined.

1556 T. Panichakul et al. / International Journal for Parasitology 37 (2007) 1551–1557

In our studies with cord blood HSCs, the cell populationsexpanded >1500-fold and reached full maturity in 20 days.From erythroid cell precursors to mature erythrocytes, cellsobtained in this way resembled those in normal blood interms of morphology, cell surface markers and other RBCcharacteristics, suggesting that they are fully functional. Ret-iculocytes were observed throughout this growth period andpeaked on day 14. The appearance of two cell surface mark-ers, DARC and GPA on day 14 reticulocytes, was a usefulindicator of the time when the erythroid cultures were suit-able for P. vivax invasion. The asynchronous maturationof the erythroid cells was an important attribute, whichallows continuous production of new reticulocytes. Thishas enabled a simplified culture procedure, and weekly addi-tions of the gRBCs were sufficient to maintain the parasitesin culture. Nevertheless, reticulocyte numbers were still lowin the culture and remain a potential limiting factor forhigher parasitemia. We showed that P. vivax merozoitesinvaded erythroblasts (Fig. 4a) and the fully mature schizo-nts were seen in erythrocytes. Previously, we have shownthat growth of P. vivax parasites required very high defor-mability of the erythrocyte membrane (Suwanarusk et al.,2004). The apparently low deformability of erythroblastsmay not support growth of the parasites to fully matureschizonts and this is one of the causes of low parasitemiain this culture. To further improve this culture condition,one should understand that asynchronous populations ofthe factor-driven erythroid cells were observed during thein vitro erythropoiesis. We suggest that the erythroblaststage should be carefully checked before use. Gradient cen-trifugation using 50–60% Percoll may then be applied to

enrich erythroblasts and provide a substantial number ofsynchronous erythroblastic stages to be used in subcultiva-tion for P. vivax culture. Normoblasts at the stages betweenpolychromatophilic and orthochromatophilic should beused in P. vivax culture as enucleation will soon ensue andyoung reticulocytes will be obtained. This approach shouldincrease parasitemia of the culture.

The continuous culture of P. vivax had been performedusing 14 parasite field isolates. The success of culturingthese P. vivax isolates with this method was 36%, giventhat we were able to culture five of the field parasite iso-lates for more than 30 days. Although reasons for the dis-appearance of parasites in culture are not fullyunderstood, this is presumably due to intrinsic differencesamong the parasite isolates. Gametocytogenesis in theearly in vitro cycles may be responsible for rapid loss ofthe asexual stages.

Compared with the P. vivax cultures using red cells fromumbilical cord blood (Udomsangpetch et al., 2007) gRBCsderived from HSCs have a wider range of young erythro-cytic stages providing more choices of target cells for par-asite invasion. The advantage of gRBCs is the expressionof receptors on the erythroid cells that are involved in inva-sion of P. vivax merozoites and the presence of appropriatehemoglobin. In addition, most reticulocytes in the cordblood are in a later stage and it is not possible to obtainany stages of erythroblasts from cord blood, whereas RBCsderived from HSCs can provide erythroblasts and newlyenucleated reticulocytes. Moreover, RBCs derived fromHSCs in the first week of culture can be cryopreserved asstock for 6–8 months with 100% cell viability. After thawing,

T. Panichakul et al. / International Journal for Parasitology 37 (2007) 1551–1557 1557

the erythroblasts continue their maturation to erythrocytesthat can be used in P. vivax culture.

One important finding of our study is the demonstrationthat P. vivax can invade erythroblasts or nucleated red cells(Fig. 4a, 1–3), which are normally in the bone marrow.Plasmodium vivax has been observed in the bone marrowof patients who exhibited signs of dyserythropoiesis,including irregularly shaped nuclei, multi-nuclearity andcytoplasmic stippling of some erythroblasts (Wickramasin-ghe and Abdalla, 2000). In addition, there is a case reportof one leukaemia patient who got vivax parasite infectionfrom an unrelated donor bone marrow transplant (Donnellet al., 1998). Another malaria patient who had traveled toSoutheast Asia was infected with P. vivax and showed pan-cytopenia in blood and bone marrow (Yamakawa et al.,1989). The presence of parasites in bone marrow and theability to infect erythroblasts, which is accompanied withsimultaneous destruction of the infected cells, can probablybe attributed to some characteristic pathological effects,including anemia, associated with this hemotropic parasite.

In conclusion, this technique is useful to study funda-mental biology of the parasite, its antigen and host–para-site interaction, which will support vaccine developmentand drug resistance testing.

Acknowledgements

The authors thank Drs. S. Hongeng, A. Jaovisidha, P.Butthep, Ms. P. Khumkomkul, and Mr. J. Opasnawakun,Ramathibodi Hospital, Faculty of Medicine, Mahidol Uni-versity, Bangkok, Thailand for providing cord blood spec-imens and for excellent technical assistance. We also thankDrs. O. Kaneko and J.H. Adams for critical comments onthe manuscript. This work was supported by Mahidol Uni-versity, Grants from The Fogarty International Center,USA (D43TW006,571), and the US Military InfectiousDiseases Research Program (MIDRP) USA.

Appendix A. Supplementary data

Supplementary data associated with this article can befound, in the online version, at doi:10.1016/j.ijpara.2007.05.009.

References

Chotivanich, K., Silamut, K., Udomsangpetch, R., Stepniewska, K.A.,Pukrittayakamee, S., Looareesuwan, S., White, N.J., 2001. Ex-vivoshort-term culture and developmental assessment of Plasmodium vivax.Trans. R. Soc. Trop. Med. Hyg. 95, 677–680.

Cowman, A.F., Crabb, B.S., 2006. Invasion of red blood cell by malariaparasites. Cell 124, 755–766.

Donnell, J.O., Goldman, J.M., Wagner, K., Ehinger, G., Martin, N.,Leahy, M., Kariuki, N., Dokal, I., Roberts, I., 1998. Donor-derived

Plasmodium vivax infection following volunteer unrelated bone mar-row transplantation. Bone Marrow Transpl. 21, 313–314.

Gao, Q., Tharavanip, S., Khusmith, S., Li-qin, W., Looareesuwan, S.,Wei-gi, Z., 1995. Two-site pan-species monoclonal antibody ELISAdetection of blood stage malaria antigen. Southeast. Asian J. Trop.Med. Public Health 26, 740–743.

Giarratana, M.C., Kobari, L., Lapillonne, H., Chalmers, D., Kiger, L.,Cynober, T., Marden, M.C., Wajcman, H., Douay, L., 2005. Ex vivogeneration of fully mature human red blood cells from hematopoieticstem cells. Nat. Biotechnol. 23, 69–74.

Golenda, C.F., Li, J., Rosenberg, R., 1997. Continuous in vitro propa-gation of the malaria parasite Plasmodium vivax. Proc. Natl. Acad. Sci.USA 94, 6786–6791.

Horuk, R., Chitnis, C.E., Darbonne, W.C., Colby, T.J., Rybicki, A.,Hadly, T.J., Miller, L.H., 1993. A receptor for the malaria parasitePlasmodium vivax: the erythrocyte chemokine receptor. Science 261,1182–1184.

Kitchen, S.K., 1938. The infection of reticulocytes by Plasmodium vivax.Am. J. Trop. Med. 18, 347–353.

Mendis, K.N., Sina, B.J., Marchesini, P., Carter, R., 2001. The neglectedburden of Plasmodium vivax malaria. Am. J. Trop. Med. Hyg. 64, 97–106.

Miller, L.H., Mason, S.J., Dvorak, J.A., McGinnis, M.H., Rothman, I.K.,1975. Molecular insights into receptors used by malaria parasites forerythrocyte invasion. Science 189, 561–563.

Mitchell, G.H., Hadley, T.J., McGinniss, M.H., Klotz, F.W., Miller, L.H.,1986. Invasion of erythrocytes by Plasmodium falciparum malariaparasites: evidence for receptor heterogeneity and two receptors. Blood67, 1519–1521.

Mons, B., Collins, W.E., Skinner, J.C., van der Star, W., Croon, J.J., vander Kaay, H.J., 1988. Plasmodium vivax: In vitro growth andreinvasion in red blood cells of Aotus nancymai. Exp. Parasitol. 66,183–188.

Osawa, M., Hanada, K., Hamada, H., Nakauchi, H., 1996. Long-termlymphohematopoietic reconstitution by a single CD34 low/negativehematopoietic stem cell. Science 273, 242–245.

Pasvol, G., Weatherall, D.J., Wilson, R.J., 1980. The increased suscep-tibility of young red cells to invasion by the malarial parasitePlasmodium falciparum. Br. J. Haematol. 45, 285–295.

Pasvol, G., Wilson, R.J., Smalley, M.E., Brown, J., 1978. Separation ofviable schizont-infected red cells of Plasmodium falciparum fromhuman blood. Ann. Trop. Med. Parasitol. 72, 87–88.

Russell, B.M., Udomsangpetch, R., Riechmann, K.H., Kotecha, B.M.,Colman, R.E., Sattabongkot, J., 2003. Simple in vitro assay fordetermining the sensitivity of Plasmodium vivax isolateds from freshhuman blood to antimalaria in areas where P. vivax is endemic.Antimicrob. Agents Chemother. 47, 170–173.

Suwanarusk, R., Cooke, B.M., Dondorp, A.M., Silamut, K., Satt-abongkot, J., White, N.J., Udomsangpetch, R., 2004. The deforma-bility of red blood cells parasitized by Plasmodium falciparum and P.

vivax. J. Infect. Dis. 189, 190–194.Trager, W., Jensen, J.B., 1976. Human malaria parasites in continuous

culture. Science 193, 673–675.Udomsangpetch, R., Somsri, S., Panichakul, T., Chotivanich, K., Siric-

haisinthop, J., Yang, Z., Cui, L., Sattabongkot, J., 2007. Short-termin vitro culture of field isolates of Plasmodium vivax using umbilicalcord blood. Parasitol. Int. 56, 65–69.

Wickramasinghe, S.N., Abdalla, S.H., 2000. Blood and bone marrowchanges in malaria. Bailliere’s Clin. Haematol. 13, 277–299.

Yamakawa, H., Kiyotaki, M., Hattori, Y., Obana, M., Matsuoka, Y.,Irimajiri, S., 1989. A case of Plasmodium vivax malaria complicatedwith pancytopenia due to hypoplasia of the bone marrow. Kans-enshogaku Zasshi. 63, 1043–1046.