Clin Exp Immunol 1995; 100:239-245

Dichotomy of the T cell response to Leishmania antigens in patients sufferingfrom cutaneous leishmaniasis; absence or scarcity of Thl activity is associated

with severe infections

A. GAAFAR*t, A. KHARAZMIt, A. ISMAIL¶, M. KEMPt, A. HEYt, C. B. V. CHRISTENSENt,M. DAFALLA§, A. Y. EL KADAROt, A. M. EL HASSAN§ & T. G. THEANDERt *Institute of TropicalMedicine, MRC, Khartoum, Sudan, tCentre for Medical Parasitology, Institute for Medical Microbiology and Immunology,University of Copenhagen and Department of Infectious Diseases and Clinical Microbiology, National University Hospital of

Denmark (Rigshospitalet), Copenhagen, Denmark, Departments of tMedicine and §Pathology, Faculty of Medicine, University ofKhartoum, and ¶Faculty of Medicine, Wadi El Niel University, Khartoum, Sudan

(Acceptedfor publication 10 January 1994)

SUMMARY

The T cell response was studied in 25 patients suffering from cutaneous leishmaniasis caused byLeishmania major with severe (n = 10) and mild (n = 15) disease manifestations. Peripheral bloodmononuclear cells (PBMC) from the patients were activated by sonicates of Leishmaniapromastigotes (LMP) and amastigotes (LDA), and the surface protease gp63. The proliferativeresponses to Leishmania antigens were lower in patients with severe disease than in patients withmild disease (P = 0 01-0-05), and such a difference was not observed in the response to purifiedprotein derivative of tuberculin (PPD) or tetanus toxoid (TT). LMP-induced interferon-gamma(IFN--y) production was lower in patients with severe than in patients with mild disease (P < 0 05).When the IL-4 and IFN-y responses of each patient were considered, two response patterns wereobserved in the cultures activated by the Leishmania sonicates. One response pattern wascharacterized by high production of IFN-y without production of IL-4 (a Thl-like pattern), theother was characterized by low IFN-y levels which in most cases were associated with IL-4production (not a Th I -like pattern). These patterns could not be distinguished when the cells fromthe same donors were stimulated by TT and PPD. The percentages of patients with a Thl-likeresponse pattern after stimulation by LMP in patients with severe and mild disease manifestationswere 30% and 80%, respectively. This difference was statistically significant (P = 0.034).

Keywords cutaneous leishmaniasis human T cell subsets

INTRODUCTION

Members of the genus Leishmania are protozoan parasitesinfecting and multiplying inside the phagolysosomes of macro-phages in the mammalian host. Diseases caused by Leishmaniaparasites, the leishmaniases, constitute serious health problems,which affect millions of people in the tropical and subtropicalparts of the world [1]. The clinical manifestations of leish-maniasis range from self-healing cutaneous ulcers (cutaneousleishmaniasis) to fatal systemic disease (visceral leishmaniasis,kala-azar) depending on the infecting parasite species and theimmune response of the host [2].

The most common clinical manifestation of L. major issimple cutaneous leishmaniasis (CL) characterized by an

Correspondence: Thor Theander, Centre of Medical Parasitology,Panum Building 24.2, Blegdamsvej 3, DK-2200 Copenhagen N,Denmark.

ulcerated lesion(s) which heals spontaneously leavingdepressed scar(s) and long-lasting immunity [3]. However, insome patients complications may occur as the parasite meta-stasizes through the lymphatics to the lymph node, leading tothe formation of subcutaneous nodules and/or enlargement ofthe regional lymph nodes [4,5]. Severe disfigurement mightresult from large ulcers. Recently, an epidemic of CL hit thecentral Sudan along the Nile River in Greater Khartoum andspread to the North [6,7]. The disease was caused by L. majorzymodeme LON-1 [8,9].

In experimental murine models the outcome of L. majorinfection depends on the activation of one of the two subsets ofT helper/inducer cells, Thl and Th2 [10]. Thl cells produceinterferon-gamma (IFN--y), IL-2, and lymphotoxin, whereasTh2 cells secrete IL-4, IL-5, IL-6, and IL-10. Functionally, Thlcells have mainly been associated with DTH reaction, whereasTh2 cells have been related to B cell help. When L. major

239

A. Gaafar et al.

infection in mice results in activation of Thl cells the micerecover and are resistant to reinfection, whereas activation ofTh2 cells leads to progressive disease and death of the animals[11,12].

We have previously shown that both Thl- and Th2-like Tcell reactivity exist in individuals who have been drug curedfrom visceral leishmaniasis caused by L. donovani [13,14]. Th2-like cytokine production profiles have been demonstrated in Tcells from peripheral blood [15] and bone marrow [16] ofpatients suffering from kala azar. We have also shown thatthe Leishmania-specific T cells of individuals recovered fromuncomplicated CL were of the Thl-like subset [17]. In thisstudy we describe the cytokine production pattern in patientssuffering from mild and severe CL, and report that the presence

in peripheral blood of Leishmania-specific T cells with ThI-likecytokine production profile is associated with mild disease,whereas the absence or scarcity of Thl activity and thepresence of Th2 activity is associated with severe disease.

PATIENTS AND METHODS

DonorsAfter giving informed consent 25 patients with CL presentingto the Tropical Diseases Hospital, Omdurman, Sudan, were

included in the study. From each patient a clinical history wasobtained and a clinical examination was performed. Thediagnosis of CL was made on the demonstration of amasti-gotes on slit smears, impression smears of biopsies and/orhistological sections. On the basis of the clinical history andthe clinical picture on admission the patients were divided intosevere and mild cases. Patients were categorized as severe if: (i)they presented with a lesion of more than 40mm in diameter,and had a clinical history of more than 6 months; and/or (ii)had subcutaneous nodules at admission. Nine normal Danishvolunteers were included as controls not exposed to Leishmaniaantigens.

TreatmentPatients with single small lesions were not treated, as theseusually healed spontaneously. Patients with multiple lesions or

subcutaneous nodules were treated with IV sodium stiboglu-conate (Pentostam) at a dose of 10mg/kg per day for 3 weeks,although some patients with severe disease needed severalcourses of Pentostam before healing was achieved.

Blood sampling and peripheral blood mononuclear cellpreparationBlood was collected by venipuncture into heparinized vacutai-ners (Becton Dickinson Ltd., Rutherford, NJ) at the day thepatients presented to the Clinic. Peripheral blood mononuclearcells (PBMC) were isolated by Lymphoprep (Nyegaard, Oslo,Norway) density centrifugation, washed twice in RPMI 1640supplemented with 10% heat-inactivated fetal bovine serum

(FBS), and frozen at - 1960C in RPMI 1640 supplemented with10% dimethylsulphoxide and 20% FBS. The cells were frozenby a manually operated version of a recently described devicefor gradient freezing of cells under field conditions [18] andstored and transported in liquid nitrogen. Before use, the cellswere rapidly thawed and washed. The viability of the cells wasconfirmed by trypan blue exclusion.

AntigensA sonicate preparation from L. major (Baringo strain;Kenya-MHOM/KE/87/NLB 455) was made as describedpreviously [19]. The crude L. major preparation was used in a

final concentration of 70 Mg/ml.Crude soluble amastigote antigen was prepared from L.

donovani amastigotes grown in the human monocyte-like cellline, U937 (Batch no. F-8918; American Type Culture Collec-tion, Rockville, MD) as described [20]. Briefly, U937 cells weregrown. Before infection the cells were transferred to tissueculture plates (Nunc, Roskilde, Denmark) at 1 x 106 cells/mland stimulated by phorbol 12-myristate 13-acetate (PMA),causing cells to differentiate and grow as a monolayer. U937cells were infected for 24 h at a parasite:cell ratio of 10:1followed by washing to remove extracellular parasites. After6 days of incubation at 37°C, the U937 cells were harvested.Amastigote purification was performed as described [21] byhomogenization of the cells, followed by differential filtrationthrough polycarbonate filters of decreasing pore size (Isopore;Millipore, Bedford, MA). Purified amastigotes were washedthree times before sonication for 5 x 45 s (MSE-150; Soniprep,Crawley, UK). Soluble antigens were collected after centrifuga-tion (30 min, 30 000g, 4°C), and stored at -80°C until used at afinal protein concentration of 13 Mg/ml.

The abundant surface protein gp63 was purified fromL. major promastigotes as described [22]. The proteaseactivity of gp63 was inactivated by heating to 70°C for20 min and the protein was used at a final concentration of10 jig/ml.

Purified protein derivative of tuberculin (PPD) and tetanustoxoid (TT) were purchased from Statens Seruminstitut(Copenhagen, Denmark) and used at 12jig/ml and 3 jig/ml,respectively.

Lymphoproliferation and cytokine productionCells were cultured in RPMI 1640 with 10mm HEPES, 20 Upenicillin and 20 ,ig streptomycin per ml supplemented by 15%heat-inactivated, pooled, normal human serum (NHS). Thecultures were incubated at 37°C in a humidified atmospherecontaining 5% CO2. PBMC were incubated with antigen at6-6 x 105 cells/ml in volumes of 170,u1 in round-bottomedmicroculture plates (Nunc). The cultures were incubated for 7days and pulsed with 20 kl/well of 3H-thymidine (New EnglandNuclear, Boston, MA) (185 MBq/ml) for the last 24 h ofincubation. The culture supernatants were recovered andstored at -20'C for later determination of IFN-7y. The cellswere harvested onto glassfibre filters and the incorporation of3H-thymidine into DNA was determined by a Matrix f3-counter(Packard, Greve, Denmark). All proliferation assays were

performed in triplicate. For each set of samples the medianct/min was recorded. The antigen response was expressed as thedifference between antigen-stimulated cultures and culturesincubated without antigen. A response was considered measur-

able if the stimulation index (SI; ct/min in antigen-stimulatedculture/ct/min in unstimulated culture) was > 2-5.

For the measurement of IL-4 release by antigen-stimulatedcultures ofPBMC, parallel cultures were carried out for 6 daysand then pulsed with 1 sM ionomycin and 50 ng/ml PMA (bothfrom Sigma, St Louis, MO) for 24 h before the culture super-natant from triplicate wells was harvested [23].

240

Human T cell response to Leishmania antigens

Cytokine measurementsIFN-y and IL-4 in the culture supernatants pooled from thetriplicate cultures were measured by ELISA as described else-where [19,23]. The detection ranges of the ELISAs were: 1-66U/ml for IFN--y (specific activity of the reference protein:2 5 x 108 U/mg) and 30-2000 pg/ml for IL-4. Antigen-inducedcytokine production was calculated as the difference betweenantigen-stimulated and unstimulated cultures. A response was

considered measurable if the SI (cytokine level in supernatantfrom antigen-stimulated culture/cytokine level in supernatantfrom cultures without antigen) was > 2-0.

Statistical analysisThe magnitude of the responses in patients with mild and severe

disease were compared by the Mann-Whitney two-samplerank sum test. The Fisher exact test was used to compare thefrequency of Thl and non-Thl responders in patients withsevere and mild disease. P < 0.05 was considered significant.

RESULTS

ClinicalfeaturesMost patients were from villages of Khartoum region or fromthe outskirts ofOmdurman City. The patients were divided intoa group with mild and a group with severe disease according tothe size of the lesion, duration of the disease and the presence ofsubcutaneous nodules. Table 1 summarizes the clinical infor-mation of the two groups. The groups were comparable withregard to age and number of lesions. The lesions were typicallynodular or noduloulcerative, and usually involved the upper or

lower limb. In 32% of the patients subcutaneous nodules,which usually followed the lymphatics draining into the region-al lymph nodes, were present.

Proliferative responses

The proliferative responses of PBMC isolated from patientswith severe and mild disease and of Danish controls presum-

ably not previously exposed to L. major are summarized inTable 2. The responses to both the purified Leishmania antigen,gp63, and the crude Leishmania antigen preparations were

Table 1. Clinical features of cutaneous leishmaniasis patients atdiagnosis

Clinical feature Severe (n = 10) Mild (n = 15)Age* (years) 31-5 (18-52) 20-5 (14-52)Sex (F/M)t 3/7 7/8

Lesion features

Duration (weeks)t 25-3 ± 20-8 10-3 ± 4-9Number of lesions 4-9 + 4-0 6 1 + 5 5Size of lesions (mm) 41-1 ± 18 5 27-7 ± 17 4Subcutaneous nodules§ 8 (80) 0 (0)Lymphadenopathy§ 3 (30) 1 (6 7)

* Median (range).t (Female/male).tMean±s.d.§ Number of patients and (percentages).

Table 2. Antigen-induced proliferation (median kct/min (95% confi-dence interval)) in peripheral blood mononuclear cells (PBMC) frompatients with severe and mild cutaneous leishmaniasis (CL) and non-

exposed Danish controls

Antigen Severe CL (n = 10) Mild CL (n = 15) Controls (n = 9)

gp63 01 (0, 06)** 19 (0-6, 51) 0-2 (01, 03)LDAt 43 (23, 63)** 7-2 (4-6, 11-4) 0-2 (0,07)LMPt 7-7 (45, 11-4)* 12-4 (94, 159) 2-1 (0, 58)PPD 56 (19, 93) 6-4 (3-1, 110) 4-4 (2-4, 729)TT 02 (0, 64) 08 (03, 29) 24 (09, 38)

*, **Response significantly lower than in patients with mild disease(*P < 0-05; ** P < 0-01).

t, t Sonicates of Leishmania promastigote (t) or amastigote (t).

significantly lower in the group with severe disease than inthe group with mild disease. PBMC from Danish donors didnot proliferate in response to gp63 and amastigote sonicate(LDA), whereas a response to the promastigote sonicate (LMP)was found in six of nine PBMC isolates.

No differences were found between the groups in theresponse to PPD. Some of the Sudanese patients respondedto TT, and the responders were mainly women (data notshown), probably due to tetanus toxoid vaccination duringpregnancy.

IL-4 and IFN--y assaysThe IFN-7y response to the Leishmania antigens was lower inthe group with severe than in the group with mild disease (Table3), but only the difference in the response to promastigotesonicate reached statistical significance (P < 0-05). The PBMCfrom Danish donors did not produce IFN--y in response to gp63or LDA. Some Danes produced IFN--y in response to LMP,and the IFN--y response to LMP in Danes and patients withsevere disease was similar.

The IFN--y response to PPD was comparable in patientsand controls. The TT response was higher in the patients withmild disease than in the two other groups, but the differencewas not statistically significant. The difference between the

Table 3. Antigen-induced IFN--y production (median U/ml (95%confidence interval)) of peripheral blood mononuclear cells (PBMC)from patients with severe and mild cutaneous leishmaniasis (CL) and

non-exposed Danish controls

Severe CL (n = 10) Mild CL (n = 15) Danes (n = 9)Antigen IFN--y (U/ml) IFN-,y (U/ml) IFN-y (U/ml)

gp63 0 (0, 3-5) 4-4 (0-6, 8 4) 0 (0)tLDA 2-2 (0, 8-2) 8-7 (4-6, 13-5) 0 (0)**LMP 10-0 (4-6, 19-6)* 26-0 (13-5, 38 7) 12-5 (0, 26-5)**PPD 25 0 (1-4, 40-4) 24-6 (11-2, 42-8) 18 (0, 36 5)Ti 1-3 (04, 10-6) 5.57 (19, 99) 22 (0, 27)

**Production statistically significantly lower than in patientswith mild disease (*P < 0-05, **P < 0 01).

tn = 6.PPD, Purified protein derivative; TT, tetanus toxoid.

241

A. Gaafar et al.

Table 4. Antigen-induced IL-4 (median pg/ml (95% confidence inter-val)) peripheral blood mononuclear cells (PBMC) from patients withsevere and mild cutaneous leishmaniasis (CL) and non-exposed Danish

controls

Severe CL (n = 10) Mild CL (n = 15) Danes (n = 9)Antigen IL-4 (pg/ml) IL-4 (pg/ml) IL-4 (pg/ml)

gp63 10 (0 20) 0 (0 0) 0 (0 40)*LDA 25 (0-120) 0 (0-45) 0 (0 40)LMP 65 (0-130) 0 (0-55) 7 (0-100)PPD 64 (0 109) 0 (0-16) 17 (0 145)TT 0 (0-90) 0 (0-35) 360 (0 1610)*

*n = 6.PPD, Purified protein derivative; TT, tetanus toxoid.

response (Table 3) probably reflected a difference in the sex

ratio between those with mild and those with severe disease.The IL-4 (Table 4) responses were generally low, but PBMC

from some donors, mainly among the patients with severe

disease, produced IL-4 in response to the Leishmania anti-gens. The differences in Leishmania antigen-induced IL-4production between patients with mild and severe diseasewere not statistically significant. PPD induced only sporadicIL-4 production. A few of the Sudanese donors and approxi-mately half of the Danes had a measurable IL-4 response to TT.

Figures 1 and 2 show the corresponding IL-4 and IFN--yresponses of each patient. Most responses to Leishmaniaantigens could be divided into two response patterns (Fig. 1):(i) those with a considerable IFN-7y production and no IL-4(Thl-like); and (ii) those with lacking or moderate IFN--yproduction often associated with some IL-4 production (notThl-like). Most of the patients with mild disease (open circles)showed a Thl-like responsiveness, as opposed to the patientswith severe disease (closed circles), in whom a non-Thl-likeresponse pattern was most frequent. When the response of thepatients was defined as ThI-like and non-ThI-like on the basisof IFN--y production as indicated in Fig. 1 (dotted lines), thefraction of patients with ThI-like response pattern to Leishma-nia antigens was higher among those with mild diseasecompared with those with severe symptoms (Table 4). The

a) Fb)

-*-j

300

2000

100 h-

0 1 510 02

0 5 10 15 20 25

responses to PPD and TT could not be divided into distinctresponse patterns, and no correlation between the responsive-ness and disease severity was evident (Fig. 2).

DISCUSSION

Leishmania parasites cause human disease ranging from self-healing cutaneous ulcers to fatal systemic infections. In addi-tion many individuals become infected without developingdisease [24-27]. CL caused by L. major is often an uncompli-cated disease, characterized by a localized skin lesion whichheals spontaneously. However, in some cases complicationsmight occur as the infection results in large ulcers or spreadsthrough the lymphatics to the lymph nodes. To correlate Thl-and Th2-like immune responses to the clinical disease, we

attempted to identify patients who had difficulties in control-ling inflammation and parasite multiplication. Hence patientswith large lesions and a long clinical history and/or apparentspread of the parasite as evidenced by swelling of subcutaneousnodules were defined as suffering from severe disease. Thisclassification of the patients is probably not accurate. Forinstance, a patient with mild disease might later developsevere symptoms, and signs of severity could be caused bycomplications due to secondary infections. There was a higherproportion of men in the group with severe CL than in thegroup with mild disease. Furthermore, the duration of diseasewas shorter among the women than among the men (data notshown). These differences might be due to a tendency forwomen to seek treatment earlier than men.

The proliferative response to Leishmania antigens was lowerin PBMC from patients with severe disease than in PBMC frompatients with mild disease. This finding is in agreement with thefindings by Jaffe etal. [28] and Frankenburg etal. [29], whoreported on Leishmania antigen-specific proliferative response inpatients with mild CL caused by L. major, and the findings ofCastes et al., and Caceres-Dittmar et al., who found thatLeishmania antigen-induced PBMC proliferative response cor-

related with disease severity in South American patients sufferingfrom CL [30-32]. The Danish controls did not respond to LDAand gp63 antigen, wkhereas LMP as reported previously [33,34]activated the PBMC of some non-exposed donors.

Leishmania antigen-induced IFN-,y responses were higher inpatients with mild than in patients with severe disease, and the

F(c).

.~~~~~~

of 0,[Mo006 o!boE loo ol- 0oi (d cis ato 0 o lo o

0 5 10 15 20 25 0 10 20 30 40 50 60

IFN-y (U/mi)

Fig. 1. IFN--y and IL-4 production by peripheral blood mononuclear cells (PBMC) from cutaneous leishmaniasis (CL) patients in

response to the Leishmania antigens gp63 (a), amastigote sonicate (LDA) (b), and promastigote sonicate (LMP) (c). 0, Patients withmild disease; 0, patients with severe disease.

242

Human T cell response to Leishmania antigens 243

400 - although the lack of a reliable method to measure antigen-(a) induced IL-4 by human T cells has made it difficult to assess the

balance between Thl/Th2-like activity in cultures of humanPBMC. In this study we used an amplification system involving

300 - pulsing of cultures with PMA and ionomycin for IL-4 produc-tion [231. The antigen-induced IL-4 was calculated as the

difference between values from antigen-stimulated culturesand cultures incubated without antigen. By use of this

E 200 _ao method it has previously been shown that TT-induced IL-4

production correlated with the TT vaccination status of the

donors [23]. The IL-4 secretion measured in this system can

mainly be ascribed to CD2+CD4+ T cells [39].

Some of the CL patients produced IL-4 in response to the100 - Leishmania antigens, but no difference could be detected

0 between patients with mild and severe disease. However,

0 * when the IL-4 and IFN--y responses of each individual patientwere considered, two response patterns emerged. The response

0°0 0 was characterized by either a high IFN-7y response (Thl-like) or

0 000 O O O 0 a low IFN--y response, which occasionally was associated with*t1 * * ° -0 |0 IL-4 production (not Thl-like). It is interesting that the non-

0 5 10 15 20 25 Thl-like response pattern found in some of the CL patients was

not found in a previous study [17] of individuals recovered fromCL. These donors all responded with high production of IFN--y

400 - after stimulation by a Leishmania promastigote sonicate. The(b) PPD and TT cytokine profiles did not typify into distinct

response patterns (Fig. 2). We found this reassuring, since itindicated that the results obtained with Leishmania antigens

300 were not due to an artefact arisen during the cultivation of thePBMC.

0 When the response patterns found in the cultures activatedby Leishmania sonicates were defined on the basis of IFN--y

E 200 _ production, we found an association between the responsepattern and disease severity (Fig. 1, Table 5). Thl-like response

patterns were mainly found among patients with mild disease,whereas the response pattern described as non-Thl-like was

3 most common in patients with severe disease. These results are

100 3 in concordance with the findings of Pirmez etal. [40], who

* Table 5. Number of individuals with a Thl-like

000peripheral blood mononuclear cell (PBMC)

0 CD 000 oo 0 cm response to the Leishmania antigens, gp63, LDA,* 1I * 1 and LMA in patients suffering from and mild

0 10 20 30 40 50 60 cutaneous leishmaniasis, respectively

IFN-y (U/m)

Fig. 2. IFN--y and IL4 production by peripheral blood mononuclearcells (PBMC) from cutaneous leishmaniasis (CL) patients in response tothe purified protein derivative (PPD) (b) and tetanus toxoid (TT) (a). 0,Patients with mild disease; 0, patients with severe disease.

difference was statistically significant when comparing theresponses to LMP (P < 0-05). The difference in the capacityto produce IFN--y is interesting, since IFN-7 is a key factor inactivating macrophages to kill the parasite [35,36], and exo-

genous IFN--y has been found efficient in the treatment of CL[37] and in combination with pentavalent antimony in thetreatment of visceral leishmaniasis (VL) [38].

Evidence is accumulating that a Thl/Th2 dichotomy in theT cell response to Leishmania exists in humans [13-17],

Number (and percentage) withThI-like response*

Disease severity

Antigen Severe (n = 10) Mild (n = 15)

gp63 2 (20) 9 (60)LDA 3 (30) 11* (73)LMP 3 (30) 12** (80)

*Thl-like responsiveness was defined accordingto the production of IFN--y (Fig. 1).

** Number statistically significantly higherthan in patients with severe disease (*P = 0-049;**P = 0-034, Fisher's exact test).

244 A. Gaafar et al.

investigated skin biopsies obtained from patients suffering L.braziliensis infections, and found increased levels of mRNAcoding for IL-2, IFN--y and lymphotoxin in patients withlocalized CL, and high levels of mRNA for IL-4 in patientswith mucocutaneous manifestations.

In conclusion, we found that the in vitro function of PBMCof patients suffering from CL differed between those with mildand those with severe disease manifestations. Leishmaniaantigen-induced proliferation and IFN--y production werelower in the patients with severe disease than in the patientswith mild disease. Moreover, Leishmania-induced cytokineproduction of each patient fell into one of two patterns,characterized as Thl-like or non-Thl-like. A Thl-like responsepattern was associated with mild disease manifestations.

ACKNOWLEDGMENTSDr M. Hag-Ali, Dr Suad Sulaiman and Professor James Jensen areacknowledged for making our work possible at the MRC-NIH labora-tory in the Sudan. The excellent technical assistance of Jette Dalsten isgreatly appreciated. This work was supported by grant 104.Dan.8.L/401 from the Danish International Development Agency (DANIDA)and the Danish Biotechnology program and NIH grant AI-16312.

REFERENCES

1 WHO Expert Committee on Leishmaniasis. The leishmaniasis.Report of a WHO Expert Committee. Geneva: WHO, 1984.

2 Heyneman D. Immunology of leishmaniasis. Bull WHO 1971;44:499-514.

3 Guirges SY. Natural and experimental re-infection of man withOriental sore. Ann Trop Med Parasitol 1971; 65:197-205.

4 Bryceson ADM. Tropical dermatology: cutanous leishmaniasis.British J Dermatol 1976; 94:22.

5 Kubba R, El-Hassan AM, Al-Gindan Y etal. Dissemination incutaneous leishmaniasis I. Subcutaneous nodules. Int J Dermatol1987; 26:300-4.

6 El-Safi SH, Peters W. Studies on the leishmaniasis in the Sudan. 1.Epidemic of cutaneous leishmaniasis in Khartoum. Trans R SocTrop Med Hyg 1991; 85:44-47.

7 Kadaro AY, Ghalib HW, Ali MS etal. Prevalence of cutaneousleishmaniasis along the Nile River North of Khartoum (Sudan) inthe aftermath of an epidemic in 1985. Am J Trop Med Hyg 1993;48:44-49.

8 Homeida MMA, Sadig M, Ghalib HW, Gaafar A, Harron DWG.Cutaneous leishmaniasis epidemic in the Sudan. Int J Pharmacol1987; 1:142-4.

9 El-Safi SH, Peters W, El Toum B et al. Studies on the leishmaniasisin the Sudan. Clinical and parasitological studies on cutaneousleishmaniasis. Trans R Soc Trop Med Hyg 1991; 85:457-64.

10 Mosmann TR, Cherwinsky H, BondMW et al. Two types ofmurinehelper T cell clone. I. Definition according to profiles of lymphokineactivities and secreted proteins. J Immunol 1986; 136:2348-57.

11 Scott P, Pearce E, Cheever AW, Coffman RL, Sher A. Role ofcytokines and CD4+ T-cell subsets in the regulation of parasiteimmunity and disease. Immunol Rev 1989; 112:161-82.

12 Locksley RM, Scott P. Helper T-cell subsets in mouse leishmaniasis:induction, expansion and effector function. Immunoparasitol Today1991; 1:58-61.

13 Kemp M, Kurtzhals JAL, Bentdzen K et al. Leishmania donovani-reactive ThI- and Th2-like T-cell clones from individuals who haverecovered from visceral leishmaniasis. Infect Immun 1993; 61:1069-73.

14 Kurtzhals JAL, Hey AS, Schaefer KU etal. Dichotomy of thehuman T-cell response to Leishmania antigens. II. Absent or Th2

like response to gp63 and ThI like response to lipophosphoglycanassociated protein in cells from cured visceral leishmaniasis patients.Clin Exp Immunol 1994; 96:416-21.

15 Ghalib HW, Piuvezam MR, Skeiky YA-W etal. Interleukin 10production correlates with pathology in human Leishmania dono-vani infections. J Clin Invest 1993; 92:324-9.

16 Karp CL, El-Safi SH, Wynn TA et al. In vivo cytokine profiles inpatients with kala-azar. Marked elevation of both interleukin 10and interferon gamma. J Clin Invest 1993; 91:1644-8.

17 Kemp M, Hey AS, Kurtzhals JAL et al. Dichotomy of the human T-cell response to Leishmania antigens. 1. Thl-like response toL. major promastigote antigens in individuals recovered fromcutaneous leishmaniasis. Clin Exp Immunol 1994; 96:410-5.

18 Hviid L, Albeck G, Hansen B, Theander TG, Talbot A. Controlled-gradient cryopreservation of mononuclear cells: description of anew method suitable for field conditions, and comparison with aconventional technique. J Immunol Methods 1993; 157:135-42.

19 Kemp M, Theander TG, Handman E et al. Activation of human Tlymphocytes by Leishmania lipophosphoglycan. Scand J Immunol1991; 33:219-24.

20 Martinez S, Looker DL, Marr JJ. A tissue culture system for thegrowth of several species of Leishmania: growth kinetics and drugsensitivities. Am J Trop Med Hyg 1988; 38:304-7.

21 Glaser TA, Wells SG, Spithill TW etal. Leishmania major and L.donovani: a method for rapid purification of amastigote. ExpParasitol 1991; 71:343-5.

22 Bouvier J, Etges RJ, Bordier C. Identification and purification ofmembrane and soluble forms of the major surface protein ofLeishmania promastigotes. J Biol Chem 1985; 260:15504-9.

23 Kurtzhals JAL, Hansen MB, Hey AS, Poulsen LK. Measurement ofantigen-dependent interleukin-4 production by human peripheralblood mononuclear cells. Introduction of an amplification stepusing ionomycin and phorbol myristate acetate. J Immunol Meth-ods 1992; 156:239-45.

24 Pampligione S, Manson-Bahr PEC, Giungi F et al. Studies onMediterranean leishmaniasis. 2. Asymptomatic cases of visceralleishmaniasis. Trans Roy Soc Trop Med Hyg 1974; 68:447-53.

25 Badar6 R, Jones TC, Carvalho EM etal. New perspectives on asubclinical form of visceral leishmaniasis. J Infect Dis 1986;154:207-14.

26 Sacks DL, Lal SL, Shrivastava SN, Blackwell J, Neva FA. Ananalysis of T cell responsiveness in Indian Kala-Azar. J Immunol1987; 138:908-13.

27 Kurtzhals JAL, Hey AS, Theander TG et al. Cellular and humoralimmune reactivity of a population in Baringo District, Kenya, toLeishmania promastigote lipophosphoglycan. Am J Trop Med Hyg1992; 46:480-8.

28 Jaffe CL, Shor R, Trau H, Passwell JH. Parasite antigens recognizedby patients with cutaneous leishmaniasis. Clin Exp Immunol 1990;80:77-82.

29 Frankenburg S, Jonas F, Gross A, Klaus S. Development of in vitroparameters of cell-mediated immunity in the course of humancutaneous leishmaniasis infection. Am J Trop Med Hyg 1993;48:512-8.

30 Castes M, Agnelli A, Verde 0, Rondon AJ. Characterization of thecellular immune response in American cutaneous leishmaniasis.Clin Immunol Immunopathol 1983; 27:176-86.

31 Castes M, Cabrera M, Trujillo D, Convit J. T-cell subpopulationsexpression of interleukin-2 receptor, and production of interleukin-2 and gamma interferon in human American cutaneous leishma-niasis. J Clin Microbiol 1988; 26:1207-13.

32 Caceres-Dittmar G, Tapia FJ, Sanchez MA et al. Determination ofthe cytokine profile in American cutaneous leishmaniasis using thepolymerse chain reation. Clin Exp Immunol 1993; 91:500-5.

33 Kemp M, Hansen MB, Theander TG. Recognition of Leishmaniaantigens by T-lymphocytes from non-exposed individuals. InfectImmun 1992; 60:2246-51.

Human T cell response to Leishmania antigens 245

34 Akuffo HO, Britton SF. Contribution on non-Leishmania-specificimmunity to resistance to Leishmania infections in humans. ClinExp Immunol 1992; 87:58-64.

35 Murray HW, Rubin BY, Rothermel CD: Killing of intracellularLeishmania donovani by lymphokine-stimulated human mononuc-lear phagocytes. Evidence that interferon-gamma is the activatinglymphokine. J Clin Invest 1983; 72:1506-10.

36 Douvas GS, Looker DL, Vatter AE, Crowle AJ. Gamma-interferonactivates human macrophages to become tumoricidal and leish-manicidal but enhances replication of macrophage-associatedmycobacteria. Infect Immun 1985; 50:1-8.

37 Harms G, Zwingenberger K, Chehade AK etal. Effects of intra-

dermal gamma-interferon in cutaneous leishmaniasis. Lancet 1989;i: 1287-92.

38 Badaro R, Falcoff E, Badaro FS et al. Treatment of visceralleishmaniasis with pentavalent antimony and interferon gamma.N Eng J Med 1991; 322:16-21.

39 Kurtzhals JAL, Kemp M, Poulsen LK et al. Interleukin-4 andinterferon-gamma production by Leishmania stimulated peripheralblood mononuclear cells from nonexposed individuals. Scand JImmunol 1995; 41:(in press).

40 Pirmez C, Yamamura M, Uemura K, Paes-Oliverira M, Conseicao-Silva F, Modlin RL. Cytokine patterns in the pathogenesis ofhuman leishmaniasis. J Clin Invest 1993; 91:1390-5.