IDENTIFICATION OF IMMUNODOMINANT 2-RY~~~~~U~~ MUSCULI ANTIGENS RECOGNIZED BY MONOCLONAL ANTIBODY AND CURATIVE IMMUNOGLOBULIN G2a ANTIBODY K. T. Y. SHAW,*~ I. T. SHAW,$ P. RYAN,* M. M. STEVENSON* and P. A. L. KONGSHAVN* *Montreal General Hospital Research Institute, 1650 Cedar Avenue, Montreal, Quebec, Canada H3G lA4 and Departments of Physiology and Medicine, McGill University, Montreal, Quebec, Canada H3G lY6 $Department of Neurology and Neurosurgery, Montreal Neurological Institute, Montreal, Quebec, Canada H3A 2B4 (Received 24 July 1991; accepted 11 February 1992) Ah&PC&-SHAW K. T. Y., SHAWI. T., RYANP., STEVENSON M. M. and KONGSHAVN P. A. L. 1992. Identification of immunodominant Trypanosoma musculi antigens recognized by monoclonal antibody and curative immuno~obulin G2a antibody. ~nter~t~onal Journa~~orF~r~~tology 22: 603-612. Try~ffnosorna musculi obtained from normal or irradiated (900 rad) hosts or from in vitro cultures were lysed and analysed by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE). Similar protein banding patterns with a molecular weight (mol. wt) range from 34 to 68 kDa were observed between the two bloodstream forms. In comparison, lysates of cultured parasites showed a unique banding pattern of antigens within the same mol. wt range. Western blot of bloodstream form lysates, probed with immune plasma (IP), revealed a wide range of parasite proteins. However, when probed with the IgGZa-enriched fraction of IP, a major band of approximately 66 kDa was detected on the blot. Several bands of higher mol. wt were also observed. When anti-T. musculimonoclonal antibodies were used to probe the blot, the 66 kDa protein was again recognized. Using indirect fluorescence, live bloodstream form parasites were analysed by flow cytometry and the ~66 protein was determined to be a surface molecule. Finally, lysates of ?Smethionine-la~lled trypanosomes were immunopr~ipitated with Sepharose linked anti-T. musculi monoclonal antibodies and the eluted ligand analysed by SDS-PAGE and autoradiographed. The 66 kDa band was identified, therefore confirming that this protein was of parasite origin. INDEX KEY WORDS: Trypanosomu muscuii; antigen; antibody.
INTRODUCTION Trypanosoma musculi, a natural protozoan parasite of
mice, causes a self-limiting infection of approximately 3 weeks duration (Targett & Viens, 1975; Viens, Targett, Leuchars & Davies, 1974). During the course of parasitaemia, three phases may be distinguished: an exponential growth phase, a plateau phase and an elimination phase. During this Iattermost phase, specific immunity is activated and mediated through an as yet undefined mechanism. Previous studies from our laboratory have shown that this process is B cell dependent (Vargas, Del, Viens & Kongshavn, 1984). The necessity for antibodies in the curative m~hanism was demonstrated in experiments whereby the passive transfer of immune plasma obtained from a cured mouse was found to be able to mediate trypanosome
~To whom all correspondence should be addressed at: Montreal General Hospital Research Institute, 1650 Cedar Avenue, Montreal, Quebec, Canada H3G lA4.
clearance in vivo and to mediate trypanosome killing in vitro (Wechsler & Kongshavn, 1984). In order to identify the curative element, immune plasma was initially purified using protein A Sepharose and the active fraction found to be enriched in immunoglobulins of the subclasses G2a and G3 (Wechsler & Kongshavn, 1985). Upon further purification using the curative fraction was affinity columns, characterized as being an immunoglobulin of the IgG2a subclass (Wechsler & Kongshavn, 1985). The identification of parasite antigens by the host, especially those recognized by a curative antibody, is an initial step towards understanding host-parasite interactions. In earlier studies, Khazindar & Dusanic (1982) reported the serological and antigenic differences between bloodstream and culture forms of T. musculi, Dywer & D’Alesandro (1976a,b) have shown that the surface of T. musculi, like other trypanosome species, appears to be covered by glycoprotein. More recently, Samarawickrema & Howell (1990) have examined the surface coat of bloodstream 403
K. T. Y. SHAW et al.
604
forms of the parasite and have identified two stagespecific polypeptides. Apart from these studies, little else is known about these trypanosomes. The objective of the present study, therefore, was to examine the antigens expressed on T. musculi, both bloodstream and culture forms, and to identify those antigen(s) recognized by the curative IgG2a antibody. MATERIALS
AND METHODS
Mice and parasites. For the present studies, a T. musculi clone was stabilized according to standard procedure, in
order to establish a line of parasites with identical antigenic minimizing components, thereby the presence of antigenically different variants in the source of trypanosomes being used. A single trypanosome was expanded in irradiated &lo-week-old male BlO.D2/oSn mice (Jackson Laboratories Inc., Bar Harbor, ME) and, at peak parasitaemia (lo6 parasites per ml blood), the mice were sacrificed and blood from each mouse divided into 1 ml aliquots. One clone was arbitrarily selected and used for all experiments. Female C3H/HeN retired breeders (Charles River, St. Constant, Quebec, Canada) were infected intraperitoneally with 5 x IO4 trypanosomes and used for the preparation of parasite antigen. The continuance of the stabilate was maintained through biweekly passage of 5 x lo4 viable T. musculi injected intravenously into &IO-week-old male A/J mice (Jackson Laboratories Inc., Bar Harbor, ME) for no more than 3 months. The parasites were enumerated as previously described (Wechsler & Kongshavn, 1985). T. musculi antigen. The parasites used for the preparation of antigen were obtained from three sources: bloodstream forms (BSF) from an untreated host, bloodstream forms from an irradiated host (BSF-X) and culture forms (CF). The blood from normal infected animals was collected from the retroorbital sinus of infected C3H/HeN mice 9-12 days postinoculum (p.i.), and the trypanosomes were partially purified from red blood cells by differential centrifugation (Wechsler & Kongshavn, 1985). They were further purified by passage through a DEAE-cellulose (Whatman Chromedia DE52, Whatman BioSystems Ltd, Maidstone, Kent, U.K.) column to separate out contaminating white blood cells (Lanham & Godfrey, 1970). The trypanosomes from irradiated hosts were purified similarly from blood taken 8-12 days p.i. Culture forms of T. musculi were prepared according to methods previously described (Wechsler & Kongshavn, 1986). The three preparations were each resuspended in phosphate-buffered saline glucose, pH 8.0 at a concentration of 10’ trypanosomes per ml and to each one was added an equal volume of lysis buffer [l% SDS, I mhl-Tris, 150 mMNaCI, 2.5 mr+EDTA, 0.05% sodium azide, 1 mMphenylmethylsulphonyl fluoride (PMSF)]. The suspension was stored at - 20°C and pelleted (25Og for 15 min) before use. Plasmapreparation.Immune plasma (IP) obtained from a cured mouse was prepared using methods previously described (Wechsler & Kongshavn, 1985). Irradiation. Mice used for the preparation of a T. musculi clone were irradiated I day prior to use with 500 rad using methods described elsewhere (Ulczak, Ghadirian, Skamene, Blackwell & Kongshavn, 1989).
Preparation of immune plasma fractions. IP was separated into IgG2a-enriched and IgGZa-depleted fractions using methods previously described (Wechsler & Kongshavn, 1986). Monoclonal antibodies. Anti-T. musculi monoclonal antibodies were raised using the method of Oi & Herzenberg (1980). The monoclonal antibodies were linked either to Pierce AminoLink beads (Pierce, Rockford, IL) or to Sepharose 4B beads (Pharmacia Canada Inc., Baie D’Urte, Quebec, Canada) according to manufacturer’s instructions. F23.1, an anti-T cell receptor monoclonal antibody (Staerz, Rammensee, Benedetto & Bevan, 1985) linked to Sepharose was provided by Dr T. Owens, Montreal Neurological Institute and Hospital, Montreal, Quebec, Canada. JR/ MAb, an anti-DNA IgGl monoclonal antibody was provided by Dr J. Rauch, Montreal General Hospital Research Institute, Montreal, Quebec, Canada. SDS-PAGE. T. musculi lysates were run on SDS-PAGE gels (10.0%) under reducing conditions using the discontinuous buffer system of Laemmli (1970). The samples were diluted with equal volumes of 2 x sample buffer (2-Bmercaptoethanol, 0.0625 hr-Tris-HCl, pH 6.8, 10% SDS, 10% glycerin) and boiled at 100°C for 2-3 min before loading. One lane containing 1 pg of standard mol. wt protein markers (Bio-Rad, Richmond, CA) was diluted in 10 ~1 of sample buffer. Electrophoresis was performed at 125 V for 75 min in a Bio-Rad Mini Protean II Slab cell. The gels were either immunoblotted as described below, or stained with 5% Coomassie blue in 10% acetic acid, 40% methanol, and then destained with 10% acetic acid, 50% methanol. Immunoblotting. Following electrophoresis, proteins of the bloodstream form preparation were transferred onto nitrocellulose by electroelution at 150 mA using the Bio-Rad Trans-Blot apparatus (Bio-Rad, Richmond, CA). Two lanes, one containing standards and the other containing parasite proteins, were cut from the blot and the efficiency of transfer was determined by amido black stain. The remaining nitrocellulose blot was blocked with 5% casein, 0.25 M-Tris, pH 7.2, 0.15 M-NaCI for 1 h at 37’C. The lanes within each blot were cut, washed three times with wash buffer (0.25 MTris, 0.15 M-NaCl, pH 7.2) and incubated with whole or fractionated IP (1:lOO dilution) or monoclonal antibodies (I:10 dilution) overnight at 37°C. After washing three times with wash buffer, the strips were immersed in 5% casein, PBS to which was added goat anti-mouse Ig (I:1000 dilution, Southern Biotechnology Associates Inc., Birmingham, AL) and incubated for 3 h at 37°C. The replicas were rinsed five times with the wash buffer and bound antibody was detected using AP colour development, Fast Red (Bio-Rad, Richmond, CA). Immunoprecipitation and autoradiography. Internal labelhng of trypanosomes and subsequent immunoprecipitation were performed using the method described by Parish, Morrison & Pearson (1985) with minor modifications. Briefly, 0.5 ml of a T. musculi preparation (partially purified as described earlier, 10’ per ml) was added to each well of a six-well plate (Limbro, Flow Laboratories Inc., Mclean, VA) containing 4.5 ml of RPM1 1640 containing only 5% of normal methionine concentration (Gibco Laboratories Life Technologies Inc., Grand Island, NY). An aliquot of 200 &i
T. musculi
antigens identified by antibody
of’%-methionine (specificactivity 0.6-4 Ci rnM_‘,Amersham Corp., Oakville, Ontario, Canada) was added to each well and the plate was covered and incubated for 16 h at 37°C. After incubation, the radiolabelled trypanosomes were harvested, pooled, washed with RPM1 1640, centrifuged at 1OOOgfor 10 min and resuspended in 200 ~1 of the same medium. In order to lyse the parasites, 300 ~1 of lysis buffer [5 mM-Tris-HCl, pH 9.2, 1 mM-EDTA, 1% Nonidet P-40 (NP40) 1 mM-PMSF and 0.1 mM-tosyl-Llysine chloromethylketone (TLCK)] was added and the suspension incubated at 37°C for 15 min. The insoluble material was removed by microcentrifugation and the supernatant dialysed against 2 1of Tris-EDTA-NaC1 (TEN) buffer, pH 8.5 (20 mM-Tris-HCl, pH 7.4, 1 mi+EDTA, 100 mM-NaCl and 0.1% NP-40) at 4°C for 24 h. The dialysate was until use. microcentrifuged and stored at -70°C Immunoprecipitations were performed using 200 pl aliquots of radiolabelled lysates. A 100 pl sample of 50% (v/v) AminoLink or Sepharose linked anti-Z’. musculi monoclonal antibodies was microcentrifuged for 5-10 min and the overlying buffer removed. Sepharose linked anti-T cell receptor monocolonal antibodies were used as control and prepared similarly. Aliquots of the trypanosome lysate were added to each set of beads and lightly vortexed before incubating on ice for 2.5 h. After incubation, each mixture was washed three times with TEN buffer, resuspended in 200 ~1 of the same buffer and placed on microcolumns (Pearson & Young, 1980). The column was dried by centrifugation at 750g for 5 min after which 4t3-60 ~12 x SDS sample buffer was added and the column was incubated at room temperature (RT) for 30 min. Finally the column was recentrifuged for 10 min and the eluted fraction collected and analysed by SDS-PAGE. The gel was stained, dried and autoradiographed on Kodak X-OMAT S film. Dor blot. A sample of bloodstream form trypanosomes lysate was divided into two 100 ~1 aliquots. Each sample was microcentrifug~ for 5 min and its overlying supernatant removed. They were then resuspended in 100 ~1 of lysis buffer f 10% SDS. A portion of each sample was aliquoted into additional samples of 1:5 and 1:50 dilution. Nitrocellulose strips (0.2 cm in width x 7 cm in length) were cut and 2 ~1 of each sample (l:l, 1:5, 1:50 f SDS) was applied, using an eppendorf pipette, onto each strip. The nitrocellulose strip was allowed to dry and the samples were applied twice more using the same procedure. After blocking for 1 h with 10% goat serum (Gibco Laboratories Life Technologies Inc., Grand Island, NY) in PBS, pH 7.4, each strip was washed three times with wash buffer (described above) and incubated with whole or fractionated IP or NMP or anti-T. mus~ff IgGfa monoclonal antibody or anti-DNA IgGl monoclonal antibody (at the dilutions described above in 1% goat serum/ PBS) overnight at 37°C. The strips were subsequently washed three times and incubated with goat anti-mouse Ig (1:tOOO dilution in 1% goat serum/PBS) for 3 h at 37°C. The blots were rinsed five times with wash buffer and developed using AP colour development, Fast Red. ~rnrn~~o~~~rescence and cyto~~oromet~y. Bloodstream form trypanosomes were probed indirectly with fluorescent antibodies and analysed by flow cytometry using the method described by Cashman, Loertscher, Nalbantoglu, Shaw,
605
Kascsak, Bolton & Bendheim (1990). Briefly, fresh purified parasites were washed with PBS (PH 7.0) and divided into five aliquots of lo6 trypanosomes. All samples were washed three times with PBS between each subsequent incubation. The samples were incubated with 10% normal goat serum for 30 min to reduce non-specific immuno~obulin binding. The aliquots were then incubated at 4°C for 30 min with either anti-T. musculi IgG2a monoclonal antibody, IP or NMP or anti-CD4 monoclonai antibody (Leu 3a, Becton Dickinson, Mountain View, CA) at I: 1 dilutions. After incubation for 30 min with the appropriate second antibody, FITC-labelled anti-mouse immunoglobulin (Tago Immunologicals, Inter Medico, Markham, Ontario, Canada), the tryp~osomes were analysed by flow cytometry with an argon laser FACScan instrument equipped with Consort 30 and LYSYS II software (Becton Dickinson, Mountain View, CA). As a control, samples of freshly obtained human peripheral lymphocytes were run concurrently with the parasite samples under similar conditions. RESULTS SDS-PAGE analysis of lysedparasite preparations In order to characterize the antigens of T. musculi, parasite lysates were subjected to SDS-PAGE. This
method allowed the visual determination of differences between the three types of parasite preparations. After electrophoresis, analysis of each Iysate revealed numerous bands covering a wide range of molecular weights in all samples (Fig. 1). Bloodstream form lysates from normal mice (BSF, lane 1) have a complex pattern enriched in molecules of mol. wt 34-68 kDa. Similarly, bloodstream forms from an irradiated host (BSF-X, lane 2) have an almost identical pattern. The majority of bands in both lanes migrate at the same mobility. In comparison, culture forms (lane 3) of T. muscufi have a different banding pattern ofmol. wt 36 68 kDa. Many of these antigens are shared with the other two preparations; however, unique bands are also visible. The culture forms appear to be quantitatively restricted in proteins of very high or very low mol. wts. Western blot analysis using immuneplasma
Following protein separation by SDS-PAGE, T. musculi antigens were transferred onto nitrocellulose. Western blots were probed with either whole or fractionated IP in order to identify the immunogenic parasite proteins (Fig. 2). On immunoblots of trypanosome lysates (T) probed with IP, a series of bands was revealed. In contrast, the curative IgG2aenriched fraction (ELU) of IP identified with high intensity a single distinct band migrating at 66 kDa. Several higher mol. wt bands also showed faint staining. A similar pattern was observed using IgG2adepleted fractions (FT) of IP. When lysates were probed with normal mouse plasma (NMP), no bands
K.
T. Y.
SHAW et al.
FIG. 1. SDS-PAGE analysis of whole T. musculi lysates. Preparations of bloodstream form trypanosomes from normal mice (BSF), bloodstream form trypanosomes from irradiated mice (BSF-X) and culture form trypanosomes (CF), were lysed and separated on a 10% gel under reducing conditions. Mol. wt markers are indicated on the left of the gel.
IP
FT
ELlJ
IP
NMP
.
..
97.466.2i’
.
,w *
42.7-
,-
*;a
.
I
f
T
T
1
T
M
A
Frc. 2. Western blot analysis of T. musculi lysates probed with immune plasma and its fractions. After SDS-PAGE separation, trypanosome lysates (T) were transferred onto nitrocellulose and probed with whole immune plasma (IP), IgGZa-depleted IP fraction (FI) and IgG2a-enriched IP fraction (EIJJ). A control lane was probed with normal mouse plasma (NMP). Immunoblots of mouse albumin (A) and malaria antigens (M) were also probed with IP. Mol. wt standards are indicated on the left and were superimposed from a portion of the blot that was stained with amido black.
T. musculi antigens identified by antibody
El
60’7
DS
97.4-
66.2-
. T
1
FIG. 3. Detection of T. musculi antigens using mono-
42.7-
clonal antibodies. Trypanosome lysates (T) were separated by SDS-PAGE, transferred onto nitrocellulose and probed with anti-T. musculi monoclonal antibodies El, H2, CS and D5. Mol. wt standards are indicated on the left and were
superimposed from a portion of the blot that was stained with amido black. 31 .o-
were detected. As further controls, mouse albumin {A) and a lysate of malaria antigens (M) were transferred onto nitrocellulose and probed with immune plasma. These blots exhibited negative binding, confirming that the specificity of immune plasma and its fractions was only towards T. musculi antigens. Western blot analysis using monoclona~ antibodies
In this series of experiments, high affinity anti-T. antibodies of three different IgG sub-classes (El and H2 are both of the IgG2a subclass while C6 and DS are of the IgG3 and IgGl subclass, respectively) were used to probe the Western blots of trypanosome lysates (Fig. 3). Interestingly, the pattern of binding is identical in all four preparations. The major band of approximately 66.2 kDa was distinctIy recognized. A small doublet of lower mol. wt was also faintly detected. No reaction was observed using a control monoclonal antibody (an IgG2a antibody to T cell receptor, data not shown) nor when preparations of mouse albumin and lysed malaria antigen were substituted for the trypanosomes. The reactivity of these monoclonal antibodies is, therefore, restricted to a single T. muscu~~antigen.
musculi monoclonal
of radiolabe~led T. musculi In order to confirm that the 66 kDa band was of parasite, not host, origin, a preparation of “Smethionine-labelled trypanosomes was lysed and ~mmunopreeipitation
FIG. 4. Autoradiograph of [‘SS]-methionine-labelled proteins of whole T. muxuli and of immunoprecipitates obtained by using AminoLink and/or Sepharose linked anti-l: musculi monoclonal antibodies. Lunea, whole T. musculilysate. Lane 6, anti-T. musculi monoclonal antibody imm~opre~ipitate. Lane c, anti-T cell receptor monoclonal antibody immunopr~pitate. Mol. wt standards are indicated on the left and were superimposed from a portion of the gel that was stained with Coomassie Brilliant blue.
immunoprecipitated with anti-T. musculi monoclonal antibody linked to AminoLink or Sepharose beads. Following SDS-PAGE and autoradiography, a series of bands was revealed when the complete lysate was analysed (Fig. 4, lane a). However, the 66 kDa band was detected, as well as several other bands including a 50 kDa protein, in the immunoprecipitate obtained from the anti-T. muscufi monoclonal antibodies. This
K. T. Y.
608
Determination
of Native
in Lysed T.musculi
SHAW
et al.
and Non-native
Antigen
Preparation
Epitopes (~SDS)
1:1(-I (-1
noSDS
(+)
SDS
1:l (+I
1:5(-J
1:5(t)
1:50(-I
1:5o(+)
FIG. 5. Dot blot analysis of T. musculi antigens. Lysates of bloodstream form trypanosomes were resuspended in 100 ~1 aliquots f 10% SDS. The samples were diluted (1: 1, 1:5, 1:50) and applied to nitrocellulose strips. The strips were probed with whole immune plasma (IP), IgG2a-enriched IP fraction (ELU), IgGZa-depleted fraction (FT) and anti-T. musculi IgG2a monoclonal antibody (MAb). Control lanes were probed with normal mouse plasma (NMP), anti-DNA IgGl monoclonal antibody (JR/MAb), second antibody alone (2”) or buffer (buffer).
lower mol. wt protein could possibly be a breakdown product of the 66 kDa molecule. Sepharose linked anti-T cell receptor monoclonal antibodies were also incubated with the trypanosome lysate as a negative control and analysis of the immunoprecipitate did not detect any proteins (lane c). Dot blot analysis of T. musculi antigens In order to determine the presence of native or SDS epitopes, lysates of T. musculi containing no or 10% SDS were analysed by dot blot (Fig. 5). When the nitrocellulose strip was probed with whole IP, a positive dot was revealed in the undiluted SDS-free sample. This identical dot was also detected on the strip probed with IgG2a-enriched IP fraction (ELU) and it was faintly recognized on the strip probed with the FT fraction. When the blot was probed with antiT. musculi monoclonal antibody (MAb), the nonSDS, undiluted trypanosome sample was also
revealed. In parasite samples which were either diluted or contained SDS, no dot was observed. As controls, the samples were also probed with NMP or anti-DNA monoclonal antibody (JR/MAb) but they did not show any binding. Flow cytometry analysis of T. musculi In order to confirm the surface location of p66 protein, the trypanosomes were indirectly labelled with fluorescent antibody and analysed by flow cytometry (Fig. 6). The population of trypanosomes was selected for viability (indicated by propidium iodide uptake, data not shown). The positive detection of surface protein was characterized by a change in histogram profile from control data (T. muscufi + FITC alone, Fig. 6A). An increase in protein concentration was indicated by a shift in histogram profile with a corresponding increase in mean fluorescence. When the parasites were incubated with anti-
609
T. mu.m& antigens identified by antibody
B
1
E
7 I
F
H
FIG. 6.
K. T. Y. SHAW et al.
610
T. ~us~uli monoclonal antibody (which specifically recognizes the ~66 protein), a unique histogram profile was observed (Fig. 6C). The low mean fluorescence indicated that the parasite population in general expressed ~66 in low abundance. As a positive control, parasites incubated with IP also showed a similar profile but was greatly shifted towards the right indicating greater recognition of a larger quantity of proteins (~66 and others) (Fig. 6E). In control samples, namely trypanosomes incubated with NMP (Fig. 6G) or anti-CD4 mon~lonal antibody (data not shown), there was no significant change in histogram profile or mean fluorescence. As a further control, human lymphocytes were incubated with anti-T. musculi monoclonal antibody (Fig. 6D); there were no detectable changes in histogram profile from that observed in the sample containing cells and FITC alone (Fig. 6B). This confirms that the identification of ~66 on the trypanosomes, as indicated by a change in histogram profile, was not an artifact caused by the addition of antibody. A shift in histogram profile was observed in samples of cells incubated with IP {Fig. 6F); this was not significant as an identical shift was also seen in cells incubated with NMP (Fig. 6H). DISCUSSION
In order to examine the antigens of T. musculi, we analysed three preparations of parasites obtained from different sources: normal mice, irradiated mice and in vitro cultures. Each preparation contained different proportions of T. rn~&~li at each of the various stages of growth. Bloodstream forms obtained 9-12 days p.i. at the height of the plateau phase of infection, were exclusively monomorphic adult trypomastigotes. In vitro cultures, in contrast, have a 100% multiplicative parasite population. A mixed popufation of 35% young and dividing forms and 65% adult forms was observed in the blood of irradiated animals. The molecular nature of the proteins in each lysate was determined by SDS-PAGE analysis. The antigenic components of both bloodstream forms appeared simiiar but varied in the intensity of staining of some bands. This was related to the differences in protein concentrations between the lysates. Analysis of culture forms also revealed a banding pattern containing a variety of proteins, but, unlike either bloodstream form preparations, lacked the very high and very low mol. wt proteins. Overall,
the antigens from multiplicative forms and adult forms appear to have some shared antigens as well as unique proteins. This is not unusual and has been observed in other trypanosomes and parasite organisms (Medina-Acosta, Karess, Schwartz & Russell, 1989; Parish et al., 1985; Scharfstein, Schechter, Senna, Peralta, Mendonqa-Previato & Miles, 1986). Having established that the lysates contain a wide range of proteins, the next step was to identify those recognized by the curative IgG2a antibody. During the elimination phase of parasitaemia, the effective acquired immune response is directed against the adult parasite forms, therefore, in all Western blots, bloodstream forms from normal mice (100% adults) were examined with immune plasma and its fractions. In the immunoblots probed with IP, a host of antigens was recognized and the banding pattern was observed to be similar to the SDS-PAGE of whole lysates. This was not surprising since we would expect the host to be able to elicit an antibody response against a wide variety of parasite antigens. When IP was separated into fractions, blots probed with the curative IgGZaenriched fraction revealed a major protein of 66 kDa (~66). Several higher mol. wt bands were also visible. When probed with the IgGZa-depleted fractions, a similar pattern was observed suggesting that other specific antibody isotypes present in the plasma also recognized the ~66. However, we have been able to show that these antibodies depleted of the IgG2aenriched fraction are not effective in elimination of parasitaemia in vivo (Wechsler & Kongshavn, 1984). Samarawickrema & Howell (1990) have also reported the identification of a 66-70 kDa molecule recognized by radioimmunoprecipitation and affinity-eluted immune serum. A second polypeptide of higher mol. wt, 88-92 kDa, was also revealed, thus substantiating our results. Since high mol. wt proteins are visible on the SDSPAGE and Western blots, the ~66 could be a subunit of a large aggregate structure. In order to determine the unit structure of the protein, the trypanosome lysate was passed through a high pressure liquid chromatography gel filtration column (data not shown). A bovine serum albumin standard (66 kDa single unit protein) was passed through the column initially as a control and its retention time in the column determined. The active trypanosome fraction obtained from this purification (as detected by
FIG. 6. Flow cytometry analysis of T. muset&.Fluorescence histograms of trypanosomes gated from a freshly isolated sample (A, C, E, G) and lymphocytes from human peripheral blood (B, D, F, H). The samples were incubated with FITC alone (A, B), anti-T. musculimonoclonal antibody + FITC (C, D), IP + FITC (E, F) or NMP + FITC (G, H). The ordinate axis represents the number of cellular events and the abscissa represents the arbitrary mean fluorescence units as determined by FACScan software.
7’. ~~~u~i antigens
enzyme-linked immunosorbent assay) eluted with an identical retention time as the standard. This implies that ~66 has a monomeric structure. In other experiments (data not shown) IP was used to probe Western blots of lysates obtained from cultured trypanosomes and from irradiated hosts. As with results described above, the ~66 as well as several other antigens were identified in both preparations. In finding p66 in the culture form preparation, one can assume that it is not an adult stage specific molecule. These results do not support the studies by Khazindar & Dusanic (1982) who showed that immunization of mice with bloodstream form trypanosomes (from irradiated hosts) was able to elicit a protective response while immunization with culture forms did not. From their results, one might conclude that the relevant antigen recognized by the immune system was adult stage specific. Alternatively, the concentration of the immunogenic molecule in the culture forms may not have been at a sufficient level necessary to generate protective immunity in their experiments. Samarawickrema & Howell (1990) have reported the stage specificity of the two major polypeptides precipitated by immune serum. When parasites were obtained from the growth and early plateau stages of infection, an 88-92 kDa protein was recognized. By mid-plateau phase onwards, a 66-70 kDa protein was present and the first protein could not be detected. Referring to our studies, the presence of ~66 protein could be detected on the Western blots of lysed culture form trypanosome preparations. The recognition of p66 on these immunoblots is most likely due to the higher concentration of yourig and dividing parasites in our samples as compared to those in the paper. Also, the IgG2a-enriched fraction used by us could have been more concentrated than that used by Samarawickrema & Howell (1990) and, therefore, was more sensitive at detecting the p66 protein at low quantities. In later experiments a panel of T. musculi monoclonal antibodies was used to probe the Western blots. The antibodies were selected for their reactivity against trypanosome antigens and for their different isotopes: IgG2a, IgGl and IgG3. Unlike IP, passive transfer of these antibodies into infected mice did not alter the course of infection (unpublished observations). Thus the mere recognition of the p66 by monoclonal antibodies is not sufficient in bringing about parasite elimination and this is observed in the schistosome model as well (Kelly, Simpson, Fox, Phillips & Smithers, 1986). Interestingly, the major protein identified in all blots was the ~66. A faint doublet below the band was also observed which may correspond to a breakdown product. Since all the monoclonal antibodies recognized the p66 antigen with the same intensity, this strongly suggests that it
identified by antibody
611
is the immunodominant protein on the trypanosome. Since the p66 was a major protein recognized by the curative immunoglobulin, it was important to confirm the parasite origin of the protein. The radiolabelled p66 antigen was immunoprecipitated by the anti-T. muxdi monoclonal antibodies and shown to be present on the autoradiograph. This confirms that the molecule is parasite, and not host, derived. A radiolabelled protein of 50 kDa mol. wt was observed by us in the autoradiograph. Similarly, in the studies of Samarawickrema & Howell (1990), a 50 kDa band was also detected in their autoradiographs, as well as in Western blots probed with normal and immune sernm followed by anti-mouse IgG as a second antibody. They have postulated that this protein is the host Ig heavy chain which binds non-specifically to the parasite surface. From the results of our experiments, it is unlikely that the 50 kDa band detected in the autoradiograph is not an immunoglobulin heavy chain since the trypanosomes in these studies were biosynthetically labelled with “S-methionine (and not surface iodinated as in their experiments). The labelled peptide which is identified on the autoradiograph must therefore be parasite derived. The 50 kDa band could be a breakdown product of the ~66 but it is also likely that the 50 kDa band contains a native epitope recognized by the curative antibody. To confirm this, a dot blot of parasite lysates containing no or 10% SDS was probed with anti-parasite antibodies. A positive reaction was observed only in samples which were SDS-free, suggesting that native epitopes were also recognized by the antibodies. In earlier experiments, immunoblots of Triton X114-extracted T. musculi probed with IP also revealed the p66 protein. The molecule was identified with the membrane-associated fraction of the solution. In order to confirm the surface location of the protein, live trypanosomes were incubated with antiT. musculi monoclonal antibody (anti-p66) and analysed by flow cytometry. The ~66 protein could be detected on the surface of the trypanosomes but not in large quantities. When the concentration of antibody was increased, there was no correlating increase in p66 detection. This suggests that the concentration of the protein on the trypanosome surface is limited. In summary, the trypanosome antigen recognized by the curative immunoglobulin G2a antibody in T. musculi infection appears to be a monomeric 66 kDa protein. As observed in both SDS-PAGE analysis and immunoprecipitation experiments, the protein is not a major component of the parasite and comprises only a small proportion of total protein content. Despite this, different isotypes of anti-T. musculi monoclonal antibodies bind specifically to the molecule confirming that it contains immunodominant antigenic epitope(s). Further characterization of p66 is currently in progress.
K. T. Y. SHAW et al.
612 Acknowledgement-This work was supported Research Council, Grant 5448.
by the Medical
REFERENCES CASHMANN. R., LOERT~CHERR., NALBANTOCLUJ., SHAW I., KASCSAK R. J., BOLTON D. C. & BENDHE~MP. E. 1990. Cellular isoform of the scrapie agent protein participates in lymphocyte activation. CeN61: 185-192. DYWER D. M. & D’ALESANDRO P. A. 1976a. The cell surface of Trypanosoma musculi bloodstream forms: I. Fine structure and cytochemistry. Journal of Protozoology 23: 75-83 DYWER D.M. & D’ALESANDROP.A. 1976b. The cell surface of Trypanosoma musculi bloodstream forms: II Lectin and immunologic studies. Journalof Protozoology 23: 262-27 1. KELLY C., SIMPSONA. J. F., Fox E., PHILLIPSS. M. & SMITHERS S. R. 1986. The identification of Schistosoma mansoni surface antigens recognized by protective monoclonal antibodies. Parasite Immunology 8: 193-198. KHAZINDAR S. H. & DUSANIC D. G. 1982. Serological and vaccination studies with bloodstream and culture forms of Trypanosoma musculi. International Journal for Parasitology 12: 257-264. LAEMMLIU. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227: 680-685. LANHAMS. M. &GODFREY D. G. 1970. Isolation of Salivarian trypanosomes from man and other mammals using DEAE-cellulose. Experimental Parasitology 28: 263-274. MEDINA-ACOSTAE., KARESS R. E., SCHWARTZH. & RUSSELL D. G. 1989. The promastigote surface protease (gp63) of Leishmania is expressed but differentially processed and localized in the amastigote stage. Molecular and Biochemical Parasitology 37: 263-274. 01 V. T. & HERZENBERC L. A. 1980. Immunoglobulinproducing hybrid cell lines. In: SelectedMethods in Cellular Immunology (Edited by MISHELL B. B. & SHIGII S. M.), pp. 357-372. W. H. Freeman, San Francisco. PARISH N. M., MORRISON I. & PEARSON T. W. 1985. Identification of an antigen specific to Trypanosoma congolense by using monoclonal antibodies. Journal of Immunology 134:593-597. PEARSONT. W.&YOUNG J. R. 1980. Analytical techniques for cell fraction. XXVIII. Dissection of complete antigenic
mixtures using monoclonal antibodies and two dimensional gel electrophoresis. Analytical Biochemistry 101: 377-386. SAMARAWICKREMA N. A. & HOWELL M. J. 1990. The surface coat of bloodstream forms of Trypanosoma musculi from mice. International Journalfor Parasitology 20: 1055-1062. SCHARFSTEINJ., SCHECH~ERM., SENNA M., PERALTA J. M., MENDONCA-PREVIATO L. & MILES M. A. 1986. Trypanosoma cruzi: characterization and isolation of a 57/ 51 000 M.W. surface glycoprotein (gp57/51) expressed by epimastigotes and bloodstream trypomastigotes. Journal ofImmunology 137: 1336-1341. STAERZ U. D., RAMMENSEEH., BENEDETTOJ. D. & BEVAN M. J. 1985. Characterization of a murine monoclonal antibody specific for an allotypic determinant of T cell antigen receptor. Journal of Immunology 134:39944000. TARGETT G. A. T. & VIENS P. 1975. The immunological response of CBA mice to Trypanosoma musculi: elimination of the parasite from the blood. International Journal for Parasitology 5: 231-234. ULCZAK 0. M., GHADIRIANE., SKAMENEE., BLACKWELLJ. M. & KONGSHAVNP. A. L. 1989. Characterization of protective T cells in the acquired response to Leishmania donovani in genetically determined cure (H-2b) and noncure (H-2”) mouse strains. Infection and Immunity 57: 2892-2899. VARGAS F., DEL C., VIENS P. & KONGSHAVNP. A. L. 1984. Trypanosoma musculi infection in B-cell-deficient mice. Infection and Immunity 44: 162-167. VIENS P., TARGETTG. A. T., LEUCHARSE. & DAVIESA. J. S. 1974. The immunological response of CBA mice to Trypanosoma musculi. I. Initial control of the infection and the effect of T cell deprivation. Clinical and Experimental Immunology 16~279-294. WECHSLER D. S. & KONGSHAVN P. A. L. 1984. Cure of Trypanosoma musculi infection by heat-labile activity in immune plasma. Infection and Immunity 44756-759. WECHSLER D. S. & KONGSHAVN P. A. L. 1985. Characterization of antibodies mediating protection and cure of Trypanosoma musculi infection in mice. Infection and Immunity 481787-794. WECHSLER D. S. & KONGSHAVNP. A. L. 1986. Heat-labile IgG2a antibodies affect cure of Trypanosoma musculi infection in C57B/6 mice. Journal of Immunology 137:2968-2972.
INTRODUCTION Trypanosoma musculi, a natural protozoan parasite of
mice, causes a self-limiting infection of approximately 3 weeks duration (Targett & Viens, 1975; Viens, Targett, Leuchars & Davies, 1974). During the course of parasitaemia, three phases may be distinguished: an exponential growth phase, a plateau phase and an elimination phase. During this Iattermost phase, specific immunity is activated and mediated through an as yet undefined mechanism. Previous studies from our laboratory have shown that this process is B cell dependent (Vargas, Del, Viens & Kongshavn, 1984). The necessity for antibodies in the curative m~hanism was demonstrated in experiments whereby the passive transfer of immune plasma obtained from a cured mouse was found to be able to mediate trypanosome
~To whom all correspondence should be addressed at: Montreal General Hospital Research Institute, 1650 Cedar Avenue, Montreal, Quebec, Canada H3G lA4.
clearance in vivo and to mediate trypanosome killing in vitro (Wechsler & Kongshavn, 1984). In order to identify the curative element, immune plasma was initially purified using protein A Sepharose and the active fraction found to be enriched in immunoglobulins of the subclasses G2a and G3 (Wechsler & Kongshavn, 1985). Upon further purification using the curative fraction was affinity columns, characterized as being an immunoglobulin of the IgG2a subclass (Wechsler & Kongshavn, 1985). The identification of parasite antigens by the host, especially those recognized by a curative antibody, is an initial step towards understanding host-parasite interactions. In earlier studies, Khazindar & Dusanic (1982) reported the serological and antigenic differences between bloodstream and culture forms of T. musculi, Dywer & D’Alesandro (1976a,b) have shown that the surface of T. musculi, like other trypanosome species, appears to be covered by glycoprotein. More recently, Samarawickrema & Howell (1990) have examined the surface coat of bloodstream 403
K. T. Y. SHAW et al.
604
forms of the parasite and have identified two stagespecific polypeptides. Apart from these studies, little else is known about these trypanosomes. The objective of the present study, therefore, was to examine the antigens expressed on T. musculi, both bloodstream and culture forms, and to identify those antigen(s) recognized by the curative IgG2a antibody. MATERIALS
AND METHODS
Mice and parasites. For the present studies, a T. musculi clone was stabilized according to standard procedure, in
order to establish a line of parasites with identical antigenic minimizing components, thereby the presence of antigenically different variants in the source of trypanosomes being used. A single trypanosome was expanded in irradiated &lo-week-old male BlO.D2/oSn mice (Jackson Laboratories Inc., Bar Harbor, ME) and, at peak parasitaemia (lo6 parasites per ml blood), the mice were sacrificed and blood from each mouse divided into 1 ml aliquots. One clone was arbitrarily selected and used for all experiments. Female C3H/HeN retired breeders (Charles River, St. Constant, Quebec, Canada) were infected intraperitoneally with 5 x IO4 trypanosomes and used for the preparation of parasite antigen. The continuance of the stabilate was maintained through biweekly passage of 5 x lo4 viable T. musculi injected intravenously into &IO-week-old male A/J mice (Jackson Laboratories Inc., Bar Harbor, ME) for no more than 3 months. The parasites were enumerated as previously described (Wechsler & Kongshavn, 1985). T. musculi antigen. The parasites used for the preparation of antigen were obtained from three sources: bloodstream forms (BSF) from an untreated host, bloodstream forms from an irradiated host (BSF-X) and culture forms (CF). The blood from normal infected animals was collected from the retroorbital sinus of infected C3H/HeN mice 9-12 days postinoculum (p.i.), and the trypanosomes were partially purified from red blood cells by differential centrifugation (Wechsler & Kongshavn, 1985). They were further purified by passage through a DEAE-cellulose (Whatman Chromedia DE52, Whatman BioSystems Ltd, Maidstone, Kent, U.K.) column to separate out contaminating white blood cells (Lanham & Godfrey, 1970). The trypanosomes from irradiated hosts were purified similarly from blood taken 8-12 days p.i. Culture forms of T. musculi were prepared according to methods previously described (Wechsler & Kongshavn, 1986). The three preparations were each resuspended in phosphate-buffered saline glucose, pH 8.0 at a concentration of 10’ trypanosomes per ml and to each one was added an equal volume of lysis buffer [l% SDS, I mhl-Tris, 150 mMNaCI, 2.5 mr+EDTA, 0.05% sodium azide, 1 mMphenylmethylsulphonyl fluoride (PMSF)]. The suspension was stored at - 20°C and pelleted (25Og for 15 min) before use. Plasmapreparation.Immune plasma (IP) obtained from a cured mouse was prepared using methods previously described (Wechsler & Kongshavn, 1985). Irradiation. Mice used for the preparation of a T. musculi clone were irradiated I day prior to use with 500 rad using methods described elsewhere (Ulczak, Ghadirian, Skamene, Blackwell & Kongshavn, 1989).
Preparation of immune plasma fractions. IP was separated into IgG2a-enriched and IgGZa-depleted fractions using methods previously described (Wechsler & Kongshavn, 1986). Monoclonal antibodies. Anti-T. musculi monoclonal antibodies were raised using the method of Oi & Herzenberg (1980). The monoclonal antibodies were linked either to Pierce AminoLink beads (Pierce, Rockford, IL) or to Sepharose 4B beads (Pharmacia Canada Inc., Baie D’Urte, Quebec, Canada) according to manufacturer’s instructions. F23.1, an anti-T cell receptor monoclonal antibody (Staerz, Rammensee, Benedetto & Bevan, 1985) linked to Sepharose was provided by Dr T. Owens, Montreal Neurological Institute and Hospital, Montreal, Quebec, Canada. JR/ MAb, an anti-DNA IgGl monoclonal antibody was provided by Dr J. Rauch, Montreal General Hospital Research Institute, Montreal, Quebec, Canada. SDS-PAGE. T. musculi lysates were run on SDS-PAGE gels (10.0%) under reducing conditions using the discontinuous buffer system of Laemmli (1970). The samples were diluted with equal volumes of 2 x sample buffer (2-Bmercaptoethanol, 0.0625 hr-Tris-HCl, pH 6.8, 10% SDS, 10% glycerin) and boiled at 100°C for 2-3 min before loading. One lane containing 1 pg of standard mol. wt protein markers (Bio-Rad, Richmond, CA) was diluted in 10 ~1 of sample buffer. Electrophoresis was performed at 125 V for 75 min in a Bio-Rad Mini Protean II Slab cell. The gels were either immunoblotted as described below, or stained with 5% Coomassie blue in 10% acetic acid, 40% methanol, and then destained with 10% acetic acid, 50% methanol. Immunoblotting. Following electrophoresis, proteins of the bloodstream form preparation were transferred onto nitrocellulose by electroelution at 150 mA using the Bio-Rad Trans-Blot apparatus (Bio-Rad, Richmond, CA). Two lanes, one containing standards and the other containing parasite proteins, were cut from the blot and the efficiency of transfer was determined by amido black stain. The remaining nitrocellulose blot was blocked with 5% casein, 0.25 M-Tris, pH 7.2, 0.15 M-NaCI for 1 h at 37’C. The lanes within each blot were cut, washed three times with wash buffer (0.25 MTris, 0.15 M-NaCl, pH 7.2) and incubated with whole or fractionated IP (1:lOO dilution) or monoclonal antibodies (I:10 dilution) overnight at 37°C. After washing three times with wash buffer, the strips were immersed in 5% casein, PBS to which was added goat anti-mouse Ig (I:1000 dilution, Southern Biotechnology Associates Inc., Birmingham, AL) and incubated for 3 h at 37°C. The replicas were rinsed five times with the wash buffer and bound antibody was detected using AP colour development, Fast Red (Bio-Rad, Richmond, CA). Immunoprecipitation and autoradiography. Internal labelhng of trypanosomes and subsequent immunoprecipitation were performed using the method described by Parish, Morrison & Pearson (1985) with minor modifications. Briefly, 0.5 ml of a T. musculi preparation (partially purified as described earlier, 10’ per ml) was added to each well of a six-well plate (Limbro, Flow Laboratories Inc., Mclean, VA) containing 4.5 ml of RPM1 1640 containing only 5% of normal methionine concentration (Gibco Laboratories Life Technologies Inc., Grand Island, NY). An aliquot of 200 &i
T. musculi
antigens identified by antibody
of’%-methionine (specificactivity 0.6-4 Ci rnM_‘,Amersham Corp., Oakville, Ontario, Canada) was added to each well and the plate was covered and incubated for 16 h at 37°C. After incubation, the radiolabelled trypanosomes were harvested, pooled, washed with RPM1 1640, centrifuged at 1OOOgfor 10 min and resuspended in 200 ~1 of the same medium. In order to lyse the parasites, 300 ~1 of lysis buffer [5 mM-Tris-HCl, pH 9.2, 1 mM-EDTA, 1% Nonidet P-40 (NP40) 1 mM-PMSF and 0.1 mM-tosyl-Llysine chloromethylketone (TLCK)] was added and the suspension incubated at 37°C for 15 min. The insoluble material was removed by microcentrifugation and the supernatant dialysed against 2 1of Tris-EDTA-NaC1 (TEN) buffer, pH 8.5 (20 mM-Tris-HCl, pH 7.4, 1 mi+EDTA, 100 mM-NaCl and 0.1% NP-40) at 4°C for 24 h. The dialysate was until use. microcentrifuged and stored at -70°C Immunoprecipitations were performed using 200 pl aliquots of radiolabelled lysates. A 100 pl sample of 50% (v/v) AminoLink or Sepharose linked anti-Z’. musculi monoclonal antibodies was microcentrifuged for 5-10 min and the overlying buffer removed. Sepharose linked anti-T cell receptor monocolonal antibodies were used as control and prepared similarly. Aliquots of the trypanosome lysate were added to each set of beads and lightly vortexed before incubating on ice for 2.5 h. After incubation, each mixture was washed three times with TEN buffer, resuspended in 200 ~1 of the same buffer and placed on microcolumns (Pearson & Young, 1980). The column was dried by centrifugation at 750g for 5 min after which 4t3-60 ~12 x SDS sample buffer was added and the column was incubated at room temperature (RT) for 30 min. Finally the column was recentrifuged for 10 min and the eluted fraction collected and analysed by SDS-PAGE. The gel was stained, dried and autoradiographed on Kodak X-OMAT S film. Dor blot. A sample of bloodstream form trypanosomes lysate was divided into two 100 ~1 aliquots. Each sample was microcentrifug~ for 5 min and its overlying supernatant removed. They were then resuspended in 100 ~1 of lysis buffer f 10% SDS. A portion of each sample was aliquoted into additional samples of 1:5 and 1:50 dilution. Nitrocellulose strips (0.2 cm in width x 7 cm in length) were cut and 2 ~1 of each sample (l:l, 1:5, 1:50 f SDS) was applied, using an eppendorf pipette, onto each strip. The nitrocellulose strip was allowed to dry and the samples were applied twice more using the same procedure. After blocking for 1 h with 10% goat serum (Gibco Laboratories Life Technologies Inc., Grand Island, NY) in PBS, pH 7.4, each strip was washed three times with wash buffer (described above) and incubated with whole or fractionated IP or NMP or anti-T. mus~ff IgGfa monoclonal antibody or anti-DNA IgGl monoclonal antibody (at the dilutions described above in 1% goat serum/ PBS) overnight at 37°C. The strips were subsequently washed three times and incubated with goat anti-mouse Ig (1:tOOO dilution in 1% goat serum/PBS) for 3 h at 37°C. The blots were rinsed five times with wash buffer and developed using AP colour development, Fast Red. ~rnrn~~o~~~rescence and cyto~~oromet~y. Bloodstream form trypanosomes were probed indirectly with fluorescent antibodies and analysed by flow cytometry using the method described by Cashman, Loertscher, Nalbantoglu, Shaw,
605
Kascsak, Bolton & Bendheim (1990). Briefly, fresh purified parasites were washed with PBS (PH 7.0) and divided into five aliquots of lo6 trypanosomes. All samples were washed three times with PBS between each subsequent incubation. The samples were incubated with 10% normal goat serum for 30 min to reduce non-specific immuno~obulin binding. The aliquots were then incubated at 4°C for 30 min with either anti-T. musculi IgG2a monoclonal antibody, IP or NMP or anti-CD4 monoclonai antibody (Leu 3a, Becton Dickinson, Mountain View, CA) at I: 1 dilutions. After incubation for 30 min with the appropriate second antibody, FITC-labelled anti-mouse immunoglobulin (Tago Immunologicals, Inter Medico, Markham, Ontario, Canada), the tryp~osomes were analysed by flow cytometry with an argon laser FACScan instrument equipped with Consort 30 and LYSYS II software (Becton Dickinson, Mountain View, CA). As a control, samples of freshly obtained human peripheral lymphocytes were run concurrently with the parasite samples under similar conditions. RESULTS SDS-PAGE analysis of lysedparasite preparations In order to characterize the antigens of T. musculi, parasite lysates were subjected to SDS-PAGE. This
method allowed the visual determination of differences between the three types of parasite preparations. After electrophoresis, analysis of each Iysate revealed numerous bands covering a wide range of molecular weights in all samples (Fig. 1). Bloodstream form lysates from normal mice (BSF, lane 1) have a complex pattern enriched in molecules of mol. wt 34-68 kDa. Similarly, bloodstream forms from an irradiated host (BSF-X, lane 2) have an almost identical pattern. The majority of bands in both lanes migrate at the same mobility. In comparison, culture forms (lane 3) of T. muscufi have a different banding pattern ofmol. wt 36 68 kDa. Many of these antigens are shared with the other two preparations; however, unique bands are also visible. The culture forms appear to be quantitatively restricted in proteins of very high or very low mol. wts. Western blot analysis using immuneplasma
Following protein separation by SDS-PAGE, T. musculi antigens were transferred onto nitrocellulose. Western blots were probed with either whole or fractionated IP in order to identify the immunogenic parasite proteins (Fig. 2). On immunoblots of trypanosome lysates (T) probed with IP, a series of bands was revealed. In contrast, the curative IgG2aenriched fraction (ELU) of IP identified with high intensity a single distinct band migrating at 66 kDa. Several higher mol. wt bands also showed faint staining. A similar pattern was observed using IgG2adepleted fractions (FT) of IP. When lysates were probed with normal mouse plasma (NMP), no bands
K.
T. Y.
SHAW et al.
FIG. 1. SDS-PAGE analysis of whole T. musculi lysates. Preparations of bloodstream form trypanosomes from normal mice (BSF), bloodstream form trypanosomes from irradiated mice (BSF-X) and culture form trypanosomes (CF), were lysed and separated on a 10% gel under reducing conditions. Mol. wt markers are indicated on the left of the gel.
IP
FT
ELlJ
IP
NMP
.
..
97.466.2i’
.
,w *
42.7-
,-
*;a
.
I
f
T
T
1
T
M
A
Frc. 2. Western blot analysis of T. musculi lysates probed with immune plasma and its fractions. After SDS-PAGE separation, trypanosome lysates (T) were transferred onto nitrocellulose and probed with whole immune plasma (IP), IgGZa-depleted IP fraction (FI) and IgG2a-enriched IP fraction (EIJJ). A control lane was probed with normal mouse plasma (NMP). Immunoblots of mouse albumin (A) and malaria antigens (M) were also probed with IP. Mol. wt standards are indicated on the left and were superimposed from a portion of the blot that was stained with amido black.
T. musculi antigens identified by antibody
El
60’7
DS
97.4-
66.2-
. T
1
FIG. 3. Detection of T. musculi antigens using mono-
42.7-
clonal antibodies. Trypanosome lysates (T) were separated by SDS-PAGE, transferred onto nitrocellulose and probed with anti-T. musculi monoclonal antibodies El, H2, CS and D5. Mol. wt standards are indicated on the left and were
superimposed from a portion of the blot that was stained with amido black. 31 .o-
were detected. As further controls, mouse albumin {A) and a lysate of malaria antigens (M) were transferred onto nitrocellulose and probed with immune plasma. These blots exhibited negative binding, confirming that the specificity of immune plasma and its fractions was only towards T. musculi antigens. Western blot analysis using monoclona~ antibodies
In this series of experiments, high affinity anti-T. antibodies of three different IgG sub-classes (El and H2 are both of the IgG2a subclass while C6 and DS are of the IgG3 and IgGl subclass, respectively) were used to probe the Western blots of trypanosome lysates (Fig. 3). Interestingly, the pattern of binding is identical in all four preparations. The major band of approximately 66.2 kDa was distinctIy recognized. A small doublet of lower mol. wt was also faintly detected. No reaction was observed using a control monoclonal antibody (an IgG2a antibody to T cell receptor, data not shown) nor when preparations of mouse albumin and lysed malaria antigen were substituted for the trypanosomes. The reactivity of these monoclonal antibodies is, therefore, restricted to a single T. muscu~~antigen.
musculi monoclonal
of radiolabe~led T. musculi In order to confirm that the 66 kDa band was of parasite, not host, origin, a preparation of “Smethionine-labelled trypanosomes was lysed and ~mmunopreeipitation
FIG. 4. Autoradiograph of [‘SS]-methionine-labelled proteins of whole T. muxuli and of immunoprecipitates obtained by using AminoLink and/or Sepharose linked anti-l: musculi monoclonal antibodies. Lunea, whole T. musculilysate. Lane 6, anti-T. musculi monoclonal antibody imm~opre~ipitate. Lane c, anti-T cell receptor monoclonal antibody immunopr~pitate. Mol. wt standards are indicated on the left and were superimposed from a portion of the gel that was stained with Coomassie Brilliant blue.
immunoprecipitated with anti-T. musculi monoclonal antibody linked to AminoLink or Sepharose beads. Following SDS-PAGE and autoradiography, a series of bands was revealed when the complete lysate was analysed (Fig. 4, lane a). However, the 66 kDa band was detected, as well as several other bands including a 50 kDa protein, in the immunoprecipitate obtained from the anti-T. muscufi monoclonal antibodies. This
K. T. Y.
608
Determination
of Native
in Lysed T.musculi
SHAW
et al.
and Non-native
Antigen
Preparation
Epitopes (~SDS)
1:1(-I (-1
noSDS
(+)
SDS
1:l (+I
1:5(-J
1:5(t)
1:50(-I
1:5o(+)
FIG. 5. Dot blot analysis of T. musculi antigens. Lysates of bloodstream form trypanosomes were resuspended in 100 ~1 aliquots f 10% SDS. The samples were diluted (1: 1, 1:5, 1:50) and applied to nitrocellulose strips. The strips were probed with whole immune plasma (IP), IgG2a-enriched IP fraction (ELU), IgGZa-depleted fraction (FT) and anti-T. musculi IgG2a monoclonal antibody (MAb). Control lanes were probed with normal mouse plasma (NMP), anti-DNA IgGl monoclonal antibody (JR/MAb), second antibody alone (2”) or buffer (buffer).
lower mol. wt protein could possibly be a breakdown product of the 66 kDa molecule. Sepharose linked anti-T cell receptor monoclonal antibodies were also incubated with the trypanosome lysate as a negative control and analysis of the immunoprecipitate did not detect any proteins (lane c). Dot blot analysis of T. musculi antigens In order to determine the presence of native or SDS epitopes, lysates of T. musculi containing no or 10% SDS were analysed by dot blot (Fig. 5). When the nitrocellulose strip was probed with whole IP, a positive dot was revealed in the undiluted SDS-free sample. This identical dot was also detected on the strip probed with IgG2a-enriched IP fraction (ELU) and it was faintly recognized on the strip probed with the FT fraction. When the blot was probed with antiT. musculi monoclonal antibody (MAb), the nonSDS, undiluted trypanosome sample was also
revealed. In parasite samples which were either diluted or contained SDS, no dot was observed. As controls, the samples were also probed with NMP or anti-DNA monoclonal antibody (JR/MAb) but they did not show any binding. Flow cytometry analysis of T. musculi In order to confirm the surface location of p66 protein, the trypanosomes were indirectly labelled with fluorescent antibody and analysed by flow cytometry (Fig. 6). The population of trypanosomes was selected for viability (indicated by propidium iodide uptake, data not shown). The positive detection of surface protein was characterized by a change in histogram profile from control data (T. muscufi + FITC alone, Fig. 6A). An increase in protein concentration was indicated by a shift in histogram profile with a corresponding increase in mean fluorescence. When the parasites were incubated with anti-
609
T. mu.m& antigens identified by antibody
B
1
E
7 I
F
H
FIG. 6.
K. T. Y. SHAW et al.
610
T. ~us~uli monoclonal antibody (which specifically recognizes the ~66 protein), a unique histogram profile was observed (Fig. 6C). The low mean fluorescence indicated that the parasite population in general expressed ~66 in low abundance. As a positive control, parasites incubated with IP also showed a similar profile but was greatly shifted towards the right indicating greater recognition of a larger quantity of proteins (~66 and others) (Fig. 6E). In control samples, namely trypanosomes incubated with NMP (Fig. 6G) or anti-CD4 mon~lonal antibody (data not shown), there was no significant change in histogram profile or mean fluorescence. As a further control, human lymphocytes were incubated with anti-T. musculi monoclonal antibody (Fig. 6D); there were no detectable changes in histogram profile from that observed in the sample containing cells and FITC alone (Fig. 6B). This confirms that the identification of ~66 on the trypanosomes, as indicated by a change in histogram profile, was not an artifact caused by the addition of antibody. A shift in histogram profile was observed in samples of cells incubated with IP {Fig. 6F); this was not significant as an identical shift was also seen in cells incubated with NMP (Fig. 6H). DISCUSSION
In order to examine the antigens of T. musculi, we analysed three preparations of parasites obtained from different sources: normal mice, irradiated mice and in vitro cultures. Each preparation contained different proportions of T. rn~&~li at each of the various stages of growth. Bloodstream forms obtained 9-12 days p.i. at the height of the plateau phase of infection, were exclusively monomorphic adult trypomastigotes. In vitro cultures, in contrast, have a 100% multiplicative parasite population. A mixed popufation of 35% young and dividing forms and 65% adult forms was observed in the blood of irradiated animals. The molecular nature of the proteins in each lysate was determined by SDS-PAGE analysis. The antigenic components of both bloodstream forms appeared simiiar but varied in the intensity of staining of some bands. This was related to the differences in protein concentrations between the lysates. Analysis of culture forms also revealed a banding pattern containing a variety of proteins, but, unlike either bloodstream form preparations, lacked the very high and very low mol. wt proteins. Overall,
the antigens from multiplicative forms and adult forms appear to have some shared antigens as well as unique proteins. This is not unusual and has been observed in other trypanosomes and parasite organisms (Medina-Acosta, Karess, Schwartz & Russell, 1989; Parish et al., 1985; Scharfstein, Schechter, Senna, Peralta, Mendonqa-Previato & Miles, 1986). Having established that the lysates contain a wide range of proteins, the next step was to identify those recognized by the curative IgG2a antibody. During the elimination phase of parasitaemia, the effective acquired immune response is directed against the adult parasite forms, therefore, in all Western blots, bloodstream forms from normal mice (100% adults) were examined with immune plasma and its fractions. In the immunoblots probed with IP, a host of antigens was recognized and the banding pattern was observed to be similar to the SDS-PAGE of whole lysates. This was not surprising since we would expect the host to be able to elicit an antibody response against a wide variety of parasite antigens. When IP was separated into fractions, blots probed with the curative IgGZaenriched fraction revealed a major protein of 66 kDa (~66). Several higher mol. wt bands were also visible. When probed with the IgGZa-depleted fractions, a similar pattern was observed suggesting that other specific antibody isotypes present in the plasma also recognized the ~66. However, we have been able to show that these antibodies depleted of the IgG2aenriched fraction are not effective in elimination of parasitaemia in vivo (Wechsler & Kongshavn, 1984). Samarawickrema & Howell (1990) have also reported the identification of a 66-70 kDa molecule recognized by radioimmunoprecipitation and affinity-eluted immune serum. A second polypeptide of higher mol. wt, 88-92 kDa, was also revealed, thus substantiating our results. Since high mol. wt proteins are visible on the SDSPAGE and Western blots, the ~66 could be a subunit of a large aggregate structure. In order to determine the unit structure of the protein, the trypanosome lysate was passed through a high pressure liquid chromatography gel filtration column (data not shown). A bovine serum albumin standard (66 kDa single unit protein) was passed through the column initially as a control and its retention time in the column determined. The active trypanosome fraction obtained from this purification (as detected by
FIG. 6. Flow cytometry analysis of T. muset&.Fluorescence histograms of trypanosomes gated from a freshly isolated sample (A, C, E, G) and lymphocytes from human peripheral blood (B, D, F, H). The samples were incubated with FITC alone (A, B), anti-T. musculimonoclonal antibody + FITC (C, D), IP + FITC (E, F) or NMP + FITC (G, H). The ordinate axis represents the number of cellular events and the abscissa represents the arbitrary mean fluorescence units as determined by FACScan software.
7’. ~~~u~i antigens
enzyme-linked immunosorbent assay) eluted with an identical retention time as the standard. This implies that ~66 has a monomeric structure. In other experiments (data not shown) IP was used to probe Western blots of lysates obtained from cultured trypanosomes and from irradiated hosts. As with results described above, the ~66 as well as several other antigens were identified in both preparations. In finding p66 in the culture form preparation, one can assume that it is not an adult stage specific molecule. These results do not support the studies by Khazindar & Dusanic (1982) who showed that immunization of mice with bloodstream form trypanosomes (from irradiated hosts) was able to elicit a protective response while immunization with culture forms did not. From their results, one might conclude that the relevant antigen recognized by the immune system was adult stage specific. Alternatively, the concentration of the immunogenic molecule in the culture forms may not have been at a sufficient level necessary to generate protective immunity in their experiments. Samarawickrema & Howell (1990) have reported the stage specificity of the two major polypeptides precipitated by immune serum. When parasites were obtained from the growth and early plateau stages of infection, an 88-92 kDa protein was recognized. By mid-plateau phase onwards, a 66-70 kDa protein was present and the first protein could not be detected. Referring to our studies, the presence of ~66 protein could be detected on the Western blots of lysed culture form trypanosome preparations. The recognition of p66 on these immunoblots is most likely due to the higher concentration of yourig and dividing parasites in our samples as compared to those in the paper. Also, the IgG2a-enriched fraction used by us could have been more concentrated than that used by Samarawickrema & Howell (1990) and, therefore, was more sensitive at detecting the p66 protein at low quantities. In later experiments a panel of T. musculi monoclonal antibodies was used to probe the Western blots. The antibodies were selected for their reactivity against trypanosome antigens and for their different isotopes: IgG2a, IgGl and IgG3. Unlike IP, passive transfer of these antibodies into infected mice did not alter the course of infection (unpublished observations). Thus the mere recognition of the p66 by monoclonal antibodies is not sufficient in bringing about parasite elimination and this is observed in the schistosome model as well (Kelly, Simpson, Fox, Phillips & Smithers, 1986). Interestingly, the major protein identified in all blots was the ~66. A faint doublet below the band was also observed which may correspond to a breakdown product. Since all the monoclonal antibodies recognized the p66 antigen with the same intensity, this strongly suggests that it
identified by antibody
611
is the immunodominant protein on the trypanosome. Since the p66 was a major protein recognized by the curative immunoglobulin, it was important to confirm the parasite origin of the protein. The radiolabelled p66 antigen was immunoprecipitated by the anti-T. muxdi monoclonal antibodies and shown to be present on the autoradiograph. This confirms that the molecule is parasite, and not host, derived. A radiolabelled protein of 50 kDa mol. wt was observed by us in the autoradiograph. Similarly, in the studies of Samarawickrema & Howell (1990), a 50 kDa band was also detected in their autoradiographs, as well as in Western blots probed with normal and immune sernm followed by anti-mouse IgG as a second antibody. They have postulated that this protein is the host Ig heavy chain which binds non-specifically to the parasite surface. From the results of our experiments, it is unlikely that the 50 kDa band detected in the autoradiograph is not an immunoglobulin heavy chain since the trypanosomes in these studies were biosynthetically labelled with “S-methionine (and not surface iodinated as in their experiments). The labelled peptide which is identified on the autoradiograph must therefore be parasite derived. The 50 kDa band could be a breakdown product of the ~66 but it is also likely that the 50 kDa band contains a native epitope recognized by the curative antibody. To confirm this, a dot blot of parasite lysates containing no or 10% SDS was probed with anti-parasite antibodies. A positive reaction was observed only in samples which were SDS-free, suggesting that native epitopes were also recognized by the antibodies. In earlier experiments, immunoblots of Triton X114-extracted T. musculi probed with IP also revealed the p66 protein. The molecule was identified with the membrane-associated fraction of the solution. In order to confirm the surface location of the protein, live trypanosomes were incubated with antiT. musculi monoclonal antibody (anti-p66) and analysed by flow cytometry. The ~66 protein could be detected on the surface of the trypanosomes but not in large quantities. When the concentration of antibody was increased, there was no correlating increase in p66 detection. This suggests that the concentration of the protein on the trypanosome surface is limited. In summary, the trypanosome antigen recognized by the curative immunoglobulin G2a antibody in T. musculi infection appears to be a monomeric 66 kDa protein. As observed in both SDS-PAGE analysis and immunoprecipitation experiments, the protein is not a major component of the parasite and comprises only a small proportion of total protein content. Despite this, different isotypes of anti-T. musculi monoclonal antibodies bind specifically to the molecule confirming that it contains immunodominant antigenic epitope(s). Further characterization of p66 is currently in progress.
K. T. Y. SHAW et al.
612 Acknowledgement-This work was supported Research Council, Grant 5448.
by the Medical
REFERENCES CASHMANN. R., LOERT~CHERR., NALBANTOCLUJ., SHAW I., KASCSAK R. J., BOLTON D. C. & BENDHE~MP. E. 1990. Cellular isoform of the scrapie agent protein participates in lymphocyte activation. CeN61: 185-192. DYWER D. M. & D’ALESANDRO P. A. 1976a. The cell surface of Trypanosoma musculi bloodstream forms: I. Fine structure and cytochemistry. Journal of Protozoology 23: 75-83 DYWER D.M. & D’ALESANDROP.A. 1976b. The cell surface of Trypanosoma musculi bloodstream forms: II Lectin and immunologic studies. Journalof Protozoology 23: 262-27 1. KELLY C., SIMPSONA. J. F., Fox E., PHILLIPSS. M. & SMITHERS S. R. 1986. The identification of Schistosoma mansoni surface antigens recognized by protective monoclonal antibodies. Parasite Immunology 8: 193-198. KHAZINDAR S. H. & DUSANIC D. G. 1982. Serological and vaccination studies with bloodstream and culture forms of Trypanosoma musculi. International Journal for Parasitology 12: 257-264. LAEMMLIU. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227: 680-685. LANHAMS. M. &GODFREY D. G. 1970. Isolation of Salivarian trypanosomes from man and other mammals using DEAE-cellulose. Experimental Parasitology 28: 263-274. MEDINA-ACOSTAE., KARESS R. E., SCHWARTZH. & RUSSELL D. G. 1989. The promastigote surface protease (gp63) of Leishmania is expressed but differentially processed and localized in the amastigote stage. Molecular and Biochemical Parasitology 37: 263-274. 01 V. T. & HERZENBERC L. A. 1980. Immunoglobulinproducing hybrid cell lines. In: SelectedMethods in Cellular Immunology (Edited by MISHELL B. B. & SHIGII S. M.), pp. 357-372. W. H. Freeman, San Francisco. PARISH N. M., MORRISON I. & PEARSON T. W. 1985. Identification of an antigen specific to Trypanosoma congolense by using monoclonal antibodies. Journal of Immunology 134:593-597. PEARSONT. W.&YOUNG J. R. 1980. Analytical techniques for cell fraction. XXVIII. Dissection of complete antigenic
mixtures using monoclonal antibodies and two dimensional gel electrophoresis. Analytical Biochemistry 101: 377-386. SAMARAWICKREMA N. A. & HOWELL M. J. 1990. The surface coat of bloodstream forms of Trypanosoma musculi from mice. International Journalfor Parasitology 20: 1055-1062. SCHARFSTEINJ., SCHECH~ERM., SENNA M., PERALTA J. M., MENDONCA-PREVIATO L. & MILES M. A. 1986. Trypanosoma cruzi: characterization and isolation of a 57/ 51 000 M.W. surface glycoprotein (gp57/51) expressed by epimastigotes and bloodstream trypomastigotes. Journal ofImmunology 137: 1336-1341. STAERZ U. D., RAMMENSEEH., BENEDETTOJ. D. & BEVAN M. J. 1985. Characterization of a murine monoclonal antibody specific for an allotypic determinant of T cell antigen receptor. Journal of Immunology 134:39944000. TARGETT G. A. T. & VIENS P. 1975. The immunological response of CBA mice to Trypanosoma musculi: elimination of the parasite from the blood. International Journal for Parasitology 5: 231-234. ULCZAK 0. M., GHADIRIANE., SKAMENEE., BLACKWELLJ. M. & KONGSHAVNP. A. L. 1989. Characterization of protective T cells in the acquired response to Leishmania donovani in genetically determined cure (H-2b) and noncure (H-2”) mouse strains. Infection and Immunity 57: 2892-2899. VARGAS F., DEL C., VIENS P. & KONGSHAVNP. A. L. 1984. Trypanosoma musculi infection in B-cell-deficient mice. Infection and Immunity 44: 162-167. VIENS P., TARGETTG. A. T., LEUCHARSE. & DAVIESA. J. S. 1974. The immunological response of CBA mice to Trypanosoma musculi. I. Initial control of the infection and the effect of T cell deprivation. Clinical and Experimental Immunology 16~279-294. WECHSLER D. S. & KONGSHAVN P. A. L. 1984. Cure of Trypanosoma musculi infection by heat-labile activity in immune plasma. Infection and Immunity 44756-759. WECHSLER D. S. & KONGSHAVN P. A. L. 1985. Characterization of antibodies mediating protection and cure of Trypanosoma musculi infection in mice. Infection and Immunity 481787-794. WECHSLER D. S. & KONGSHAVNP. A. L. 1986. Heat-labile IgG2a antibodies affect cure of Trypanosoma musculi infection in C57B/6 mice. Journal of Immunology 137:2968-2972.
