Major surface proteins and antigens on the different in vivo and in vitro forms of Trypanosoma cruzi

The surface proteins of six stages of Trypanosoma cruzi were labeled by Iodogen-catalized surface iodination and analyzed by one and two dimensional polyacrylamide gel electrophoresis. These stages included bloodstream trypomastigotes, culture-form trypomastigotes, amastigotes, staphylomastigotes, epimastigotes, and metacyclic trypomastigotes. Antigens recognized by serum antibodies were detected by Western blotting against serum from mice hyperimmunized against bloodstream trypomastigotes. Bloodstream trypomastigotes, culture-form trypomastigotes and staphylomastigotes contained several surface proteins of molecular weight (Mr) 90 000 and isoelectric points (pI) between 5.0 and 7.5. Western blotting reveals that at least two proteins of 90 000 Mr represent the major antigens seen on bloodstream trypomastigotes, culture-form trypomastigotes, staphylomastigotes and amastigotes. However, a 90 000 Mr protein was not detected by either Western blotting or surface iodination on epimastigotes or metacyclic trypomastigotes. The major surface proteins on these latter two stages were represented by several 72 000 Mr proteins with pI values between 5.2 and 5.8. An interesting result of this survey is that a 90 000 Mr surface antigen was present on staphylomastigotes, a stage which can be grown in cell free medium.


INTRODUCTION
Trypanosoma cruzi is characterized by a complex life cycle expressing multiple forms of development in both the vertebrate host and the invertebrate vector [ 1].The cell surface proteins of several of these different forms have been analyzed with regard to their relative molecular weights (Mr) and recognition by serum antibodies from infected hosts [2][3][4][5][6].These studies indicate that both qualitative and quantitative differences exist among the cell surface proteins of the various stages of the life cycle of T. cruzi.Although there is general agreement that the major antigenic species on the surface of the bloodstream trypomastigote is an approx.90 000 M r glycoprotein, it is unclear whether this protein is present on all developmental stages of the parasite.
In view of this uncertainty as well as to provide necessary information for future studies on the regulation of expression ofgene sequences which encode T. cruzi surface antigens, we chose to investigate the surface components of six observed morphological forms of T. cruzi.In this article, we define by the criteria of relative molecular weight, isoelectric point and recognition by serum antibodies from T. cruzi infected mice, the proteins present on the surface of the following in vivo or in vitro forms of T. cruzi: (1) bloodstream trypomastigotes isolated from infected mice; (2) culture-form trypomastigotes grown in vitro in vertebrate tissue culture cells; (3) amastigotes isolated from in vitro cultured vertebrate tissue culture cells; (4) staphylomastigotes grown in vitro under acellular culture conditions, (5) epimastigotes grown in monophasic culture and (6) metacyclic trypomastigotes obtained by transformation of epimastigotes in an aceUular medium.

Parasites. T. cruzi, Peru strain
Growth and isolation of developmental forms.Mice, C57BL/6, were inoculated with 104 bloodstream trypomastigotes and bled about 12 days later by heart puncture using trisodium citrate saline as an anticoagulant [11].The bloodstream trypomastigotes were separated from red blood cells by a Percoll step gradient [12], rinsed twice in trisodium citrate saline and separated from host cells and platelets by passage over a 5 ml DEAE-ceilulose column equilibrated with PSG (I = 0.206.) Culture-form trypomastigotes were grown in cultures of Va-13 cells, an SV-40 virus transformed WI-38 human lung cell line obtained from the American Type Culture Collection.Va-13 cells were infected by exposure to bloodstream trypomastigotes overnight.The tissue cells were rinsed, resupplied with medium and continuously subcultured with the periodic addition of uninfected tissue cells.Yield was about 5 )< 107 culture-form trypomastigotes per 75 cm 2 flask per day after the culture was established.Culture-form trypomastigotes were pelleted by centrifugation twice in PSG (I = 0.217).Remaining host cells and cellular debris were removed by passage over a DEAE-cellulose column described above.
Amastigotes were harvested from infected Va-13 cells as follows.Infected monolayers were washed five times with cold phosphate-buffered saline pH 7.4 (PBS), the cells were removed by scraping and resuspended in 5 ml PBS.DNAse (100 lag ml-l), RNAse A (100 lag ml-t), and phenylmethylsulfonylfluoride (2 mM) were added and the cells disrupted for 60 s at full speed in a stainless steel Eberbach micro blender.Amastigotes were recovered by centrifugation onto a 20% isotonic metrizamide cushion [13].The amastigotes were removed and resuspended in 5 ml PBS containing DNAse (50 lag ml-I), RNAse (50 lag ml-1), and phenylmethylsulfonylfluoride(2 mM) and allowed to set on ice for 10 rain.The cells were then recentrifuged onto a metrizamide cushion, recovered and washed with cold PBS.
Staphylomastigotes were obtained by transformation of culture-form trypomastigotes in ML-15H [ I0]. Staphylomastigote masses from 5 day cultures were separated from culture-form trypomastigotes by centrifugation at 50 )< g for 1 min.The cell pellet was resuspended in 10 ml ML-15H and washed twice with cold PBS.Staphylomastigotes obtained by this procedure were contaminated with less than 1% cultureform trypomastigotes.Subculturing into ML-15H at 3 day intervals resulted in log phase growth with less than 1% contamination with epimastigotes or trypomastigotes.
Epimastigotes were grown in THOSMEN [ 14] at 28°C without shaking.Log phase cells were greater than 99% epimastigotes.Cells were harvested at 800 X g at 4°C.
Metacyclic trypomastigotes were produced by transformation of epimastigotes by the procedure of Sullivan [15] and were separated from remaining epimastigotes by the procedure of A1-Abbassy et al [16].Cells were greater than 98% metacyclics after these procedures.lodination.Cell surface proteins were radiolabeled with Na~25I by a modification of the Iodogen procedure of MarkweU and Fox [17].Cells were washed three times in PBS containing 1% glucose, and resuspended in 0.9 ml PBS containing 1% glucose.Iodogen (10 lag) was dissolved in dichloromethane and dried in a glass test tube containing about fifteen 4-5 mm glass beads.Na125I (1.0 mCi) was added to the 0.9 ml cell suspension and the cells were incubated with the Iodogen coated glass beads for 20 min at room temperature with gentle shaking.By phase contrast microscopy cells were normal and those with a flagellum were motile.The cells were washed 3 times in PBS containing 1% glucose and solubilized in membrane dissolving buffer on ice for 20 min with occasional gentle mixing.Nuclei and particulate matter were removed by successive centrifugation of 2 000 and 15 000 rpm for 10 min each at 4°C.The supernatant was removed and stored in aliquots which were either used immediately or frozen at -70°C with glycerol added to a final 10% concentration.
Protein concentration was determined with a Bio-Rad Protein Assay Kit.The micro assay was performed according to the manufacturer's instruction.lsoelectric focusing, first dimension.Electrophoresis was performed according to O'Farrel et al [7] with the following modifications.Four tube gels (10 cm X 0.5 cm) were prepared as follows: 1.5 g urea, 340 lal 10% Triton X-100, 400 lal of a 30% acrylamide-5.3%N,N'-diallyltartardiamide solution and 15 lal 10% ammonium persulfate were added to 720 ~tl H~O.The solution was degased and 282 I~1 of Pharmalyte 3-10 was added.The gels were overlayed with 8 M urea and polymerized h.Samples containing 100 000-200 000 cpm were mixed with 5 lal (51xg each protein) of BRL High Molecular Weight Marker Kit and solid urea added to 8 M. The sample was loaded on the gel and overlayed with 4 M urea and electrofocused at 8°C, 500 V for h then increased to 1 000 V for 1 h.The upper chamber (cathode) contained 0.01 M ethylenediamine and the lower chamber (anode) contained 0.01 M iminodiacetic acid.After focusing the gels were removed and equilibrated in 2.3% SDS, 63 mM Tris-HCl, 5% 2-mercaptoethanol, 10% (w/v) glycerol, 0.01% bromophenol blue for 20 min at room temperature and either loaded immediately onto a second dimension gel or frozen at -70°C and used within 2 weeks.

SDS gels, second dimension.
A 10% acrylamide gel (1.5 mm X 10 cm X 14 cm) was prepared as follows: the lower gel (6.4 ml lower gel buffer [7], 8.5 ml of a 25.8% acrylamide-4.2%N,N'diallyltartardiamide solution, 10.6 ml H20, 150 lal 10% ammonium persulfate, and 15 lal TEMED) was poured, overlaid with sec-butanol saturated with one quarter strength lower gel buffer and polymerized overnight at room temperature.The upper gel solution (2.5 ml upper gel buffer [7], 1.5 ml of a 30% acrylamide-1.5%N,N'-diallyltartardiamide solution, 6 ml H20, 30 la110% ammonium persulfate, and 10 lal of TEMED), was layered on the lower gel, and polymerized for to 2 h.The equilibrated isoelectric focusing gel was layered on the second dimension gel, sealed in place with a 1% agarose solution, and electrophoresed at 20 mA at 4°C until the tracking dye ran off the bottom.
One dimensional SDS gels.Lower and upper gels were made as described above for second dimension gels.Samples containing 50 000-100 000 cpm were mixed with an equal volume of 2 X SDS sample buffer and boiled 3 min before loading.Gels were electrophoresed as described above.

Use of internal standards in 2 dimensional gels. Three proteins supplied in the BRL
High Molecular Weight Protein Marker Kit had pl values within the range of the isoelectric focusing gel: phosphorylase b 92 500 M r, pI= 6.9; bovine serum albumin 68 000 M r , pl ---5.8; and ovalbumin 45 000 M r, pl = 5.4.The addition of marker proteins at 5 lag each did not interfere with running of the gel sample and each was detectable by staining (Fig. 1).These provided internal standards which, when compared to the autoradiograph of the gel, allowed an accurate assignment ofpI and M r values to spots present on the autoradiographs.Accordingly, it allowed a direct comparison of autoradiographs of different samples.

RESULTS
lnternal standards.In an attempt to assign more accurately pl and M r values to proteins detected by two dimensional gel electhrophoresis, a system of internal standards was employed.As shown in Fig, 1, the position in the 2nd dimension gel of three of the proteins used as internal standards was logarithmically related to the apparent molecular weights of these proteins and linearly related to their isoelectric points.Thus, by employing internal protein standards both p! and M r values can accurately be assigned to the surface proteins detected by two dimensional gel analysis.
Analysis of iodinated cell surface proteins.One and two dimensional gel electrophoretic profiles of radioiodinated cell surface proteins of the six different stages of T. cruzi are shown in Fig. 2. The apparent molecular weights and pl values of the individual protein species identified by this approach are presented in Table I.The major surface proteins which were detected by this method were relatively acidic with p/values in the range of 5.2 to 6.6.As shown in Fig. 2  stream and culture-form trypomastigotes were quite similar.Both forms possess several major proteins of 90 000 M r with pI values in the 5.5 to 6.6 range.This observation is in agreement with previous studies which have identified a 90 000 M r protein on the surface of one or both of these forms [2][3][4][5][6].However, this is the first indication that the single band at 90 000 Mr, previously seen only on one dimensional gel profiles, represents multiple protein species with the same molecular weight.Several surface components present on the bloodstream trypomastigotes are absent on the culture-form trypomastigotes.In particular, a major surface component of 68 000 M r and several proteins of > 100 000 M r were observed only on the bloodstream trypomastigotes.Considering the 68 000 M r protein's molecular weight and pI, it is likely that this component is bovine serum albumin introduced during the isolation of the bloodstream trypomastigotes.It is also likely that several of the >100 000 M r proteins represent contaminants that arise from platelets which remain in the preparation [3].A major protein of 90 000 My was also observed on the staphylomastigotes, a form which arises by transformation of either bloodstream or culture-form trypomastigotes in acellular culture conditions to an amastigote like form capable of cellular division [10].However, unlike bloodstream and culture-form trypomastigotes only one protein was observed at 90 000 M r by two dimensional gel analysis.Surprisingly, the metacyclic trypomastigote which, like the bloodstream and culture-form trypomastigotes, is an invasive form of the parasite, showed a surface protein pattern quite distinct from that of either of the latter forms.In fact, the profile of the iiii!iiii!ii Fig. 2. One and two dimensional gels 125 • of I-labeled surface proteins of various forms of T. cruzi.• designates the positions of the internal standards in the second dimension gel as determined by staining with Coomassie Blue (see Fig. 1).The arrow indicates the position of the protein that reacted with antibody (See Fig. 4).metacyclic trypomastigote surface proteins was remarkably similar to that of the epimastigote form.In both the epimastigote and metacyclic trypomastigote several major surface proteins were clustered in the range of 72 000 to 77 000M r with plvalues between 5.2 and 6.9.The metacyclic trypomastigote profile did, however, show a major surface component at 85 000 M r which is absent from the epimastigote profile.
In neither the metacyclic trypomastigote nor the epimastigote profile were we able to detect a 90 000 M e protein similar to that observed in the bloodstream trypomastigote and culture-form trypomastigote profiles.
Identification of immunodominant surface antigen.Although Nogueira et al. [3] also reported the absence of a 90 000 M r protein on the surface of the epimastigote and metacyclic trypomastigote, several other laboratories have clearly identified a major 90 000 M r protein on the surface of the epimastigote [2,5,6].The latter studies are particularly relevant since, collectively, they demonstrate that the 90 000 M r surface polypeptide on bloodstream trypomastigote, culture-form trypomastigote and epimastigote is the immunodominant surface component recognized by antibodies in human IgG obtained from Chagasic patients, serum from mice infected with T. cruzi and rabbit anti-trypomastigote serum.One notable possibility that could account for the variance between our results and those of other investigators is that the 90 000 M r protein is present in the epimastigote and metacyclic trypomastigote but is refractory to labeling by the conditions employed in this study.This possibility is directly amenable to investigation by Western blot analysis.Each of the six forms of T. cruzi were individually solubilized and the lysates were subjected to one dimensional gel electrophoresis.The fractionated proteins were transferred to nitrocellulose by electroelution [18] and reacted with hyperimmune mouse serum.As shown in Fig. 3, T. cruzi specific antibodies present in the sera of infected mice reacted strongly with a protein(s) of 90 000 M r in four of the six forms of the parasite.No reaction, however, was observed with either the epimastigote or metacyclic trypomastigote forms.This result is consistent with the surface protein labeling studies and further suggests that this antigen is not present in either the epimastigote or metacyclic trypomastigote.Since the two dimensional gel profiles of the bloodstream trypomastigotes and culture-form trypomastigotes identified several proteins of 90 000 Mr, it was desirable to determine how many of these proteins were recognized by antibodies in the hyperimmune mouse sera.As shown in Fig. 4, a Western blot analysis of a two dimensional gel of bloodstream trypomastigotes indicates that two major proteins of 90 000 M r and pl 5.6--5.8were reactive while the 90 000 M r protein at pi6.2 (Fig. 2) was not recognized by the antiserum.Interestingly, one of these proteins possesses similar M r and pI values to the 90 000 M r protein observed on the surface of staphylomastigotes by both surface labeling studies and Western blot analysis.If further analysis confirms that the 90 000 M r antigen on the staphylomastigotes is identical to one of those present on trypomastigotes, the comparative ease with which staphylomastigotes may be cultured would make this form a good candidate for use in the isolation of the antigen in large quantities.

DISCUSSION
In general, our results are in agreement with those of Noguiera et al. [3,4], who have identified a 90 000 M r antigen on the surface of bloodstream trypomastigotes but have not detected a similar antigen on the surface of either the epimastigote or metacyclic trypomastigote.Both results are, however, at variance with reports on the presence of a 90 000 M r antigen of the surface of epimastigotes [2,5,6].Although the basis for these differences has not yet been determined, several possible explanations have been examined.
Zingales et al. [6] suggest that the surface proteins of T. cruzi are more amenable to detection when t3q rather than 125I is employed as the radionuclide.Whether or not this is the case, we do not favor the possibility that the 90 000 M r peptide is present on the surface of the epimastigote but is refractory to detection by the radioiodination procedures used in our studies.The results of the Western blot analysis (Fig. 3) clearly show that the 90 000 M r antigen is found in four of the stages examined but is absent in the epimastigote and metacyclic trypomastigote.Since this approach does not rely on the cellular location of the antigen for detection, it strongly argues against the presence of the 90 000 M r polypeptide as an abundant antigen either on the surface or as an internal protein in the epimastigote and metacyclic trypomastigote forms.
We have also examined the possibility that the appearance of the 90 000 M r surface antigen on epimastigotes depends on cell growth (i.e.log vs stationary) and/or culture conditions (Beard and Manning, manuscript in preparation).Although qualitative and quantitative differences in peptides present on the surface of epimastigotes of the Y and Peru strain were observed, no 90 000 M r surface peptide was detected by these studies.We have also isolated the surface proteins of both the epimastigote and culture-form trypomastigote of the Peru strain by iminobiotin-avidin affinity chromatography ([20] and Beard and Manning, manuscript in preparation).In these preparations the 90 000 M r antigen represents the major surface protein from the culture-form trypomastigote but is not detected in the epimastigote.
Morel et al. [2 l] report the interesting finding that individual clones of laboratory stocks of both the Y and C1 strains of T. cruzi exhibit heterogeneity in their K-DNA restriction endonuclease patterns.They also report similar results for putative Y strain cultures obtained from different laboratories.In view of these results, it is conceivable that the Y and C 1 strains used by the various investigators for surface protein analysis exhibit different biological properties; one of which may be the expression of the gene sequence encoding the 90 000 M r protein in the epimastigote stage.Studies which will test this possibility are in progress, In conclusion, we have presented evidence that the surface protein patterns of the bloodstream and culture-form trypomastigotes are quite similar with both forms having several 90 000 M r surface peptides not all of which are recognized by hyperimmune mouse serum.The epimastigote and metacyclic trypomastigote surface protein patterns are also comparable and are distinctly different from those of the bloodstream and culture-form trypomastigotes.The major surface components on the epimastigote and metacyclic trypomastigote are a group of proteins of 72-77 000 M r.Proteins of similar M r and pI were not observed in the bloodstream or culture-form trypomastigotes.In addition, the 90 000 M r immunoreactive protein(s) present on the bloodstream, culture-form trypomastigotes, staphylomastigotes and amastigotes could not be detected on epimastigotes or metacyclic trypomastigotes by either the surface labeling techniques or Western blot analysis with hyperimmune mouse serum.

Fig. 1 .
Fig. 1.Two dimensional gel profile of internal protein standards stained with Coomassie Blue.Inset shows the relationship between distance migrated by phosphorylase b (92 500 Mr) , bovine serum albumin (68 000 M) and ovalbumin (43 000 Mr) in the first dimension (IEF) and their respective pI values (A).

TABLE I -
Cell surface proteins on six stages of T. cruzi *Major proteins.