Crossreactions and sequence homologies between recombinant polypeptides from Leishmania aethiopica and human IgG and IgM

FEMS Immunology and Medical Microbiology 17 (1997) 11^19

Crossreactions and sequence homologies between recombinant polypeptides from Leishmania aethiopica and human IgG and IgM Arve Osland a

a ;b ;

*, Senait Ashena¢

1 ;b

, Hege K. Vefring a , Terje E. Michaelsen

c

Department of Clinical Chemistry, Central Hospital in Rogaland, Armauer Hansens vei 20, 4003 Stavanger, Norway

b c

Armauer Hansen Research Institute, P.B. 1005, Addis Ababa, Ethiopia

Department of Vaccinology, National Institute of Public Health, Geitemyrsveien 75, 0462 Oslo, Norway

Received 27 August 1996; revised 28 October 1996; accepted 29 October 1996

Abstract

Recombinant DNA fragments M (154 bp) and G (206 bp), coding for recombinant polypeptides that crossreact with human IgM and IgG, have been isolated from a genomic library of Leishmania aethiopica. Epitope scanning of the two recombinant polypeptides, using overlapping octapeptides, revealed several crossreactive epitopes present in both recombinant proteins. By comparing amino acid sequences, similar sequences in human W and Q immunoglobulin heavy chains were identified. One of the parasite octapeptides is identical to an octapeptide in Q1 covering the Gm(a) allotypic marker. Expression of both the M and G fragments was detected in the parasites by RT-PCR of total mRNA, using primers specific for these fragments. Preliminary data showed that the presence of autoimmune anti-IgG antibodies was more pronounced in sera from patients with diffuse cutaneous leishmaniasis than in sera from patients with localised cutaneous leishmaniasis. We suggest that these immunoglobulin-crossreacting epitopes potentially might contribute to the induction of rheumatoid factors and be involved in the interplay between the parasite and the host immune system. Keywords :

Immunoglobulin crossreaction;

Leishmania aethiopica

; Gm(a) allotype

1. Introduction

Leishmania aethiopica is the causative agent of the two forms of leishmaniasis found in the highlands of Ethiopia [1,2], namely di¡use cutaneous leishmaniasis (DCL) and localised cutaneous leishmaniasis (LCL). The ¢rst form shows multiple skin lesions on the face, trunk and extremities and is usually

* Corresponding author. Tel.: +47 51 51 95 02; fax: +47 51 51 99 07; e-mail: [email protected] 1 Present address: Department of Pathology, Medical Faculty, Addis Ababa University, Addis Ababa, Ethiopia.

not self-healing whereas LCL is mostly seen as single lesions that are self-healing over time. During screening of an L. aethiopica genomic library with human antibodies [3], we discovered and isolated several clones that antigenically crossreacted with human IgG and IgM. Several reports have indicated that molecular mimicry between host and pathogen may be involved in the development of di¡erent clinical manifestations in various diseases, for example in acute rheumatic fever and reactive arthritis after enteric infection [4], in leprosy [5] as well as in AIDS [6]. In leishmaniasis, autoantibodies against several nuclear proteins have been found in

0928-8244 / 97 / $17.00 Copyright ß 1997 Federation of European Microbiological Societies. Published by Elsevier Science B.V. PII S 0 9 2 8 - 8 2 4 4 ( 9 6 ) 0 0 1 0 0 - 9

FEMSIM 710 9-4-97

12

A. Osland et al. / FEMS Immunology and Medical Microbiology 17 (1997) 11^19

high titres in sera from patients with visceral leishmaniasis [7] and rheumatoid factors (RF) are often found after several infections, including patients with visceral leishmaniasis [8] and new world mucocutaneous leishmaniasis [9]. However, no antigens from Leishmania that crossreact with human IgG have been identi¢ed. In this study we describe the identi¢cation and characterisation of L. aethiopica recombinant clones expressing multiple epitopes that speci¢cally bind to antibodies against human IgM and IgG. 2. Materials and methods

as previously described [3]. When the clones were tested with secondary antibody only, the ¢lters were incubated in the conjugate solution (anti-IgM or anti-IgG, diluted 1:4000 with OVA/TBS) for 1 h, washed with TBS and ¢nally incubated in NBT/ BCIP substrate solution (Promega, Madison, WI, USA). Lysogens were made and ammonium sulfate precipitates of the recombinant antigens were prepared as described [10], and the recombinant antigens were further puri¢ed by immunoa¤nity chromatography using anti-L-galactosidase columns (1 ml) (Promega) as described by the manufacturer. SDS-PAGE and Western blotting were carried out as described before [11,12].

2.1. Organisms and antibodies

2.3. DNA preparation and analysis

Parasite isolates GG (DCL) and K499 (LCL) were isolates from the district of Ocholo in Ethiopia. Promastigotes were grown in RPMI 1640 supplemented with 10% heat-inactivated fetal calf serum, 100 U/ml penicillin, 100 Wg/ml streptomycin and 2 mM L-glutamine at 25³C. The parasites were grown to stationary phase before harvesting. The antibodies used were goat anti-human IgG (anti-IgG) and goat anti-human IgM (anti-IgM), both were a¤nity puri¢ed IgG fractions and conjugated with alkaline phosphatase (Calbiochem-Behring, La Jolla, CA, USA). Human IgG was from Kabi-Pharmacia, Uppsala, Sweden and human IgM was from Sigma, St. Louis, IL, USA. All sera were obtained from the serum bank at Armauer Hansen Research Institute, Addis Ababa and were from patients who were clinically and histopathologically diagnosed with either DCL or LCL after the guidelines of the Research Committee, ALERT Hospital, Addis Ababa.

DNA from L. aethiopica was puri¢ed by CsClethidium bromide density gradient ultracentrifugation as previously described [3]. DNA from L. major, strain 5ASKH, was a generous gift from L. Al-Rustamani, London, UK. The DNA inserts of the V gt11 clones were ampli¢ed using the polymerase chain reaction (PCR) [3,13]. 106 ^107 phages from either isolated plaques or high titre lysates of each clone were added to 100 Wl of PCR bu¡er containing 1 mM each of deoxynucleotides and 100 nM V gt11 forward and reverse primers (Promega) and Taq polymerase (4 U) (Promega). Thirty cycles were performed using the following temperature pro¢le: 30 s at 95³C, 30 s at 55³C and 2 min at 68³C. The ampli¢ed inserts were subsequently analysed by agarose gel electrophoresis using HaeIII digested xX174 DNA (Pharmacia) as size markers. DNA probes were made from PCR ampli¢ed inserts of clones by random priming (Multiprime DNA labelling kit, Amersham, UK), with [32P]dCTP (3000 Ci mmole31 , Amersham) as substrate. Plaque lifts were performed as described [14] and prehybridisation (1 h at 42³C) and hybridisation (20 h at 42³C) conditions included 50% formamide, 5U Denhardt's mix, 5USSC, 0.1% SDS and 100 Wg/ml of denatured salmon sperm DNA. Filters were washed with 4USSC containing 0.1% SDS at 37³C. Autoradiograms were made using Kodak X-Omat ¢lm. Single-stranded templates for DNA sequencing of clones M and G were prepared by PCR ampli¢ca-

2.2. Screening of gene library and preparation of recombinant antigen

Genomic libraries of L. aethiopica isolates 080 (DCL) and 1467 (LCL) were made separately in V gt11 [3]. Brie£y, L. aethiopica DNA was completely digested with either AluI or RsaI. The two digests were mixed, EcoRI linkers (8-, 10- and 12-mers) added and the fragments ligated to V gt11 arms, with subsequent packaging and amplifying. The screening of libraries with antibodies was performed

FEMSIM 710 9-4-97

A. Osland et al. / FEMS Immunology and Medical Microbiology 17 (1997) 11^19

Fig. 1. Arrays of recombinant antigens from

L. aethiopica

13

clones probed with antibodies against human IgG and IgM. Drops containing

20^50 plaque forming units of four M clones and two G clones were arrayed on lawns of

E. coli,

blotted onto nitrocellulose ¢lters and

probed with anti-IgG and anti-IgM.

V

gt11 reverse primer (ob-

manufacturer. Peptide synthesis was performed on

tained from the Biotechnology Centre, University

an activated cellulose membrane with Fmoc amino

of Oslo, Norway) and were isolated using magnetic

acids [15]. Octapeptides, with an overlap of four ami-

beads with streptavidin (Dynal A/S, Oslo, Norway).

no acids, were synthesised and after blocking of the

tion with biotinylated

Sequencing was performed with T7 polymerase and primers (24-mers, forward and reverse) from Promega, using an automatic DNA sequencer. DNA sequencing was performed at the DNA sequencing laboratory, Biotechnology Centre, University of Oslo.

2.4. Test for the presence of M and G fragments in parasite DNA The M and G fragments were ampli¢ed from parasite DNA by PCR using the following primers : 5P-TTGTGTACCGTCTCGGGTTGG-3P and

(forward)

5P-CGATGTCTGCGAAAGTTTGTTGC-3P

(reverse) for the M fragment and 5P-TTCCTCCACTCGACATAGGGCTCG-3P (forward) and 5P-ATGGTGAAGGGCACAGCACAGACG-3P (reverse) for the G fragment (see Fig. 2). PCR was performed in 10 mM Tris pH 8.3, 50 mM KCl, 0.5

WM

primers,

0.2 mM each of dATP, dTTP, dGTP and dCTP and 30 units/ml of Taq polymerase (Gibco). The MgCl2 concentration was 2.5 mM when amplifying the M fragment and 1.75 mM for the G fragment. Ampli¢cation was carried out for 32 rounds of the following cycle : 30 s at 96³C, 50 s at 60³C and 1 min at 72³C. The fragments were analysed by agarose gel electrophoresis.

2.5. Epitope scanning

Fig. 2. DNA sequence of

Epitope scanning was performed using the SPOTs

L. aethiopica

inserts in clones M and

G. The translated polypeptides are shown below and the octapeptides that most likely contain the epitopes are underlined and

peptide synthesis kit (Cambridge Research Biochem-

marked. Primers used for ampli¢cation of parasite DNA and

icals, Ltd., Cambridge, UK) as recommended by the

cDNA are indicated above the nucleotide sequence.

FEMSIM 710 9-4-97

14

A. Osland et al. / FEMS Immunology and Medical Microbiology 17 (1997) 11^19

Fig. 3. Epitope scanning of rAgM (A) and rAgG (B). Overlapping octapeptides were synthesized as shown above and the peptide numbers correspond to the numbers of the spots below.

membrane, the octapeptides were tested for presence of crossreactive epitopes using anti-IgM and antiIgG (1:2000 in blocking bu¡er) and NBT/BCIP as substrate. 2.6. Expression of M and G genes

Expression of the M and G gene fragments was detected by reverse transcription (RT)-PCR, using the ABI Prism 310 Genetic Analyzer (Perkin Elmer, Foster City, CA) to detect DNA fragments ampli¢ed from speci¢c mRNAs. Parasites (200 Wl culture) were harvested by centrifugation, resuspended in 150 Wl RPMI medium and 50 Wl RPMI was added containing either 106 leukocytes (bu¡y coat), 40% normal human serum or no additions. The preparations were incubated at room temperature for 6 h, centrifuged and the cells were lysed and total RNA was isolated using the RNeasy kit (Qiagen Ltd., Dorking, UK). mRNA

was isolated from total RNA using magnetic beads with oligo-dT (Dynal A/S) as described by the manufacturers and E. coli tRNA (Sigma) was added to the preparations to a concentration of 1 Wg/ml. Reverse transcription of either total RNA or total mRNA was performed in 50 mM Tris pH 8.3, 75 mM KCl, 3 mM MgCl2 , 1 mM dithiothreitol, 1 mM each of dATP, dTTP, dGTP and dCTP, 120 Wg/ml random hexamers, 1.4 units/ml M-MLV reverse transcriptase (Gibco BRL, Paisley, UK) and 1400 units/ml RNasin (Promega) for 2 h. PCR of cDNA was performed as described for the ampli¢cation of the M and G fragments from parasite DNA except that £uorescence labelled (TET) forward primers (Perkin Elmer) were added to the PCR reactions to a concentration of 50 nM. After ampli¢cation, 1 Wl of the PCR reaction was added to 12 Wl water containing 0.5 Wl TAMRA Gene Scan 500 size standard (Perkin Elmer) and placed on the ABI Prism 310. Capillary electrophoresis was carried

FEMSIM 710 9-4-97

A. Osland et al. / FEMS Immunology and Medical Microbiology 17 (1997) 11^19

15

out with 2.5% Gene Scan polymer (Perkin Elmer) for 10 min using the GS native C module. 2.7. Test for anti-IgG and anti-IgM antibodies in patient sera

To test for antibodies of the IgM class against IgG, 5 Wg IgG was spotted on nitrocellulose ¢lters and blocked overnight with OVA/TBS containing normal goat serum (diluted 1:100). The ¢lters were then washed with TBS and incubated for 1 h with patient sera diluted 1:100 with OVA/TBS (six DCL and six LCL sera were tested). After washing with TBS, the ¢lters were incubated with anti-IgM conjugate (diluted 1:4000 in OVA/TBS) and processed as described for the screening procedure. The presence of antibodies of the IgG class against IgM was tested using the same procedure, but with 2 Wg IgM spotted on ¢lter and anti-IgG conjugate in the last incubation. 2.8. Computer analysis and nucleotide accession numbers

DNA and protein homology searches were done in EMBL 42 and SwissProt 31 using the program FASTA in PC-Gene 6.8 (Intelligenetics Inc., Mountain View, CA). The computer programs FSTPSCAN and PALIGN in PC-Gene were used to identify similar amino acid sequences in recombinant antigens and immunoglobulins. The G sequence has been assigned EMBL accession number X78587 and the M sequence EMBL accession number X78588.

Fig. 4. A: Alignment of the epitope containing octapeptides in rAgM (Epi M1^3) and rAgG (Epi G1^4) with sub-sequences in the W and Q chains. Also, the alignment of two sub-sequences detected by computer analysis (M8/MUC and EPI G2/MUC) are shown. The names of the H chains are from the entries in the SwissProt database, where MUC and GC are W and Q respectively. An asterisk marks identical amino acids. B: Alignment of the amino acid sequence of rAgG with sequences of human Q1.

combinant antigen from the G clones (rAgG) did not bind anti-IgM and the recombinant antigen from the M clones (rAgM) did not bind anti-IgG. SDSPAGE/Western blot analysis showed that puri¢ed rAgM and rAgG are L-galactosidase fusion proteins with apparent molecular masses of approximately 120 kDa.

3. Results

3.1. Detection of crossreacting recombinant antigens

3.2. Characterisation of the DNA inserts of clones M and G

During screening of the L. aethiopica V gt11 genomic library [3], several clones were found to bind secondary antibody conjugates (anti-IgG and antiIgM) without prior incubation with human sera to the recombinant antigens. Of a total of 51 clones that initially were isolated from the library by screening with human patient sera, six clones were shown to bind secondary antibody directly. Of these, four bound anti-IgM and two anti-IgG (Fig. 1). The re-

The sizes of the DNA inserts of the four M and two G clones were estimated by PCR ampli¢cation and agarose gel electrophoresis and the results showed that all the M clones have an identical size of approximately 150 bp, and both the G clones have an identical size of approximately 200 bp. DNA probes from one of the M and one of the G inserts were tested for crosshybridisation with the other clones on plaque lifts, and the results showed that

FEMSIM 710 9-4-97

16

A. Osland et al. / FEMS Immunology and Medical Microbiology 17 (1997) 11^19

Fig. 5. Capillary electrophoresis showing expression of the G fragment (A) and M fragment (B) in L. aethiopica. Ampli¢cation of the fragments was performed using cDNA made from total mRNA puri¢ed from L. aethiopica (GG, DCL) incubated with human leukocytes (left). The controls (right) show the ampli¢cations from the cDNA reactions without reverse transcriptase present. Capillary electrophoresis was run on the ABI Prism 310 as described in Section 2. The y-axis shows the relative £uorescence and the x-axis the size in bp, with the sizes of the Gene Scan 500 markers indicated.

the M probe crosshybridised with all the M clones, but not with the G clones, and the G probe crosshybridised with the second G clone, but not with the M clones. The DNA sequences of the M and G fragments are shown in Fig. 2. As the DNA sequencing was performed using PCR ampli¢ed inserts as template, the reading frame and orientation are known, and the translated polypeptides are shown below the DNA sequence in Fig. 2. The size of the M insert was found to be 154 bp coding for a 47 amino acid recombinant antigen, whereas the size of the G insert was found to be 206 bp, coding for a 68 amino acid recombinant antigen. Computer homology searches in the EMBL and SwissProt databases did not reveal sequences that showed any obvious homology with the nucleotide and amino acid sequences of clones M and G. Ampli¢cation of the M and G fragments from parasite DNA, using forward and reverse primers indicated in Fig. 2, showed that both these fragments were present in both DCL and LCL isolates of L. aethiopica as well as in L. major. The approximate sizes of the fragments generated were found to be 70 bp for the M fragment and 160 bp for the G fragment.

3.3. Detection of possible epitopes by peptide scanning

In order to identify oligopeptides that contain the crossreactive epitopes, epitope scanning of rAgM and rAgG was performed. As octapeptides were used to screen for epitopes, we also tried to identify octapeptides that most closely ¢t the signal patterns. The results, presented in Fig. 3, showed that three di¡erent epitopes present in rAgM (Epi M1^3) bound anti-IgM and four di¡erent epitopes in rAgG (Epi G1^4) bound anti-IgG. The octapeptides that were identi¢ed to contain these epitopes, based on signal pattern, are underlined in Fig. 2. The amino acid sequences of the octapeptides were compared with immunoglobulin sequences in the SwissProt database, in order to ¢nd similar subsequences in IgM and IgG. The most likely alignments of the octapeptides with the di¡erent immunoglobulin subsequences are shown in Fig. 4A. Epi G3 was found to be 100% identical with an octapeptide subsequence in the Q1 chain at the Gm(a) allotypic marker [16]. Similar subsequences are also present in the Q2, Q3 (Gm(non a)) and Q4 (Gm(Q4non a)) chains. Epi G4 shows 80% identity with a pentapeptide in Q3 chain, Epi G1 shows 60% identity with a pentapeptide subsequence in the Q1 and Epi M2 66% identity

FEMSIM 710 9-4-97

A. Osland et al. / FEMS Immunology and Medical Microbiology 17 (1997) 11^19

17

Fig. 6. Presence of antibodies against IgG in sera from patients with DCL and LCL. Six di¡erent DCL and six di¡erent LCL sera were tested.

with a hexapeptide subsequence in the W chain. The other octapeptides showed a lower degree of similarity with the immunoglobulin subsequences. Also, the di¡erent overlapping octapeptides in the two recombinant proteins were compared with the immunoglobulin sequences in the SwissProt database by computer analysis. A hexapeptide (HPQQTF) in rAgM (see M8 in Fig. 4A) was found to be 63% identical to a sequence in the W chain and the heptapeptide SSSSSTR (in Epi G2) was found to be 70% identical to a subsequence in the W chain (see Fig. 4A). Also a nonapeptide, covering both Epi G1 and Epi G2, was found to be 55% identical to a sub-sequence in Q1 (Fig. 4B). The alignment of the amino acid sequences of rAgG and Q1 is shown in Fig. 4B. A search in all entries in the SwissProt database for Epi G3 sub-sequence was performed. Only two entries, IgG1 from human and IgG2 from guinea pig, contained this sub-sequence. 3.4. Detection of speci¢c mRNAs

In order to investigate if the M and G fragments were expressed in the parasite, analysis by RT-PCR of puri¢ed total mRNA was performed. The primers used for ampli¢cation of the M and G fragments are shown in Fig. 2. The results, shown in Fig. 5, clearly show that speci¢c mRNAs, containing both the M and the G fragment, are detected after reverse transcription followed by PCR. Controls without reverse transcription gave much weaker signals. Also, ampli¢cations of cDNA made from total mRNA from leukocytes alone, or tRNA from E. coli, gave no signals (data not shown). The sizes of the fragments generated were found to be approximately 70 bp for

the M fragment and 160 bp for the G fragment (Fig. 5). Expression could also be detected by RT-PCR of total RNA, though in this case the signals caused by endogenous DNA were relatively stronger. Similar expression of the M and G fragments in total RNA from the LCL parasite isolate (K449) was also detected. 3.5. Presence of anti-immunoglobulin antibodies in patient sera

Preliminary tests for anti-immunoglobulin antibodies in patient sera were performed on six DCL patients and six LCL patients. The results showed that antibodies of the IgM class against IgG were predominantly present in sera from patients with DCL (Fig. 6). Such antibodies could be detected in ¢ve out of six DCL sera tested. In contrast, anti-IgG antibodies in sera from LCL patients could hardly be detected. Antibodies of the IgG class against IgM could not be demonstrated in any of these sera. 4. Discussion

Clones expressing polypeptides that antigenically crossreact with human IgM and IgG have been isolated from an L. aethiopica genomic library. Since the recombinant protein that binds anti-IgG does not react with anti-IgM and vice versa (see Fig. 1), the results indicated that this is not merely an unspeci¢c binding of goat IgG, but a speci¢c antigenantibody binding, discriminating between IgG and IgM. Both rAgM and rAgG were shown to contain sev-

FEMSIM 710 9-4-97

18

A. Osland et al. / FEMS Immunology and Medical Microbiology 17 (1997) 11^19

eral crossreactive epitopes. The degree of similarity with sub-sequences in IgM and IgG varied from 100% between an octapeptide in Q1 and Epi G3 to lower degrees of similarity. Whether the linear epitopes identi¢ed in the immunoglobulin heavy chains actually crossreact with the Leishmania epitopes remains to be tested. It might also be that the linear epitopes in the rAgG and rAgM mimic immunoglobulin conformational epitopes. However, Epi G3 is interesting being identical to an octapeptide covering residues speci¢c for the Gm(a) allotypic marker, especially because it has been shown that rheumatoid factors often have speci¢city against this marker [17]. The relatively low probability of detecting this octapeptide at random was also supported by the inability to detect this sub-sequence in the entries of the SwissProt database, except in two immunoglobulin sequences (human IgG1 and guinea pig IgG2 heavy chains). Thus, the presence of Epi G3 in Leishmania might indicate that this epitope has some functional property, and the possibility that multiple crossreacting epitopes are present in the two recombinant antigens might add to this. The actual size and which amino acids are involved in antibody binding in the di¡erent epitopes must await further analysis with peptides of smaller sizes as well as analysis with amino acid substitutions. The detection of speci¢c mRNAs was tested both in parasites alone as well as in parasites incubated with either human leukocytes or serum. In our hands the results were most consistent when analysing parasites that were incubated with human cells. This might be of signi¢cance, but might also re£ect a more e¤cient isolation of speci¢c L. aethiopica mRNA together with human leukocyte RNA (carrier e¡ect). Both the M and G fragments seem to be expressed in the parasite as speci¢c mRNAs, containing both sub-sequences, were detected by RTPCR. However, the expression seems to be rather low as endogenous DNA, containing the same fragments, was also detected in the controls without reverse transcription, though in a lower concentration. Perhaps certain factors are needed to induce expression of these genes. Whether the reading frames used by the parasites are the same as used for expressing the recombinant proteins in E. coli has not been established. However, at least for the G fragment, internal stop codons are introduced between the

primer sites using the two other reading frames (at bp 52 and 96). So, the G fragment is probably expressed in the parasite, giving rise to crossreacting epitopes, unless this sequence is part of the 5P-leader or the 3P-untranslated sequence. The detection of multiple immunoglobulin-crossreactive epitopes in recombinant proteins from Leishmania is a novel observation and these epitopes might constitute important factors in the complex interactions between the parasite and the immune system of the host [18]. Such parasite epitopes might potentially induce an autoimmune anti-immunoglobulin immune response in some patients. Autoimmunity provoked by molecular mimicry should occur only when the microbial and host determinants are similar enough to crossreact, yet di¡erent enough to break immunological tolerance [19,20]. Epi G3 should be especially interesting in this respect [17] as rheumatoid factors have been reported in both visceral [8] and new world cutaneous and mucocutaneous leishmaniasis [9]. Such autoantibodies against immunoglobulins might interfere with the network regulation of the immune response [21] and perhaps down-regulate important parts of the cellular response, thus making the host more vulnerable to the ongoing leishmanial infection. The possibility also exists that the epitopes might mimic some functional part of the immunoglobulins. However, the crossreacting epitopes described here are localised neither to the C1q binding site in IgG [22] nor to the binding site for the high a¤nity Fc receptor [23]. On the other hand, the IgM sequence that binds C1q has not yet been identi¢ed and the crossreacting epitopes might also interfere with other, less de¢ned, Fc receptors. Several models have been proposed on how infective microorganisms can take advantage of the immune response of its host to reach an equilibrium between host defence and parasite survival, thus creating a chronic state of the infection [4], and it is possible that the molecules described here might be involved in such processes. Crossreactions between immunoglobulins and microorganisms have been reported between hepatitis virus B and IgA [24] and between the SR protein of Streptococcus mutans and IgG [25], but to our knowledge this is the ¢rst report that identi¢es recombinant Leishmania polypeptides that crossreact with immunoglobulins. To further characterize these

FEMSIM 710 9-4-97

A. Osland et al. / FEMS Immunology and Medical Microbiology 17 (1997) 11^19 crossreactions, work is in progress to test the speci¢cities of the rheumatoid factors present in sera from DCL patients using well de¢ned peptides and also to

19

[10] Huynh, T.V., Young, R.A. and Davies, R.W. (1984) Construction

and

screening

cDNA

libraries

in

V

gt10

and

V

gt11. In : DNA Cloning Techniques : A Practical Approach (D. Gover, Ed.), pp. 49^78. IRL Press, Oxford.

isolate cosmid clones carrying the M and G DNA

[11] Laemmli, U.K. (1970) Cleavage of structural proteins during

fragments in order to characterise the whole genes

the assembly of the head of bacteriophage T4. Nature 227,

and perhaps get closer to their functional properties.

680^685. [12] Towbin, H., Staehlin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellu-

Acknowledgments

lose

sheets :

Procedure

and

some

applications.

Proc.

Natl.

Acad. Sci. USA 76, 4350^4354. [13] Saiki, R.K., Gelfand, D.H., Sto¡el, S., Scharf, S.J., Higuchi,

We

wish

to

express

our

appreciation

to

M.S.

Wright for help with the DNA sequencing. Armauer Hansen Research Institute was supported by Agencies for International Development of the Swedish and Norwegian Governments (SIDA and NORAD).

R., Horn, G.T., Mullis, K.B. and Erlich, H.A. (1988) Primer directed enzymatic ampli¢cation of DNA with a thermostable DNA polymerase. Science 239, 487^491. [14] Maniatis, T., Fritsch, E.F. and Sambrook, J. (1989) Molecular Cloning : A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. [15] Blankenmeyer-Menge, B., Nimtz, M. and Frank, R. (1990) An e¤cient method for anchoring Fmoc-amino acids to hy-

References

droxyl-functionalised

solid

supports.

Tetrahedron

Lett.

31,

1701^1704. [16] Kabat, E.A., Wu, T.T., Perry, H.M., Gottesman, K.S. and

[1] Bray, R.S., Ashford, M.A. and Bray, M.A. (1973) The parasite causing cutaneous leishmaniasis in Ethiopia. Trans. R. Soc. Trop. Med. Hyg. 67, 45^348.

(1987) High prevalence of elephantiasis and cutaneous leishmaniasis in Ocholo, south west Ethiopia. Ethiop. Med. J. 25, 203^207.

Isolation and characterization of recombinant antigens from

Leishmania aethiopica

that react with human antibodies. In-

fect. Immun. 60, 1368^1374.

tors reacting with `non a' and other antigens of native autologous IgG. J. Immunol. 105, 928^937. [18] Dessaint, J.P. and Capron, A. (1989) Immunode¢ciencies in

and autoimmunity. Immunol. Ser. 55, 285^313.

van der Veer, E., Chin, R.A., Lien, A. and Van Joost, T. (1992) Similarity between mycobacterial and human antigens. Br. J. Dermatol. 127, 352^9358.

from a disequilibrium between two complementary groups of molecules that mimic MHC ? Immunol. Today, 17, 114^119. [7] Argov, S., Ja¡e, C.L., Krupp, M., Slor, H. and Shoenfeld, Y. production

by

patients

infected

with

Clin. Exp. Immunol. 76, 190^197.

Rocha, M. (1983) Circulating immune complexes and rheuschistosomiasis

Wiktor,

T.J.,

Koprowski,

and

visceral

leishmaniasis.

Am. J. Trop. Med. Hyg. 32, 61^68.

normal tissues. J. Virol. 57, 397^401. [21] Jerne, N.K. (1984) Idiotypic networks and other preconceived ideas. Annu. Rev. Immunol. 79, 5^24.

on IgG. Nature 332, 738^740. [23] Duncan, A.R., Woof, J.M., Partridge, L.J., Burton, D.R. and Winter, G. (1988). Localization of the binding site for the human high-a¤nity Fc receptor. Nature 332, 563^564. [24] Pontisso, P., Ruvoletto, M.G., Tiribelli, C., Gerlich, W.H.,

B virus and IgA cross-react in their binding to the hepatocyte surface. J. Gen. Virol. 73, 2041^2045. [25] Ganglo¡, S., M'Zoughi, R., Lett, E., Scholler, M. Baer, J.

Streptococcus mutans

SR protein and human IgG crossreac-

pron, A. (1980) Circulating immune complexes and anti-IgG

tive

antibodies

thetic peptides. J. Immunol. 148, 3249^3255.

mucocutaneous

Oldstone,

Pini, A. Ogier, J. and Kein, J.P. (1992) Epitope mapping of

[9] Desjeux, P., Santoro, F., Afchain, D., Lovens, M. and Ca-

in

H.,

Ruol, A. and Alberti, A. (1992) The preS1 domain of hepatitis

[8] Carvalho, E.M., Andrews, B.S., Martinelli, R., Dutra, M. and

in

M.J.,

M.B.A. and Notkins, A.L. (1986) Molecular mimicry : Fre-

[22] Duncan, A.R. and Winter, G. (1988) The binding site for C1q

[6] Su ë sal, C., Daniel, V. and Opelz, G. (1996) Does AIDS emerge

factor

disease. Cell 50, 819^820. [20] Srinivasappa, J., Saegusa, J., Prabhakar, B.S., Gentry, M.K.,

quency of reactivity of monoclonal antiviral antibodies with

[5] Van den Akker, T.W., Naafs, B., Kolk, A.H., De Glopper-

Autoantibody

[19] Oldstone, M.B.A. (1987) Molecular mimicry and autoimmune

Buchmeier,

[4] Abu-Shakra, M. and Shoenfeld, Y. (1991) Chronic infections

matoid

of Proteins of Immunological

parasitic diseases. Immunode¢c. Rev. 1, 311^324.

[3] Osland, A., Beyene, D., Ashena¢, S. and Beetsma, A. (1992)

Leishmania.

Sequences

Interest. NIH, Bethesda, MD. [17] Gaarder, P.I. and Natvig, J.B. (1970) Hidden rheumatoid fac-

[2] Mengistu, G., Humber, D.P., Mulugeta, E. and Tadesse, M.

(1989)

Foeller, C. (1991)

leishmaniasis.

Am.

J.

Trop.

Med. Hyg. 29, 195^198.

FEMSIM 710 9-4-97

determinants,

by

using

recombinant

proteins

and

syn-