Vol. 47, No. 3INFECTION AND IMMUNITY, Mar. 1985, P. 760-7660019-9567/85/030760-07$02.00/0Copyright © 1985, American Society for Microbiology

Pathogenicity, Stability, and Immunogenicity of a Knobless Clone ofPlasmodium falciparum in Colombian Owl Monkeys

SUSAN G. LANGRETH* AND ELIZABETH PETERSONDepartment of Microbiology, Uniformed Services University of the Health Sciences, Bethesda, Maryland 20814

Received 22 October 1984/Accepted 17 December 1984

The pathogenicity, immunogenicity, and morphological stability of a knobless clone of strain FCR-3 of thehuman malaria parasite Plasmodium falkiparum was investigated in Aotus monkeys. An early knob-bearing(K+), wild-type isolate of strain FCR-3 and the D3 knobless (K-) clone were adapted to Aotus monkeyerythrocytes in continuous culture, establishing the parasites in Aotus cells without exposure to in vivo cellularor humoral immune responses. All monkeys, intact or splenectomized, which were infected with wild-typeFCR-3 adapted to Aotus cells in vitro, developed virulent infections and had to be drug treated. The intactnonsplenectomized animals which received knobless D3 cloned parasites did not develop virulent infectionseven after multiple infections. The splenectomized monkeys which received the K- D3 clone had virulentinfections. Late-stage wild-type K+ parasites sequestered in both intact and splenectomized monkeys, whereaslate-stage D3 K- parasites did not sequester in the splenectomized animals. These results suggest that twoelements affected the pathogenicity of the malaria parasites in these experiments. Knobs on K+-infectederythrocytes enabled these parasites to sequester, presumably by attachment to capillary endothelium. Whenpresent, the spleen eliminated circulating K- late-stage erythrocytes, presumably by selection on the basis oftheir nondeformability. Although clone D3 K- parasites are nonvirulent in intact monkeys, they induced someimmunological protection against challenge with wild-type K+ parasites. The surface morphology of K--infected erythrocytes remains unaltered throughout these experiments, suggesting that loss of knobs is a stablecondition.

Infection with the human malaria parasite Plasmodiumfalciparum results in visible alterations at the surface of thehost erythrocyte. These alterations, called knobs, are 70- to100-nm electron-dense cuplike protrusions beneath the eryth-rocyte unit membrane. They have been observed in infectederythrocytes (IRBCs) of both humans and Aotus monkeys invivo and in vitro (1, 15, 21, 30). They are reported to containhistidine-rich protein of parasite origin (8, 12, 13, 18) and areantigenically distinct from the rest of the erythrocyte surface(14). Individuals with immunity to P. falciparum have anti-bodies in their sera which bind to the knobs (16). The knobsappear on the red cell surface at the stage when rings matureinto trophozoites. This timing correlates with the sequestra-tion of late stages from the peripheral circulation. The knobsare believed to be the sites of attachment of IRBCs to thecapillary endothelium in the deep vasculature, thereby ob-structing blood circulation and avoiding passage through thespleen (22).With the development of methods to maintain erythro-

cytic stages of P. falciparum in continuous culture (28),parasite variants which did not produce knobs soon weredetected in isolates from several geographical areas (17).Cultures have been enriched for knobless variants by plasmaexpander selection techniques (25). Knobless (K-) single-cell isolates then have been cloned from such cultures (31).Merozoites harvested from such a knobless clone have beenused to immunize Aotus monkeys, with muramyl dipeptideas the adjuvant (29).The stability and pathogenicity of viable knobless para-

sites of P. falciparum have not been vigorously examined invivo. It is unknown whether knobless parasites are more orless virulent than wild type or whether loss of knobs is astable condition. It also is unknown whether knobless (K-)

* Corresponding author.

trophozoites and schizonts will sequester from the periph-eral circulation of an infected animal. If they do not seques-ter, it is uncertain whether they will be recognized ordestroyed by the spleen. To investigate these parameters, aknobless clone and its parent wild-type (K+) strain of P.falciparum were adapted to owl monkey erythrocytes. Hu-man malaria parasites are usually adapted to monkey cellsby serial passage through several splenectomized animals.Because of reports by J. Barnwell et al. (3, 4), M. Hommelet al. (9, 10), and P. David et al. (7) of changes in antigenicityupon passage of malaria parasites through splenectomizedprimates, we did not adapt the K- clone and K+ wild-typeparasites in this way. Instead, the parasites were adapted invitro by continuous culture in owl monkey erythrocytes byrecently developed modifications ofhuman cell culture meth-ods (23).

In this report we present evidence that a knobless clone ofP. falciparum (FCR-3 clone D3) is not pathogenic but isimmunogenic in intact Colombian owl monkeys. Late-stageparasites of the K- clone, which lack knobs to bind tocapillary endothelium, do not sequester and therefore aresubject to destruction by the spleen. In addition, the surfacemorphology of Aotus erythrocytes infected with the K-clone remains knobless in cells isolated from infected intactor splenectomized owl monkeys and after long-term culture.

MATERIALS AND METHODS

Monkeys. Adult, healthy Colombian owl monkeys (Aotustrivirgatus griseimembra) were obtained from South Amer-ican Primates, Miami, Fla.; Walter Reed Army Institute ofResearch, Washington, D.C.; Litton Bionetics, Rockville,Md.; and the National Institutes of Health, Poolesville, Md.All animals were karyotyped and determined to be type II,III, or IV (Table 1).

760

on May 29, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

P. FALCIPARUM IN COLOMBIAN OWL MONKEYS 761

TABLE 1. Comparison of parasitemias in intact andsplenectomized monkeys infected with FCR-3 wild type or FCR-3

clone D3 (K-)a

Monkeysb parasitemia (%) Drug treated

FCR-3 wild type, intact115 (M, II) 4.1 Yes5335 (F, II) 6.5 Yes5336 (F, III) 5.5 Yes7921 (F, II) 0.8 Yesc8443 (M, II) 2.4 Yesc

FCR-3 wild type, splenectomized217J (F, II) 4.4 Yes

Clone D3, intact6902 (M, II) 0 No8490 (M, II) 0 NoWR 159 (M, II) 0 No5342 (F, II) 0.02 No114 (F, IV) 0 No216J (F, III) 0.005 NoWR 157 (F, IV) 0 No

Clone D3, splenectomized209J (M, III) 7.6 YesWR 135 (M, IV) 2.9 No5341 (F, III) 7.6 Yesa Peak parasitemias for each group cannot be averaged because of the

intervention by drug treatment. Criteria for drug treatment are detailed in thetext.bSex (F, female; M, male) and karyotype are within parentheses.c Monkeys were drug treated when patent, with reticulocytes > 20%, and

hematocrit < 20.

Animals were anesthesized with 0.1 to 0.15 ml of ketamine(100 mg/ml) for venipuncture and for infection with parasi-tized erythrocytes via the femoral vein. Blood was drawnwithout anticoagulant for serum samples at 7- to 10-dayintervals. The last preimmune blood draw was 7 days beforeinfection. Blood was taken from ear pricks for daily moni-toring of parasitemias on blood films and for samples forelectron microscopy. At peak parasitemias, blood with 10%anticoagulant (heparin or citrate phosphate dextrose) wasdrawn for recovery of parasitized erythrocytes for in vitroculture, cryopreservation, and preparation of antigen slidesfor immunofluorescence assays (IFAs).

Parasites. A stabilate of P. falciparum FCR-3 (FMG) fromthe Gambia (11), cryopreserved within 2 months of itsoriginal in vitro adaptation, was used throughout theseexperiments. This parental wild-type strain FCR-3 is mor-phologically and antigenically heterogeneous (15-17). TheD3 knobless clone of strain FCR-3 was obtained by micro-scopic selection by Trager et al. (31). Both wild-type FCR-3and the D3 clone were kindly provided by William Trager.

Parasite cultivation. Parasites (FCR-3 wild type and cloneD3) grown in human 0+ erythrocytes with A' serum by theTrager-Jensen (28) method were adapted to culture in Aotusmonkey erythrocytes (23). The adapted parasites were main-tained in vitro in Aotus erythrocytes for 1 month beforemonkeys were infected. The erythrocytes of each monkeywere tested for their ability to support the growth of both theD3 clone and the wild-type strain in vitro before in vivoinfections. Erythrocytes from infected monkeys exhibitingparasitemias of 0.005 to 7.0% were placed into culture toexpand parasite populations for cryopreservation and elec-tron microscopy.

Infection of monkeys from cultured parasites. Inocula wereprepared from cultures at parasitemias of approximately 2%,with ring stages predominant. The parasitized erythrocyteswere washed in RPMI 1640 medium to remove the AB serumand were diluted in RPMI 1640 medium without serum togive an inoculum of 0.5 ml containing a specific number ofparasitized erythrocytes, usually 5 x 105. After infection ofthe animals, the extra syringes which had been preparedwith infecting doses were returned to the laboratory, and theparasitized erythrocytes were returned to culture, passingthe cells through a 23-gauge needle, to simulate conditions ofanimal infection. These inoculum controls were culturedwith 10% human AB serum at 2% hematocrit for 4 to 7 daysto ascertain viability of the infecting parasites.Approximately 3 months before infection, several mon-

keys were splenectomized (217J, 209J, WR 135, and 5341).Monkeys 115, 7921, 8443, and 217J (splenectomized) re-ceived doses of 5 x 105 FCR-3 wild-type parasites. Monkeys6902, 8490, WR 159, 5342, 209J (splenectomized), WR 135(splenectomized), and 5341 (splenectomized) received initialdoses of 5 x 105 clone D3 parasites. The intact monkeys inthis group subsequently received increasingly higher dosesof clone D3, up to 1 x 107 parasites, at approximatelymonthly intervals. Monkeys 114, 216J, and WR 157 initiallyreceived 5 x 106 clone D3 parasites, followed by two dosesof 1 x 107 clone D3 parasites at approximately monthlyintervals. Monkeys 6902 and 8490 were challenged with 5 x105 FCR-3 wild-type parasites 1 month after their secondinfection with clone D3. Monkeys 114 and 216J were chal-lenged with S x 105 FCR-3 wild-type parasites after receiv-ing three doses of clone D3. Monkeys 5335 and 5336, neverbefore exposed to P. falciparum, also were given 5 x 105wild-type parasites at the time of these challenge infections.

Parasitemias in monkeys were monitored by Giemsa-stained blood films. Counts were based on a minimum of20,000 erythrocytes. The following criteria were used fordetermining when to drug treat experimental animals: para-sitemia above 5% (if predominantly ring stages) or above 4%(if late stages prominent); reticulocytes above 20% of thetotal erythrocytes; hematocrit below 20; anorexia exceeding24 h; or moribund animal. Monkeys were treated with a totalof 55 to 65 mg of amodiaquine (no. BG58821/2977 AK;Walter Reed Army Institute of Research) given intramuscu-larly in water (50 mg/ml) over 3 to 4 days. Recrudescentmonkeys were treated when their parasitemias exceeded3.5%.

Microscopy. IFAs were performed as described previously(33), with antigen slides prepared with FCR-3 wild-typecultured parasites. Dilutions of heat-inactivated sera from1:20 to 1:10,000 were tested, and titers reported are thereciprocal of the last dilution giving positive fluorescence ofmerozoites within schizonts. For all assays, the controlnormal serum was a pool prepared from 10 naive Aotusmonkeys, screened for negative response to P. falciparumby IFA and to human erythrocyte surface proteins in ahemagglutination assay. Slides were examined by epifluo-rescence microscopy.Transmission electron microscopy (TEM) and scanning

electron microscopy (SEM) were performed on both thewild-type FCR-3 and the D3 clone parasites from cultures inhuman O+ erythrocytes, from cultures in Aotus erythrocytesbefore infection of animals, from isolates from all monkeysexhibiting parasitemias, and from isolates from infectedmonkeys which were placed back into culture in Aotuserythrocytes for 1 to 14 days. Methods for fixation andprocessing for TEM were as described previously (15). TEM

VOL. 47, 1985

on May 29, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

762 LANGRETH AND PETERSON

samples were examined in a JEOL 100CX or Philips 400transmission electron microscope, operating at 60 kV.For SEM, cultured cells were routinely fixed by a modi-

fication of the procedure of Arnold et al. (2). Cells werewashed with culture medium (RPMI 1640 medium) withoutserum, prefixed with 0.1% glutaraldehyde in RPMI 1640medium (pH 7.0) for 15 min at room temperature, and fixedwith 2.0% glutaraldehyde in phosphate-buffered saline (pH7.3) for 45 min at room temperature. Cells were resuspendedin RPMI 1640 medium and allowed to settle for 30 min ontocover slips which had been pretreated with 0.1% poly-L-ly-sine (pH 7.3) for 30 to 60 min. Unattached cells were washedoff with RPMI 1640 medium, and the cover slips were rinsedwith distilled water, dehydrated with increasing concentra-tions of ethanol, and transferred to Freon 113. Samples werethen critical-point dried from CO2 with a Polaron critical-point drying apparatus or air dried from Freon 113 in avacuum dessicator at room temperature (20). After thesamples were coated with gold-palladium, they were viewedwith a JEOL JSM-35 scanning electron microscope at anaccelerating voltage of 25 kV and recorded with PolaroidP/N 55 film. Samples of blood from infected animals werecollected in heparinized blood capillary tubes and processedfor SEM by the protocol described for culture cells exceptthat samples were usually held overnight at 4°C in fixative.For samples which contained no late-stage parasites, cellswere cultured overnight to allow parasite maturation andwere then processed for SEM.To identify parasitized cells for SEM analysis, a light

microscopy step was included in sample preparation. Afterfixation and attachment to glass cover slips, samples wererinsed with RPMI 1640 medium and stained with Giemsastock solution diluted 1:5 with phosphate buffer (pH 6.8) for10 min at room temperature, in a modification of the proce-dure of Wetzel et al. (32). After the cover slips were rinsedwith RPMI 1640 medium, they were inverted over a buffer-filled well on a microscope slide, and cells with Giemsa-stained parasites in areas near an orientation pattern werephotographed for later comparison with scanning electronmicrographs of the same cells. Routine processing for SEMthen was continued.

RESULTSPathogenicity. Both the wild-type strain FCR-3 and the D3

clone parasites were cultured continously in Aotus erythro-cytes for 1 month before infection of monkeys. The eryth-rocytes of all monkeys in these experiments supported thegrowth in vitro of both wild-type and D3 parasites understandard culture conditions (23).Course of infection in vivo. Data comparing the wild-type

(K+) and clone D3 (K-) infections in intact and splenecto-mized monkeys are presented in Table 1. All monkeys whichreceived the wild-type FCR-3 parasites developed virulentinfections (high parasitemias or reticulocytes > 20% totalred cells or both). Throughout their infections ring-stageparasites were the predominant form in the peripheral circu-lation. Monkeys 8443 and 7921 went into reticulocyte crisis(reticulocytes > 20%, 5% normoblasts, hematocrit < 20)when their parasitemias were 2.4 and 0.8%, respectively.The monkeys infected with wild-type FCR-3 required drugtreatment according to the preestablished criteria (seeabove). Two of the animals which received wild-type FCR-3subsequently recrudesced (217J splenectomized and 115),exhibiting low transient parasitemias.

After receiving a single low dose of the K- D3 clone, thesplenectomized monkey 209J developed a virulent infection

and was drug treated. This animal recrudesced three timesand also was drug treated at the third recrudescence. Oneyear after recovery from the third recrudescence, monkey209J again was infected with a low dose (5 x 105) of the K-D3 clone. The monkey became patent 12 days later. Theparasitemia peaked at 0.17% on day 17 postinfection andwas negative by day 27. One month later the monkey againwas challenged with a higher dose (1 x 106) ofD3 but did notbecome patent. Two additional splenectomized animals,naive to malaria, received a single low dose (5 x 105) of theK- D3 clone and developed virulent infections. One animal(5341) had a peak parasitemia of 7.6% and was drug treated.The second animal (WR 135) had parasitemias in excess of1% for 10 days but did not exceed 3% and was not drugtreated. Reticulocytes exceeded 20% in this animal immedi-ately after clearance of the parasites, and the animal subse-quently became chronic at low parasitemia (<0.05%).During the first experiment, four intact monkeys received

multiple doses of the K- D3 clone. Three animals (WR 159,6902, and 8490) never became patent with D3, and the fourth(5342) showed transient parasitemias never exceeding 0.02%.After receiving two doses of D3 parasites, monkeys 8490 and6902 were challenged with the wild-type FCR-3. Althoughthey developed parasitemias of 1.2 and 3.6%, respectively,the animals controlled their infections without drug treat-ment. Their hematocrits and reticulocyte percentages re-mained within acceptable ranges throughout their infections.In contrast, the naive monkey 5335 had to be drug treatedwhen its parasitemia reached 6.5%.

In a second experiment, three intact monkeys (114, WR157, and 216J) received multiple high doses of clone D3. Twoanimals (114 and WR 157) never became patent, whereasmonkey 216J exhibited a parasitemia (0.005%) on only 3days. Monkeys 114 and 216J and a naive animal (5336) weresubsequently challenged with the wild-type strain. All threebecame patent 9 to 10 days later and were drug treated 16 to17 days postchallenge when their parasitemias were 5.9, 5.9,and 5.8%, respectively. The naive animal (5336) recrudescedand was again drug treated (peak parasitemia, 3.7%). Para-sitemia did not recur in monkey 114. The recrudescentparasitemia of monkey 216J did not exceed 0.20%.An analysis was made of the proportion of late stages

present in the peripheral circulation of every animal in theseexperiments on all days when their parasitemia exceeded1%. These data are summarized in the form of a histogram(Fig. 1) and include values for eight FCR-3-infected animals(seven intact and one splenectomized) and three splenecto-mized D3-infected animals. None of the seven intact D3-infected animals ever had parasitemias exceeding 0.02%,and therefore do not provide data to be included in thissummary. Ring forms were predominant in FCR-3-infectedanimals, with >87% of the data points showing 20% or fewerlate stages (Fig. 1). In no instance did late stages exceed 50%in these animals. In contrast, the proportion of late stages inthe peripheral circulation of the D3-infected, splenectomizedanimals was frequently high, with approximately 45% of thedata points showing >50% late stages.

Antibody response. Before infection, the serum antibodytiters against P. falciparum FCR-3 were negative for allmonkeys as determined by IFA. Monkeys infected with P.falciparum FCR-3 wild-type strain in Aotus cells from cul-ture developed humoral immune responses as detected byIFA (Table 2). Monkeys infected with clone D3 also devel-oped antibody titers (Tables 2 and 3). The response to D3was dose dependent in that the monkeys receiving higheramounts of D3 (5 x 106) rapidly produced higher antibody

INFECT. IMMUN.

on May 29, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

P. FALCIPARUM IN COLOMBIAN OWL MONKEYS 763

60-t

0-

c

z

0'

U. 20

0 50

LATE STAGES (%)

100

FIG. 1. Frequency distribution of percentage of total number ofdays when infected animals had late-stage parasites (trophozoitesand schizonts) as the indicated percentage of their total IRBCs.Solid bars are total percentages for eight animals infected withFCR-3 wild type, seven intact and one splenectomized. Patternedbars are total percentages for three splenectomized animals infectedwith D3 K-. No intact animals infected with D3 developed signifi-cant parasitemias.

titers (114, 216J, and WR 157). Monkeys 8490 and 6902,which were infected twice with D3, followed by a wild-typechallenge, produced high IFA titers 1 month after challenge(Table 3). Monkeys 114 and 216J, which received three largedoses of D3 developed high IFA titers before challenge withwild-type parasites (Table 3).

Stability of phenotype. The surface morphologies of eryth-rocytes infected with wild-type and D3 parasites were stud-ied by TEM and SEM. These surface morphologies ap-peared to be the same, whether the parasites were culturedin human or Aotus erythrocytes. In TEM, most trophozoite-and schizont-IRBCs in wild-type FCR-3 cultures had dis-torted surfaces and knob structures. All trophozoite- andschizont-IRBCs observed in D3 clone cultures had relativelysmooth surfaces and lacked knobs as determined by TEM. Afew (<10%) late-stage IRBCs from FCR-3 wild-type cultures

TABLE 2. IFA titers on monkeys with virulent infections'

Titer

Monkey Infecting Preinfec- At drug treat- At 1 moParasites Ptionfec- ntdu treat- posttreat-

ment

115b Wild type (K+) <20 20 (21) 320217Jb.C Wild type (K+) <20 <20 (21) 1605335 Wild type (K+) <20 80 (25) 3205336b Wild type (K+) <20 40 (17) 6407921 Wild type (K+) <20 640 (31) 25608443 Wild type (K+) <20 320 (30) 640209Jb.C D3 clone (K-) <20 160 (33) 640a All animals were drug treated at peak parasitemia except monkeys 7921

and 8443, who were treated when reticulocytes > 20%o and hematocrits < 20.b Had recrudescences.c Monkeys 217J and 209J were splenectomized.

had the D3 surface morphology as determined by TEM asexpected, since the D3 clone was derived from the FCR-3wild-type strain.The IRBC surface morphology of the parasites isolated

from each monkey at peak parasitemias retained the samemorphology as that of the parasites in the infecting dose, asdetermined by TEM. The presence or absence of a spleenhad no observed effect on IRBC surface morphology. CloneD3-parasitized erythrocytes isolated from monkeys 5342(intact) and 209J (splenectomized) remained knobless,smooth, and undistorted (Fig. 2A). Isolates from the variousrecrudescences of monkey 209J also retained typical D3morphology. Parasitized erythrocytes isolated from mon-keys infected with FCR-3 wild type (intact 115, 5335, 7921,and 8443 and splenectomized 217J) continued to exhibit theheterogeneous surface morphologies characteristic of thisstrain. Most late-stage IRBCs showed knob-bearing, dis-torted surfaces (Fig. 2D). Isolates from monkeys 6902 and8490 after challenge with FCR-3 wild type were of K+morphology.SEM, in combination with Giemsa-stained light micros-

copy, also was used to assess the surface morphology ofcells infected with late-stage FCR-3 wild-type and D3 para-sites.Knobs were observed on the surface of most late-stage

FCR-3 wild-type infected cells when the parasites werecultured in either human or Aotus erythrocytes, and therelative shape and frequency of the knobs appear similar inboth host cell types. Samples isolated from each FCR-3-wild-type-infected animal at the peak parasitemia, before drugtreatment, showed the same predominantly knobby morphol-ogy as that seen in the culture material. A human O cellinfected with a trophozoite of the FCR-3 wild-type strain isshown in Fig. 2E and is a typical example of the K+morphology of late-stage infected cells. Figure 2F shows atrophozoite-infected Aotus erythrocyte isolated from ananimal after infection with FCR-3 wild type.

Erythrocytes infected with late stages of clone D3 para-sites were uniformly knobless and relatively undistorted.Many trophozoite-infected cells appear to have a single,large depressed area on the surface. This morphology issimilar in cultures of human and Aotus erythrocytes and inIRBCs isolated from splenectomized animals. The knoblessmorphology typical of this clone is seen in Fig. 2B and C.Figure 2B shows a D3 trophozoite-infected human O+erythrocyte in culture, and Fig. 2C shows a trophozoite-in-fected Aotus erythrocyte isolated from an animal afterinfection with clone D3.

TABLE 3. IFA titers on FCR-3 clone D3-infected intact monkeysTiter

Monkey Preinfec- At 1 mo At 1 mo At 1 motion postinfec- postinfec- postinfec-tion no. 1 tion no. 2 tion no. 3

Low dose of D3WR159 <20 20 80 1605342a <20 80 640 6408490 <20 20 20 5,120b6902 <20 20 20 640"

High dose of D3114 <20 320 1,280 2,560216Ja <20 2,560 1,280 2,560WR157 <20 1,280 2,560 2,560a Transient patency.b Infection number 3 was with wild-type parasites.

VOL. 47, 1985

on May 29, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

764 LANGRETH AND PETERSON

FIG. 2. Transmission and scanning electron micrographs of cells containing D3 K- (A, B, and C) and FCR-3 wild-type K+ (D, E, and F)late-stage parasites. (A) K- schizont-IRBC from the third recrudescence of the splenectomized monkey which received clone D3. (B) HumanO+ erythrocyte in culture containing a D3 K- trophozoite. (C) D3 trophozoite-infected Aotus erythrocyte from an infected monkey. (D) AK+ schizont-IRBC from an intact Aotus monkey infected with FCR-3 wild type; the isolate was cultured for 1 day to mature the parasites.(E) Human O+ erythrocyte in culture containing an FCR-3 wild-type trophozoite. (F) Aotus erythrocyte taken from an infected monkeycontaining an FCR-3 wild-type trophozoite. Arrowheads indicate knobs on the cell surface. Bars represent 1 ,um.

DISCUSSION

These studies in Colombian Aotus monkeys directly com-pare the pathogenicity of a heterogeneous wild-type K+strain of P. falciparum, FCR-3, and the cloned D3 K- linederived from FCR-3. This cloned line lacks the surfacedistortions and knobs of erythrocytes infected with theparental strain. The wild-type K+ strain caused virulentinfections in both intact and splenectomized monkeys, andlate-stage IRBCs were sequestered from the peripheral cir-culation regardless of whether the host monkey was sple-nectomized. The cloned K- line was virulent in splenecto-mized animals, and late-stage IRBCs did not sequester inthese monkeys. In intact animals, the K- cloned line wasessentially avirulent; parasitized cells were seen only on rareoccasions, and these few were ring forms.Two major factors appear to play crucial roles in deter-

mining virulence in these experiments: the surface of late-stage IRBCs is K+ or K-, and the animal is splenectomizedor intact. Our results suggest the following: (i) the presenceor absence of knobs determines whether late-stage parasiteswill sequester or circulate, and (ii) if late stages do circulate,the presence or absence of a spleen determines whetherthese circulating late-stage parasites will be eliminated orcontribute to a virulent infection. The mechanism of thisvery rapid splenic destruction of K- IRBCs is probably not

antibody mediated but is based on recognition of nondeform-ability such as occurs with senescent erythrocytes.

Cranston et al. (6) reported that erythrocytes infected withlate-stage P. falciparum parasites (trophozoites and schiz-onts) lose the ability to deform, as measured by sheer stressin vitro. They compared the cloned D4 line of FCR-3 (31),which is K-, with two K+ strains. They found that loss ofdeformability is related to parasite maturation in both K+and K- strains. Since the spleen is known to recognize anddestroy nondeformable erythrocytes, these authors sug-gested that the loss of deformability is the determining factorin splenic elimination of late-stage P. falciparum IRBCs.The roles of the spleen in malaria are varied and complex.

It is a major site of parasite and erythrocyte destruction (26,34), and both specific and nonspecific immune mechanismshave been demonstrated (24, 27). In addition, David et al. (7)and Hommel et al. (9, 10) have suggested that the spleen maymodulate antigenic expression, sequestration, and in vitrocytoadherence properties of erythrocytes infected with late-stage P. falciparum. These studies compared two lines ofK+ parasites which were sequentially passaged in severalsquirrel monkeys.The surface and knob morphologies of the wild-type- and

D3 K--IRBCs did not change throughout the course of ourexperiments. Thus, these major morphological structuresappear to be stable in vivo as well as in vitro. Variation of

INFECT. IMMUN.

on May 29, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

P. FALCIPARUM IN COLOMBIAN OWL MONKEYS 765

surface IRBC antigens, such as occurs in the Plasmodiumknowlesi-rhesus model, may be a cause of recrudescences inprimate malarias (5). Antigenic variation of IRBC surfacecomponents has recently been reported in P. falciparuminfections in squirrel monkeys in a cloned parasite line and inrecrudescent populations of the Indochina I strain (9). Strain-specific high-molecular-weight P. falciparum IRBC surfaceantigens which are related to in vitro cytoadherence alsohave been reported (19). The relationship of these antigensto the IRBC surface antigens in D3 K- cells and to loss ofknobs is not known. Analysis of surface antigenic determi-nants in original and recrudescent animal isolates of the D3K- line is in progress.Although parasites of the cloned D3 knobless line have

reduced pathogenicity in intact monkeys, they induced ahumoral immune response, as measured by IFA titers, andthis response was dose dependent (Table 3). To determinewhether the immune response induced by D3 was protec-tive, monkeys were subsequently challenged with homolo-gous D3 or heterologous FCR-3 wild-type parasites. Priorexposure to the D3 clone provided solid protection for thesplenectomized monkey 209J when later challenged withhomologous D3 parasites. Two experiments demonstratedpartial protection after heterologous FCR-3 wild-type chal-lenge. In experiment 1, two D3 monkeys eliminated theirFCR-3 wild-type infections without drug treatment. In ex-periment 2, the D3 monkeys received higher doses of D3parasites and developed higher antibody titers as measuredby IFA but were less protected than the monkeys in exper-iment 1 when challenged with wild-type FCR-3. The mon-keys had to be drug treated to control their initial parasit-emias from wild-type infection, but only the control naiveanimal (5336) had to be drug treated a second time becauseof a high recrudescent parasitemia. Due to the extremelylimited supply of Colombian owl monkeys, animals in exper-iments 1 and 2 differed by sex and karyotype (Table 1), andthese variables may have contributed to the observed differ-ences in protection. The genetics of the immune response inAotus monkeys have not yet been well studied. Partial ratherthan complete protection is consistent with the suggestionthat the wild-type strain is both antigenically as well asmorphologically more heterogeneous than the D3 clonedline.The adherence of late-stage K+ parasitized erythrocytes

to capillary endothelium via knobs contributes to the path-ogenesis of falciparum malaria. Sequestered parasites ob-struct local blood circulation, leading to local anoxia andischemia. In addition, these sequestered parasites avoidattack by both specific and nonspecific immunological mech-anisms, including splenic destruction. The use of knob-spe-cific antisera to neutralize knob function may be an addi-tional approach to controlling the pathogenicity of thisdisease. Parasite sequestration has been reversed in vivoand in K+ IRBC cytoadherence in vitro (7) with squirrelmonkey immune serum. Antibodies to knobs have beenidentified in the sera of immune owl monkeys (16). Thus,although most attempts to develop immunity to falciparummalaria use merozoite antigens, knob-endothelium bindingcomponents also are possible candidate antigens for a mul-tivalent vaccine.

ACKNOWLEDGMENTS

We thank William Trager for providing us with an early isolate ofwild-type FCR-3 and the D3 cloned line and William Ellis forantimalarial drugs. We are indebted to Donald Krogsted and WilliamCollins for helpful discussions and advice and to Robert Reese for

karyotyping several of the monkeys. The technical assistance of G.Patrick Riordan, Susan Kenney, and Ina Ifrim is gratefully acknowl-edged.

This work was supported by funds from the U.S. Agency forInternational Development (PASA no. DZ/DSB-0453-4-80) and theUnited Nations Development Program/World Bank/World HealthOrganization Special Programme for Research and Training inTropical Diseases (Project no. 810022).

LITERATURE CITED1. Aikawa, M., J. R. Rabbege, and B. T. WeUde. 1972. Junctional

apparatus in erythrocytes infected with malarial parasites. Z.Zellforsch. U. Mikroskop. Anat. 124:72-75.

2. Arnold, J. D., A. E. Berger, and 0. L. Allison. 1971. Someproblems of fixation of selected biological samples for SEMexamination. Scanning Electron Microscopy/1971:251-256.

3. BarnweUl, J. W., R. J. Howard, H. G. Coon, and L. H. Miller.1983. Splenic requirement for antigenic variation and expressionof the variant antigen on the erythrocyte membrane in clonedPlasmodium knowlesi malaria. Infect. Immun. 40:985-994.

4. Barnwell, J. W., R. J. Howard, and L. H. Miler. 1982. Alteredexpression of Plasmodium knowlesi variant antigen on theerythrocyte membrane in splenectomized rhesus monkeys. J.Immunol. 128:224-226.

5. Brown, K. N., and I. N. Brown. 1965. Immunity to malaria:antigenic variation in chronic infections of Plasmodium knowl-esi. Nature (London) 208:1286-1288.

6. Cranston, H. A., C, W. Boylan, G. L. Carroll4 S. P. Sutera, J. R.Williamson, I. Y. Gluzman, and D. J. Krogstad. 1984. Plasmo-dium falciparum maturation abolishes physiologic red cell de-formability. Science 223:400-403.

7. David, P. H., M. Honunel, L. H. Miller, I. Udeinya, and L. D.Oligino. 1983. Parasite sequestration in Plasmodiumfalciparummalaria: spleen and antibody modulation of cytoadherence ofinfected erythrocytes. Proc. Natl. Acad. Sci. U.S.A.80:5075-5079.

8. Hadley, T. J., J. H. Leech, T. J. Green, W. A. Daniel, M.Wahigren, L. H. Miller, and R. J. Howard. 1983. A comparisonof knobby (K') and knobless (K-) parasites from two strains ofPlasmodium falciparum. Mol. Biochem. Parasitol. 9:271-278.

9. Hommel, M., P. H. David, and L. D. Oligino. 1983. Surfacealterations of erythrocytes in Plasmodium falciparum malaria.J. Exp. Med. 157:1137-1148.

10. Hommel, M., P. H. David, L. D. Oligino, and J. R. David. 1982.Expression of strain-specific surface antigens on Plasmodiumfalciparum-infected erythrocytes. Parasite Immunol. 4:409-419.

11. Jensen, J. B., and W. Trager. 1978. Plasmodium falciparum inculture: establishment of new strains. Am. J. Trop. Med. Hyg.27:743-746.

12. Kilejian, A. 1979. Characterization of a protein correlated withthe production of knob-like protrusions on membranes of eryth-rocytes infected with Plasmodiumfalciparum. Proc. Natl. Acad.Sci. U.S.A. 76:4650-4653.

13. Kilejian, A. 1980. Homology between a histidine-rich proteinfrom Plasmodium lophurae and a protein associated with knob-like protrusions on membranes of erythrocytes infected withPlasmodium falciparum. J. Exp. Med. 151:1534-1538.

14. Kilejian, A., A. Abati, and W. Trager. 1977. Plasmodiumfalciparum and Plasmodium coatneyi: immunogenicity of "'knob-like protrusions" on infected erythrocyte membranes. Exp.Parasitol. 42:157-164.

15. Langreth, S. G., J. B. Jensen, R. T. Reese, and W. Trager. 1978.Fine structure of human malaria in vitro. J. Protozool.25:443-452.

16. Langreth, S. G., and R. T. Reese. 1979. Antigenicity of theinfected-erythrocyte and merozoite surfaces in falciparum ma-laria. J. Exp. Med. 150:1241-1254.

17. Langreth, S. G., R. T. Reese, M. R. Motyl, and W. Trager. 1979.Plasmodium falciparum: loss of knobs on the infected erythro-cyte surface after long-term cultivation. Exp. Parasitol.48:213-219.

18. Leech, J. H., J. W. Barnwell, M. Aikawa, L. H. Miller, and R. J.Howard. 1984. Plasmodium falciparum malaria: association of

VOL. 47, 1985

on May 29, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

766 LANGRETH AND PETERSON

knobs on the surface of infected erythrocytes with a histidine-rich protein and the erythrocyte skeleton. J. Cell Biol.98:1256-1264.

19. Leech, J. H., J. W. Barnwell, L. H. Miller, and R. J. Howard.1984. Identification of a strain-specific malarial antigen exposedon the surface of Plasmodiumfalciparum-infected erythrocytes.J. Exp. Med. 159:1567-1575.

20. Liepins; A., and E. deHarven. 1978. A rapid method for celldrying for scanning electron. Scanning Electron Micros-copy/1978 11:37-43.

21. Luse, S. A., and L. H. Miller. 1971. Plasmodiurn falciparummalaria: ultrastructure of parasitized erythrocytes in cardiacvessels. Am. J. Trop. Med. Hyg. 20:655-60.

22. Miller, L. H. 1969. Distribution of mature trophozoites andschizonts of Plasmodium falciparum in the organs of Aotustrivirgatus, the night monkey. Am. J. Trop. Med. Hyg.18:860-865.

23. Peterson, E., S. G. Langreth, and S. Kenney. 1984. Long termcultivation ofPlasmodiumfalciparum in Aotus trivirgatus eryth-rocytes. Am. J. Trop. Med. Hyg. 33:331-335.

24. Qdiinn, T. C., and D. J. Wyler. 1979. Intravascular clearance ofparasitized erythrocytes in rodent malaria. J. Clin. Invest.63:1187-1194.

25. Reese, R. T., S. G. Langreth, and W. Trager. 1979. Isolation ofstages of the human parasite Plasmodium falciparum fromculture and from animal blood. Bull. W.H.O. 57(Suppl. 1):53-61.

26. Seed, T. M., and J. P. Kreier. 1980. Erythrocyte destructionmechanisms in malaria, p. 1-46. In J. P. Kreier (ed.), Malaria,

vol. 2. Academic Press, Inc., New York.27. Shear, H. L., R. S. Nussenzweig, and C. Bianco. 1979. Immune

phagocytosis in murine malaria. J. Exp. Med. 149:1288-1298.28. Trager, W., and.J. B. Jensen. 1976. Human malaria parasites in

continuous culture. Science 193:673-675.29. Trager, W., H. N. Lanners, H. A. Stanley, and S. G. Langreth.

1983. Immunization of owl monkeys to Plasmodium falciparumwith merozoites from cultures of a knobless clone. ParasiteImmunol. 5:225-236.

30. Trager, W., M. A. Rudzinska, and P. C. Bradbury. 1966. Thefine structure of Plasmodium falciparum and its host erythro-cytes in natural malarial infections in man. Bull. W.H.O.35:883-885.

31. Trager, W., M. Tershakovee, L. Lyandvert, H. Statley, N.Lanners, and E. Gubert. 1981. Clones of the malaria parasitePlasmodium falciparum obtained by microscopic selection:their characterization with regard to knobs, chloroquine sensi-tivity and formation of gametocytes. Proc. Natl. Acad. Sci.U.S.A. 78:6527-6530.

32. Wetzel, B., B. W. Erickson, Jr., and W. R. Levis. 1973. Theneed for positive identification of leukocytes examined by SEM.Scanning Electron Microscopy/1973 III:535-542.

33. World Health Organization Memorandum. 1974. Serologicaltesting in malaria. Bull. W.H.O. 50:527-534.

34. Wyler, D. J., C. N. Oster, and T. C. Quinn. 1979. The role of thespleen in malaria infections, p. 183-204. In The role of thespleen in immnunology of parasitic diseases, Tropical DiseaseResearch Series No. 1. Schabe and Company, Basel.

INFECT. IMMUN.

on May 29, 2018 by guest

http://iai.asm.org/

Dow

nloaded from