E L S E V I E R Immunology Letters 48 (1995) 209-214

immunology letters

Mapping of the linear antigenic determinants from the Leishmania infantum histone H2A recognized by sera from dogs with

leishmaniasis

Manuel Soto a, Jose M. Requena", Luis Quijada a, Mercedes Garcia b, Fanny Guzman c, Manuel E. Patarroyo c, Carlos Alonso a,*

aCentro de Biologia Molecular "Severo Ochoa", Universidad Aut6noma de Madrid, Cantoblanco, 28049 Madrid, Spain bDepartamento de Medicina y Sanidad Animal, Facultad de Veterinaria, Universidad de Extremadura, 10071 Chceres, Spain

Clnstituto de Immunologia, Hospital de San Juan de Dios, Bogota, Colombia

Received 28 September 1995; revised 3 November 1995; accepted 9 November 1995

Abstract

Antibodies reacting against the H2A histone protein were frequently observed in the sera from dogs naturally infected with the protozoan parasite Leishmania infantum. Using synthetic peptides covering the complete sequence of the protein we have identified the amino terminal region, comprising from amino acids 1 to 20, and the carboxyl terminal region, comprising from amino acids 106 to 132, as conforming the antigenic determinants of the protein. Those regions, exposed in the nucleosome surface, are highly divergent in sequence relative to the mammalian H2A histones. The anti-H2A histone antibodies present in the sera of these dogs specifically recognize the L. infantum H2A histone and they do not react with mammalian histones. The present data indicate that, in spite of the evolutionary conservation of the H2A histone protein among eukaryotic organisms, the humoral response against this protein during natural infection is specifically triggered by the parasite protein antigenic determinants.

Keywords: Anti-histone antibody; Canine leishmaniasis; Synthetic peptide: B-cell epitope: Histone H2A

I. Introduction

Intracellular protozoan parasites belonging to the genus Leishmania are the etiological agents of leishma- niasis, a spectrum of diseases whose manifestations depend upon the species involved and the efficiency of the host's immune system to respond to the parasite. In the Mediterranean basin, infection with L. infantum results in the human and canine visceral leishmaniasis (VL) forms of the disease [1]. In this region VL has been recognized as an opportunistic infection in pa- tients having immune deficiency syndrome [2]. In the different forms of the leishmaniasis disease a spectrum of immune responses, both cell-mediated and humoral, are induced (see Ref. [3] for a review). The elucidation of the critical parasite antigens involved in the trigger- ing of the host immunity remains of major importance.

*Corresponding author. Tel.: + 34 1 3978473; Fax: +34 1 3974799.

0165-2478/95/$09.50 © 1995 - - Elsevier Science B.V. All rights reserved

S S D I 0165-2478(95)02473-U

Antigenic membrane molecules such as gp63 [4], gp46 [5] and LPG [6] may be of great interest because of the high copy number of these molecules on the surface of the promastigote. In addition, several intracellular proteins have also been described as immunodominant antigens during Leishmania infections. Examples of this class of antigens are kinesin [7], hsp70 [8], acidic riboso- mal proteins [9] and ribosomal protein LiP0 [10].

Recently, we have reported the isolation of the L. infantum histone H2A gene after immunoscreening of a cDNA expression library with sera from dogs with leishmaniasis [11]. In the present work we have sought to define the level and specificity of the humoral im- mune response against the H2A histone protein during the natural leishmaniasis. The data presented indicate that in canine VL the immunodominant anti-H2A re- sponse is specifically triggered by Leishmania histone H2A and that the epitopes defining the antigenic deter- minants are located in the amino and carboxyl terminal regions of the protein.

210 M. Soto et al. / Immunology Letters 48 (1995) 209-214

2. Materials and methods

2.1. Parasites and sera

Promastigotes of L. infantum (LEM 75; zymodeme l) were grown at 26°C in RPMI 1640 medium (Gibco, Paisley, UK) supplemented with 10% heat-inactivated fetal calf serum (Flow Lab., Irvine, UK). A Maracay strain of T. cruzi was also used in this study. The parasites were also grown in the same medium but supplemented with 15% heat inactivated fetal calf serum.

The sera were obtained from 31 dogs with visceral leishmaniasis (VL) from the Extremadura region of Spain. The animals were clinically and analytically eval- uated at the Department of Parasitology (Veterinary School, Extremadura University, Spain). All sera were positive when tested by indirect immunofluorescence, having titres ranging from 160 to 640. The presence of amastigote forms of the parasite in macrophages from popliteal and prescapular lymphoid nodes confirmed, moreover, the existence of Leishmania-infection in these animals. The sera from 10 healthy animals, used as negative controls, were also obtained from dogs living in the same geographical area.

2.2. Cloning and purification of rLiH2A

The cDNA coding for L. infantum LiH2A (cL71 clone) described previously [11] was cloned into the EcoRI site of the pMAL-cRI expression plasmid (Bio- labs, Inc.). The purification of the recombinant protein rLiH2A was performed according to the methodology provided by the supplier (Biolabs, Inc).

were blocked with 5% nonfat dried milk powder in PBS and 0.5% Tween 20. The filters were sequentially probed with primary and secondary antisera in block- ing solution. A peroxidase immuno-conjugate (Nordic Immunology) was used as second antibody and the specific binding was revealed with the Western blotting detection ECL system (Amersham).

2.5. FAST-ELISA measurements

The sensibilization of FAST-ELISA lids (Becton- Dickinson) was performed overnight at room tempera- ture using I00 /~1 of the antigen diluted in PBS. The antigen concentration was 1.5 and 100 /~g/ml of the recombinant protein or peptides, respectively. Previ- ously, the binding of peptides to FAST-ELISA lids was monitored by measurement of the decrease in A22o of a solution of 10/zg/ml of peptide after overnight incuba- tion with the lids. The binding capacity was approx. 100 ng/lid and no significant differences in the binding of each one of the peptides were observed. After sensi- bilization, the lids were washed with PBS and 0.5% Tween 20. Afterwards, the antigen-coated lids were incubated for 1 h with the blocking solution (5% nonfat dried milk powder in PBS and 0.5% Tween 20). The lids were immersed in the microtiter plates containing the diluted sera and incubated for 2 h at room temper- ature with shaking. After exposure to antibody, the lids were washed as described above. As secondary anti- body, horseradish peroxidase-labelled antibodies (dilu- tion 1:2000) were used. After incubation for 1 h at room temperature and washing the lids were developed by ortho-phenylenediamine, as substrate. The ab- sorbance was read at 450 nm.

2.3. Protein samples 2. 6. Affinity-purification of antibodies

Protein samples were prepared from l09 Leishmania promastigotes or a similar number of T. cruzi epi- mastigotes. The cultured parasites were washed three times with phosphate-buffered saline (PBS), incubated in ice for 15 rain in 0.2 ml of a lysis buffer containing 10 mM Tris-HC1, pH 8.0, 1% Triton X-100, 0.15 M NaC1, 1 mM phenylmethylsulfonyl fluoride and soni- cated for 10 rain in a water bath. Preparations of nuclear fractions were obtained according to the methodology described by Schreiber et al. [12]. Com- mercial preparations of calf thymus histones (type II-S) were purchased from Sigma Chemical Co. (USA).

2.4. Immunoblot analysis

The specific antibodies against the rLiH2A protein were affinity-purified from a pool of 12 positive anti- H2A sera on an antigen column. For that purpose, 1 mg of the recombinant protein was covalently bound to CNBr-activated Sepharose 4B (Pharmacia) and packed into a column. Coupling and blocking was carried out according to the manufacturers' instructions. Two ml of the mixed sera were passed through the antigen column. After washing, the specific antibodies were eluted from the column with 0.1 M glycine, pH 2.8. Finally, the antibody preparation was equilibrated to pH 7.5 with 1 M Tris-HCl. The solution of the anti- body was restored to the original volume of the pooled serum (2 ml).

SDS-PAGE on 15% acrylamide gels was performed using the Mini-protean system (BioRad). After elec- trophoresis the proteins were transferred to nitrocellu- lose membranes (Amersham). Afterwards the transfers

2. 7. Synthesis of peptides

A library of overlapping peptides covering the whole L. inJantum histone H2A sequence [11] was synthesized

M. Soto et al. / Immunology Letters 48 (1995) 209-214 211

1.2.

]) 1.0

U

(~ 0.8

0.6--

0.4-

0.2-

1 2 3 4 5 6 7 8 9 i011121314151617 18 19 20 21 22 23 24 25 26 27 28 29 30 31

Sera Fig. 1. Reactivity of leishmaniasis sera against the L. inJantum histone H2A. The reactivity of 31 sera (1:250 dilution) against the rLiH2A was assayed by FAST-ELISA. The absorbance values of sera minus the absorbance mean of ten control sera (A450 = 0.129; SD = 0.059) are represented.

by the simultaneous multiple-peptide solid-phase syn- thetic method using a polyamide resin and FMOC chemistry [13]. Purity was checked by amino acid anal- ysis and HPLC.

3. Results

3.1. Reactivity of sera from VL dogs with the recombinant L. infantum histone H2A

A cDNA coding for the L. infantum histone H2A was cloned into the pMAL-cRI expression vector to produce the recombinant protein rLiH2A. An affinity purified preparation of rLiH2A was used in FAST- ELISA assays to analyze the levels of specific antibod- ies in the sera from 31 dogs naturally infected with L. infantum (Fig. 1). It was observed that 78% of the sera showed values of anti-rLiH2A reactivity higher than the mean plus 3 S.D. of control values. Thus, given this frequency of recognition, it could be concluded that the histone H2A protein is a potent immunogen during Leishmania infection of canine hosts. Preliminary data from our laboratory appear to confirm the presence of anti-rLiH2A antibodies also in human VL sera.

3.2. Mapping of the antigenic regions of Leishmania h&tone H2A

In order to determine the region(s) involved in the antibody recognition of the Leishmania histone H2A a series of peptides overlapping by 5 amino acids were synthesized (Fig. 2). In this study, the 19 canine VL sera with the highest anti-rLiH2A reactivity values (Fig. 1) were selected and assayed by FAST-ELISA against the synthetic peptides. Among the nine overlapping peptides tested only peptides P1, P8 and P9 were recog- nized by the leishmaniasis sera (Fig. 2). Peptide P1 is formed by the first 20 amino acids of the H2A histone sequence and peptides P8 and P9 cover the 27 carboxyl- terminal amino acids. Peptide 6 was also recognized by the leishmaniasis sera but the reactivity value was very low (Fig. 2). The reactivity values of the leishmaniasis sera against the rest of the synthetic peptides was similar to the value of the background measured in the absence of antigen. Remarkably, a heterogeneous recognition of the synthetic peptides P1, P8 and P9 was observed among VL sera; while certain sera mainly recognized the amino- (P1 peptide) others only recog- nized the carboxyl-terminal region (P8/P9 peptides) of the protein. The high standard deviation of the mean value of the reactivity against each one of the peptides

212 M, Soto et al. / Immunology Letters 48 (1995) 209 214

Peptides

pl I MATPRSAKKAARKSGSKSAK 20

P2 x6 SKSAKCGLIFPVGRVGGMMR 35

P3 31 GGMMRRGQYARRIGASGAVY so

P4 46 SGAVYLAAVLEYLTAELLEL 65

ps 61 ELLEL SVKAAAQSGKKRCRL 8 0

P6 v 6 KRCRLNPRTVMLAARHDDDI 9 s

P v s I HDDD I GTLLKNVTLSHSGVV 11o

Ps i o 6 HSGVVPNI SKAMAKKKGGKK 125

P9 121 KGGKKGKATPSA 1 3 2

I I I I I |

0 . 1 0 . 2 0 . 3 0 . 4 0 . 5 0 . 6

I

0.7

Absorbance

Fig. 2. Synthetic peptides, 20-mers overlapping by five residues covering the entire L. inJantum H2A histone, were assayed by standard FAST-ELISA against 19 canine leishmaniasis sera at 1:250 dilution. The absorbance mean values of the 19 sera against each one of the peptides are represented.

is an indication of the large heterogeneity in recogni- tion of the peptides shown by the sera collection (Fig. 2).

The analysis of the reactivity values obtained against peptides P8 and P9 suggested the existence of several antigenic determinants within the carboxyl-ter- minal part of the Leishmania histone H2A. Fig. 3 illustrates the differential recognition of peptides of the carboxyl-terminal region shown by individual VL sera. Some of the sera reacted strongly with peptide P8 and weakly with peptide P9 while other sera re- acted strongly with peptide P9 and did not react with peptide PS. Thus, one of the antigenic determinants must be located in the amino-terminal half of peptide P8 and a second one in the carboxyl-terminal end of the histone H2A, present in P9. The possible presence of a third epitope in the carboxyl-terminal region was then examined by testing the reactivity of the sera against an overlapping peptide corresponding to residues 113-132 (peptide 10). From the results pre- sented in Fig. 3 showing that peptide P10 was not recognized by serum 2 while it strongly reacted against peptide P8 and that serum 4 did not react with peptide P9 while it recognized peptide P10, we deduced that the sequence ISKAMAKK could be defining a third antigenic determinant of the carboxyl- terminal region of the protein.

3.3. The anti-H2A reactivity of the Leishmaniasis sera is specifically directed against the Leishmania histone H2A

Since the histones are among the most highly conserved proteins in nature [14,15] and these proteins are known to be immunogenic in autoimmune diseases, we in- vestigated whether the anti-H2A histone antibodies induced during Leishmania infection cross-react with mammalian histones. For that purpose, antibodies reacting against the rLiH2A protein were isolated by affinity chromatography from a pool of canine leishmaniasis sera. The reactivity analysis of these anti- bodies tested on immunoblots against total parasite pro- teins, parasite nuclear extracts and calf thymus histones (Fig. 4A) showed that the anti-rLiH2A antibodies recognized a single band of about 14 kDa present in both L. infantum nuclear extracts (lane 3) and total proteins (lane 5), but that there was no reactivity against the thymus histones (lane 6). Thus, it may be concluded that the anti-histone H2A antibodies, present in the sera of VL dogs, are specifically elicited by the Leishmania histone H2A. The anti-H2A antibodies purified from the sera of Leishmania-infected dogs cross-reacted, however, with histone H2A of the trypanosomatid T. cruzi since, as Fig. 4B shows, the anti-rLiH2A puri- fied antibodies from the leishmaniasis sera reacted

M. Soto et al. / Immunology Letters 48 (1995) 209 214 213

P8

P9

PI0

H S GWP N I S KAMAKKKGGKKG

KGGKKGKATPSA

I SKAMAKKKGGKKGKATP SA

1 2 4 7 I0 Ii 14 15 17 20 21 23

P8

P9

PIO

Fig. 3. Sequence alignment of the synthetic peptides (P8, P9 and P10) derived from the carboxyl-terminal sequence of the histone H2A. The figure illustrates the reactivity values shown by 12 VL sera against the three peptides. The number of each serum corresponds to that used in Fig. 1. Black squares represent absorbance values higher than 0.3. Grey squares represent absorbance values between 0.05 and 0.3. White squares are considered as negative values ( < 0.05).

with the 14-kDa protein band in nuclear and total protein extracts from this parasite. The cross-reactivity is in agreement with the high sequence similarity exist- ing between histones H2A from L. infantum and T. cruzi [16]. Phylogenetic analysis of the histone H2A proteins has indicated that Trypanosomatids possess the more divergent histones known to date and that these organisms diverged from the main eukaryotic line earlier than any other of the organisms represented in the tree reported by Thatcher and Gorovsky [15].

4. Discussion

The data presented in this paper demonstrate that during L. infantum infection histone H2A is an impor- tant target of the humoral immune response. Our re- suits indicated that 78% of the sera from dogs with VL contain significant levels of anti-H2A antibodies. To our knowledge this is the first time that during a parasitic infection a prominent humoral response against histones has been reported. High levels of anti- histone auto-antibodies have been also'described in sera from patients with systemic lupus erythematosus and other autoimmune diseases [17], and in patients with infectious mononucleosis [18]. In the latter case the increase in anti-histone auto-antibodies appears to be due to a polyclonal B cell activation induced by Ep- stein-Barr virus. The present data indicate, on the other hand, that during Leishmania infection the anti-histone humoral response is specifically elicited by histone H2A of the parasite since the anti-histone antibodies induced during the disease did not recognize calf thymus his- tone, although they reacted with the H2A of T. cruzi, a

phylogenetically related protozoan. This correlates with the fact that the histones from Trypanosomatids pos- sess the most divergent primary structure known to date [15]. While the sequence identity among mam- malian H2A is higher than 99% [14], the L. infantum H2A has 45% of sequence identity when compared with the consensus H2A sequence [11]. The L. infantum H2A possesses a highly conserved central core region with amino- and carboxyl-regions more divergent [11]. Re- markably, we have found that the antigenic determi- nants of the Leishmania H2A are located in amino- and carboxyl-regions, i.e., the less-conserved regions of the molecule. Thus, our data indicate that during Leishma- nia infection, the anti-H2A humoral response is di- rected against the parasite histone and the host self-tolerance is not broken.

On the other hand, the fact that epitopes of L. infantum histone H2A are located in the amino- and carboxyl-terminal regions could be reflecting a more general mechanism of selective B cell activation and epitope selection in immune humoral response against histones. The finding that the anti-histone auto-anti- bodies induced during autoimmune diseases are di- rected against the histone regions that are thought to be exposed in the nucleosomal surface has led to the suggestion that during these diseases the autoimmune response is probably stimulated by the chromatin cores and that the anti-histone antibodies are triggered by the histone regions exposed in the nucleosome rather than by the free histones [19]. In the nucleosomal core, the 20 30 residues at both amino- and carboxyl-terminal regions of the histone H2A are exposed on the nu- cleosome surface, whereas the rest of the molecule is internal and it is implicated in the histone-histone and

214 M. Soto et al. / Immunology Letters 48 (1995) 209 214

A 66--

45-- 36--

29-- 24-- 20--

14--

1 2 3 4 5 6

a

B 66--

45--

36--

29-- 24-- 20--

14--

1 23 4 5 6

Fig. 4. Species-specificity of anti-rLiH2A antibodies obtained from a mixture of canine leishmaniasis sera. Proteins were electrophoresed on 15% polyacrylamide gels and blotted. (A) Lane 1, 1 /tg of the afffinity-purified MBP-rLiH2A protein; lane 2, 1 /~g of the affinity- purified MBP protein; lane 3, 10 /~g of L. infantum nuclear protein extracts; lane 4, 10/~g of L. infantum cytosolic extracts; lane 5, 20 ~g from L. infantum total proteins; lane 6, 4 pg from calf thymus histones. (B) Lanes 1 and 4, 10 pg of L. infantum or T. cruzi nuclear protein extracts, respectively; lanes 2 and 5, 10/tg of L. in/cmtum or T. cruzi cytosolic extracts, respectively; lanes 3 and 6, 20 /zg of L. infantum or T. cruzi total proteins, respectively. Both immunoblots were probed with the anti-rLiH2A antibodies purified from a mixture of canine leishmaniasis sera by affÉnity chromatography. Molecular mass markers are shown in kilodaltons.

h i s tone -DNA interactions. A m o n g the 10 overlapping

synthetic peptides tested in this study covering the complete sequence of the protein only four, namely peptides P1, P8, P9 and P10, were recognized by the sera of infected dogs. Peptide P1 conta ins the 20 amino- terminal residues of the histone H2A while peptides P8, P9 and P10 cover the 27 carboxyl- terminal amino acids of the molecule. Thus, the striking coincidence between the locat ion of the antigenic de terminants of the protein

and their predicted locat ion on the nucleosome surface could be taken as an indicat ion that the an t i -H2A ant ibodies elicited dur ing leishmaniasis are generated by a mechanism of epitope selection similar to that operat ing in a u t o i m m u n e diseases.

Acknowledgements

This work was supported by funds from the Comu- n idad A u t 6 n o m a de Madr id (160-9 and I + D0020/94), C Y C I T (SAF93-0146), and a C D T I grant to Labora to- rios LETI. A n ins t i tu t ional grant from F u n d a c i 6 n Ra- m 6 n Areces is also acknowledged.

References

[1] Bray, R.S. (1974) Ann. Rev. Microbiol. 28, 189. [2] de Gorgolas, M. and Miles, M.A. (1994) Nature 372, 734. [3] Liew, F.Y. and O'Donnell, C.A. (1993) Adv. Parasitol. 32, 161. [4] Shreffier, W.G., Burns, J.M. Jr., Badar6, R., Ghalin, H.W.,

Button, L.L., McMaster, W.R. and Reed, S.G. (1993) J. Infect. Dis. 167, 426.

[5] Lohman, K.L., Langer, P.J. and McMahon-Pratt, D. (1990) Proc. Natl. Acad. Sci. USA 87, 8393.

[6] Turco, S.J. and Decoteaux, A. (1992) Ann. Rev. Microbiol. 46, 65.

[7] Burns, J.M. Jr., Shreffier, W.G., Benson, D.R., Ghalib, H.W., Badaro, R. and Reed, S.G. (1993) Proc. Natl. Acad. Sci. USA 90, 775.

[8] MacFarlane, J.M., Blaxter, M.L., Bishop, R.P., Miles, M.A. and Kelly, J.M. (1990) Eur. J. Biochem. 190, 377.

[9] Soto, M., Requena, J.M., Garcia, M., Gomez, L.C., Navarrete, I. and Alonso, C. (1993) J. Biol. Chem. 268, 21835.

[10] Soto, M., Requena, J.M. and Alonso, C. (1993) Mol. Biochem. Parasitol. 61, 265.

[11] Soto, M., Requena, J.M., Gomez, L.C., Navarrete, I. and Alonso, C. (1992) Eur. J. Biochem. 205, 211.

[12] Schreiber. E., Matthias, P., Mfiller, M. and Schaffner, W. (1989) Nucleic Acids Res. 17, 6419.

[13] Houghten, R.A. (1985) Proc. Natl. Acad. Sci. USA 82, 5131. [14] Wells, D.E. (1986) Nucleic Acids Res. 14, rl19. [15] Thatcher, T.H. and Gorovsky, M.A. (1994) Nucleic Acids Res,

22, 174. [16] Puerta, C., Martin, J., Alonso, C. and Lopez, M.C. (1994) Mol,

Biochem. Parasitol. 64, I. [17] Tan, E.M. (1991) Cell 67, 841. [18] Garzelli, C., Manunta, M., Incaprera, M., Bazzichi, A., Conaldi,

P.G. and Falcone, G. (1992) lmmunol. Lett. 32, 111. [19] Burlingame, R.W., Boey, M.U, Starkebaum, G, and Rubin,

R.L. (1994) J. Clin. Invest. 94, 184.