Recombinant cysteine proteinase from Leishmania (Leishmania) chagasi implicated in human and dog T-cell responses

INFECTION ANDIMMUNITY, June 2005, p. 3787–3789 Vol. 73, No. 6

0019-9567/05/$08.00⫹0 doi:10.1128/IAI.73.6.3787–3789.2005

Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Recombinant Cysteine Proteinase from

Leishmania (Leishmania)

chagasi

Implicated in Human and Dog T-Cell Responses

Paulo Henrique da Costa Pinheiro,

Suzana de Souza Dias,

Kelsen Dantas Eula´lio,

Ivete L. Mendonc¸a,

Simone Katz,

and Clara Lu

´cia Barbie´ri

*

Department of Microbiology, Immunology and Parasitology, Universidade Federal de Sa˜o Paulo, Escola Paulista de Medicina, Sa˜o Paulo, SP,1_{Division of Parasitology, Faculdade NOVAFAPI, Teresina, PI,}2_{Division of Infectious and}

Parasitic Diseases, Universidade Federal do Piauı´, Teresina, PI,3_{and Department of Clinical}

Veterinary, Universidade Federal do Piauı´, Teresina, Piauı´, PI,4_Brazil

Received 20 October 2004/Returned for modification 8 December 2004/Accepted 20 January 2005

High in vitro lymphoproliferative responses were induced in humans and dogs by a recombinantLeishmania (Leishmania) chagasi cysteine proteinase, with secretion of IFN-␥ in asymptomatic subjects or of IFN-␥, interleukin 4 (IL-4), and IL-10 in oligosymptomatic subjects. In contrast, responses of symptomatic patients and dogs were lower, with production of IL-4 and IL-10.

Visceral leishmaniasis (VL) is caused byLeishmania (Leish-mania) chagasiin South America, and dogs are the main res-ervoir of the disease. Among the immunological changes found with VL, those involving T cells and interleukin 10 (IL-10) production have been correlated with pathology, whereas con-trol of the infection has been associated with the production of gamma interferon (IFN-␥) (3, 4, 5, 8, 13). Leishmania (L.)

infantumcysteine proteinases have been used for the evalua-tion of humoral and cellular immune responses in human and canine VL (9, 12). A recombinant cysteine proteinase fromL. (L.) chagasi, rLdccys1, expressed by theLdccys1gene (10) in

Escherichia coli, was shown to be a suitable tool for the diag-nosis of American VL (6). The present study demonstrates that rLdccys1 elicits cellular immune responses in naturally infected humans and dogs in different stages of VL.

L. (L.) chagasi(MHOM/BR/1972/LD) amastigotes were iso-lated from spleens of infected hamsters, as previously de-scribed (2). TheL. (L.) chagasiLdccys1 was cloned and ex-pressed as described elsewhere (6). The study population included human and canine blood samples obtained from sub-jects living in Teresina, Piauı´ State, Brazil, an area where VL is endemic, and were classified as shown in Table 1. The human and dog procedures were approved by the Ethical Committee for Human and Animal Care at the Universidade Federal de Sa˜o Paulo, Escola Paulista de Medicina. The lymphoprolifera-tive assays were performed with peripheral blood mononuclear cells (PBMC) purified by Ficoll-Hypaque density gradient cen-trifugation. For in vitro proliferation assays, the cells were cultured into 96-wells plates (1⫻106

cells/ml) in RPMI 1640 medium with 10% pooled human serum (R10) in the presence of rLdccys1 (2.5␮g/ml) orL. (L.) chagasiamastigote extracts (AM) (equivalent to 1⫻107

amastigotes). After 96 h at 37°C in 5% CO2, the cells were pulsed with 1␮Ci of [3

H]thymidine/ well for 18 h, and [3

H]thymidine uptake was determined after filtration on glass fiber filters. Lymphoproliferative responses

were expressed as stimulation indexes (SI) determined by di-viding the mean counts per minute for antigen-stimulated cells, in triplicate, by the mean counts for the control medium. Stu-dent’sttest was used with SigmaStat software to evaluate the significance of the data (P ⬍ 0.05). After stimulation with rLdccys1, PBMC from asymptomatic, oligosymptomatic, symp-tomatic, and treated patients proliferated with mean stimula-tion indexes of 8.0, 6.25, 2.45, and 4.1, respectively, whereas healthy individuals showed basal levels of lymphoproliferation. Lower stimulation was obtained for PBMC from the same patients with AM (SI of 2.5, 5.0, 2.1, and 2.0 from asymptom-atic, oligosymptomasymptom-atic, symptomasymptom-atic, and treated patients, re-spectively) (Fig. 1A). In PBMC from asymptomatic, oligo-symptomatic, and symptomatic dogs, rLdccys1 induced SI means of 11, 3.3, and 1.8, respectively, whereas lower re-sponses were obtained with AM (SI of 2.2, 1.2, and 0.6 from asymptomatic, oligosymptomatic, and symptomatic dogs, re-spectively). Corroborating our results, the lymphoproliferative responses stimulated by the recombinantL. (L.) infantum cys-teine proteinase B (rCPB) in Iranian dogs with active VL were weaker in symptomatic than in asymptomatic animals. How-ever, in contrast to our results, asymptomatic dogs showed higher lymphoproliferative responses to the totalL. (L.) infan-tumlysate than to rCPB. In addition, for asymptomatic dogs, we found that higher lymphoproliferative responses were elic-ited by rLdccys1 (SI⫽11) than by rCPB (SI⬵4.0) (9, 12). For

* Corresponding author. Mailing address: Division of Parasitology, Universidade Federal de Sa˜o Paulo, Escola Paulista de Medicina, Rua Botucatu, 862-6o andar, Sa˜o Paulo, SP, Brazil 04023-062. Phone: 55 11 5576-4532. Fax: 55 11 5571-1095. E-mail: barbiericl@ecb.epm.br.

TABLE 1. Classification of study population according to VL clinical symptoms and isolation ofL. (L.) chagasi

amastigotes from bone marrow

Patient group _symptomClinicala _isolationParasite

Noninfected (n⫽10) ⫺ No

asymptomatic (n⫽21) ⫺⫺⫹ Yes

oligosymptomatic (n⫽22) ⫺⫹⫹ Yes

symptomatic (n⫽25) ⫹⫹⫹ Yes

treated (n⫽23) ⫺ No

a_{⫺, absence of clinical signs;⫺⫺⫹, fever;⫺⫹⫹, fever and splenomegaly;}

⫹⫹⫹, weight loss, splenomegaly, and hepatomegaly.

evaluation of lymphokine production, 1⫻107_{PBMC from VL} patients and dogs were cultured in 1 ml of R10 and maintained for 72 h in the presence of 2.5␮g/ml rLdccys1. Culture super-natants were tested for IFN-␥, IL-4, and IL-10 using a specific enzyme-linked immunosorbent assay. PBMC from asymptom-atic patients secreted high levels of IFN-␥when stimulated with rLdccys1 (1.722 ng/ml). Lower levels of IFN-␥ (0.871 ng/ml) were produced by PBMC from oligosymptomatic pa-tients, whereas this lymphokine was not detected in symptom-atic individuals. IL-4 and IL-10 were secreted by PBMC from oligosymptomatic patients (0.277 ng/ml and 0.317 ng/ml, re-spectively), and a significant increase in the secretion of these two lymphokines was detected in symptomatic individuals (0.848 ng/ml and 0.725 ng/ml for IL-4 and IL-10, respectively). No lymphokine secretion was found in the PBMC supernatants from normal and treated patients (Table 2). These results corroborate literature data which show that VL patients with a controlled form of the disease display Th1-type lymphokines, whereas immunosuppression and a Th2 profile predominate in individuals with exacerbated disease (1, 8, 14). Secretion of significant levels of IFN-␥, IL-4, and IL-10 by lymphocytes from oligosymptomatic patients is in agreement with data

which showed both resistant (IFN-␥) and susceptible (IL-10) profiles in this form of VL (7). PBMC from asymptomatic and oligosymptomatic dogs secreted IFN-␥, whereas IL-4 and IL-10 were released by PBMC from symptomatic animals (Ta-ble 3). Lymphocytes from oligosymptomatic dogs did not se-crete IL-4; however, IL-10 values were comparable in the su-pernatants of PBMC from humans and dogs presenting symptomatic and oligosymptomatic signs of VL. Few studies have reported cytokine dosage in dogs with VL (15), impairing comparison with our results. The production of nitric oxide (NO) was also evaluated for the supernatants from human and dog lymphocytes stimulated by rLdccys1, by use of the Griess reagent. PBMC from asymptomatic and oligosymptomatic pa-tients secreted significant levels of NO (108.42␮M ⫾19.22

␮M and 58.71␮M⫾ 7.78␮M, respectively), whereas symp-tomatic and treated patients secreted lower NO concentrations (19.22␮M⫾ 0.8␮M and 5.02␮M⫾ 0.2␮M, respectively) (Fig. 2A). A similar relationship between NO production and the clinical signs of VL was observed with PBMC cultures from dogs stimulated with rLdccys1 (82.16 ␮M⫾ 8.33␮M, 25.21

␮M⫾7.44␮M, and 7.79␮M⫾0.34␮M from asymptomatic, oligosymptomatic, and symptomatic dogs, respectively) (Fig.

FIG. 1. Proliferative responses of PBMC recovered from normal, asymptomatic, oligosymptomatic, symptomatic, and treated humans (A) and dogs (B) in the presence of rLdccys1 and AM. Data are representative of three experiments and are shown as mean SI⫾standard deviations. An SI ofⱖ2.0 was considered a positive response (P⬍0.05).

TABLE 2. Lymphokine production by PBMC from normal and VL patients stimulated byL. (L.) chagasi rLdccysa

Clinical manifestation Cytokine production (ng/ml)

IFN-␥ IL-4 IL-10

Noninfected (n⫽10) 0.011⫾0.002 0.003⫾0.001 0.017⫾0.002 Asymptomatic (n⫽21) 1.722⫾0.133 0.077⫾0.026 0.054⫾0.031 Oligosymptomatic (n⫽

22)

0.871⫾0.085 0.277⫾0.081 0.317⫾0.005 Symptomatic (n⫽25) 0.037⫾0.006 0.848⫾0.075 0.725⫾0.122 Treated (n⫽23) 0.012⫾0.002 0.008⫾0.001 0.013⫾0.001

a_{Supernatants were used at a 1:4 dilution. Results are the means for triplicate}

wells and are representative of three experiments. Minimal values considered from lymphokine dosages were as follows: IFN-␥, 0.0153 ng/ml; IL-4, 0.016 ng/ml; IL-10, 0.016 ng/ml.

TABLE 3. Lymphokine production by PBMC from normal and VL dogs stimulated byL. (L.) chagasirLdccysa

Clinical manifestation Cytokine production (ng/ml)

IFN-␥ IL-4 IL-10

Normal (n⫽6) 0.008⫾0.003 0.012⫾0.004 0.006⫾0.002 Asymptomatic (n⫽18) 0.541⫾0.134 0.062⫾0.029 0.026⫾0.008 Oligosymptomatic (n⫽

14)

0.174⫾0.032 0.061⫾0.008 0.310⫾0.027 Symptomatic (n⫽17) 0.036⫾0.006 0.461⫾0.053 0.814⫾0.241

a_{Supernatants were used at a 1:4 dilution. Results are the means for triplicate}

wells and are representative of three experiments. Minimal values considered from lymphokine dosages were as follows: IFN-␥, 0.0153 ng/ml; IL-4, 0.016 ng/ml; IL-10, 0.016 ng/ml.

2B). The observed direct correlation between the secretion of IFN-␥and NO during the development of VL in humans and dogs has been previously reported (11, 16).

Overall, our results showed that the rLdccys1 antigen is a suitable immunological marker for several stages of VL in humans and dogs.

We thank Adriana Nunes Pinheiro and Joa˜o Batista Teles from the Instituto de Doenc¸as Tropicais Natan Portella, Teresina, Piauı´ State, Brazil, for excellent assistance.

This study was supported by FAPESP (Fundac¸a˜o de Amparo a` Pesquisa do Estado de Sa˜o Paulo), CAPES (Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico), and FACULDADE NO-VAFAPI.

REFERENCES

1.Bacellar, O., M. Barral-Neto, R. Badaro´, and E. M. Carvalho.1991. Gamma interferon production by lymphocytes from children infected withL. chagasi (L.) chagasi. Braz. J. Med. Biol. Res.24:791–795.

2.Barbie´ri, C. L., A. I. Doine, and E. Freymu¨ller.1990. Lysosomal depletion in macrophages from spleen and foot lesions ofLeishmania-infected hamster. Exp. Parasitol.71:218–228.

3.Carvalho, E. M., O. Bacellar, C. Brownell, T. Regis, R. L. Coffman, and S. G. Reed.1994. Restoration of IFN-␥production and lymphocyte proliferation in visceral leishmaniasis. J. Immunol.152:5949–5956.

4.Carvalho, E. M., A. Barral, D. Pedral-Sampaio, M. Barral-Netto, R. Badaro´, H. Rocha, and W. D. Johnson.1992. Immunologic markers of clinical evo-lution in children recently infected withLeishmania donovani chagasi. J. In-fect. Dis.165:535–540.

5.Carvalho, E. M., R. S. Teixeira, and W. D. Johnson, Jr.1981. Cell-mediated immunity in American visceral leishmaniasis: reversible immunosuppression during acute infection. Infect. Immun.22:649–656.

6.Dias, S. S., P. H. C. Pinheiro, S. Katz, M. R. M. dos Santos, and C. L. Barbie´ri.2005. A recombinant cysteine proteinase fromLeishmania( Leish-mania)chagasisuitable for serodiagnosis of American visceral leishmaniasis. Am. J. Med. Trop. Hyg.72:126–132.

7.Gama, M. E. A., J. M. L. Costa, J. C. R. Pereira, C. M. C. Gomes, and C. E. P. Corbett.2004. Serum cytokine profile in the subclinical form of visceral leishmaniasis. Braz. J. Med. Biol. Res.37:129–136.

8.Ghalib, H. W., M. R. Piuvezam, Y. A. Sheiky, M. Siddig, F. A. Hashim, A. M. el-Hassan, D. M. Russo, and S. G. Reed.1993. Interleukin 10 production correlates with pathology in humanLeishmania donovaniinfections. J. Clin. Investig.92:324–329.

9.Nakhaee, A., T. Taheri, M. Taghikhani, M. Mohebali, A.-H. Salmanian, N. Fasel, and S. Rafati.2004. Humoral and cellular immune responses against Type I cysteine proteinase ofLeishmania infantumare higher in asymptom-atic than symptomasymptom-atic dogs selected from a naturally infected population. Vet. Parasitol.119:107–123.

10.Omara-Opyene, A. L., and L. Gedamu.1997. Molecular cloning, character-ization and overexpression of two distinct cysteine protease cDNAs from

Leishmania donovani chagasi. Mol. Biochem. Parasitol.90:247–267. 11.Panaro, M. A., A. Acquafredda, S. Lisi, D. D. Lofrumento, T. Trotta, R.

Satalino, M. Saccia, V. Mitolo, and O. Brandonisio.1999. Inducible nitric oxide synthase and nitric oxide production inLeishmania infantum-infected human macrophages stimulated with interferon-gamma and bacterial lipo-polysaccharide. Int. J. Clin. Lab. Res.29:122–127.

12.Rafati, S., A. Nakhaee, T. Taheri, A. H. Ghashghaii, A. H. Salmanian, M. Jimenez, M. Mohebali, S. Masina, and N. Fasel.2003. Expression of cysteine proteinase type I and II ofLeishmania infantumand their rec-ognition by sera during canine and human visceral leishmaniasis. Exp. Parasitol.103:143–151.

13.Rezai, H. R., S. M. Ardehali, G. Amirhakimi, and A. Kharazmi.1978. Immunological features of kala-azar. Am. J. Trop. Med. Hyg.27:1079–1083. 14.Sang, D. K., J. H. Ouma, C. C. John, C. C. Whalen, C. L. King, A. A. F. Mahmoud, and F. P. Heinzel.1999. Increased levels of soluble interleukin-4 receptor in the sera of patients with visceral leishmaniasis. J. Infect. Dis. 179:743–746.

15.Santos-Gomes, G. M., R. Rosa, C. Leandro, S. Cortes, P. Roma˜o, and H. Silveira.2002. Cytokine expression during the outcome of canine experi-mental infection byLeishmania infantum. Vet. Immunol. Immunopathol. 88:21–30.

16.Voldoukis, I., J. C. Drapier, A. K. Nu¨ssler, Y. Tselentis, O. A. Silva, M. Gentilini, D. M. Mossalayi, L. Monjour, and B. Dugas.1996. Canine visceral leishmaniasis: successful chemotherapy induces macrophage antileishmanial activity via theL-arginine nitric oxide pathway. Antimicrob. Agents

Che-mother.40:253–256.

Editor:W. A. Petri, Jr.

FIG. 2. NO production. PBMC isolated from human (A) and dog (B) blood were cultured for 72 h in the presence of 2.5␮g/ml rLdccys1, and NO was assayed in the cellular supernatants as described previously. Results are shown as the means⫾standard deviations and are representative of three experiments.