Purification and Characterization

Adenylate cyclase activity associated with Trypanosoma cruzi sedimentable fractions was solubilized by treatment with the non-ionic detergent Lubrol PX and 0.5 M-(NH4)2SO4. The following hydrodynamic and molecular parameters were established for a partially purified enzyme-detergent complex: sedimentation coefficient 6.2 S; Stokes radius 5.65 nm; partial specific volume 0.83 ml/g; Mr 244000; frictional ratio 1.33. A Mr of about 124000 was calculated for the detergent-free protein from these parameters. The pl of this enzyme activity was 6.2. A monoclonal antibody to T. cruzi adenylate cyclase was obtained, which inhibited cyclase activities from several lower eukaryotic organisms. The T. cruzi adenylate cyclase was further purified by using this antibody in immunoaffinity chromatographic columns. Fractions obtained after this chromatography showed, on SDS/polyacrylamide-gel electrophoresis, a main polypeptide band with an apparent Mr of about 56000, which specifically reacted with the monoclonal antibody.


INTRODUCTION
The enzyme system responsible for cyclic AMP synthesis in lower eukaryotic organisms may be an important tool for functional and structural studies of the catalytic entity of adenylate cyclase. In Neurospora crassa, Saccharomyces cerevisiae, Mucor rouxii, Trypanosoma cruzi and Blastocladiella emersonii, adenylate cyclase activity has two interestingproperties: dependency on Mn2+ and insensitivity to fluoride (Flawia & Torres, 1972;Paveto et al., 1975;Varimo & Londesborough, 1976;da Silveira et al., 1977;Lopez-Gomez et al., 1978). Moreover, only crude fresh preparations from Neurospora and yeast show a significant activity in assays performed with Mg2+ and guanine nucleotides (Casperson et al., 1983;Rosenberg & Pall, 1983).
In Neurospora crassa, the adenylate cyclase catalytic entity, which has a loose association with membranes, may interact with regulatory factors from avian erythrocyte membranes. The reconstituted cyclase system is active with Mg2+ and shows stimulatory responses to guanine nucleotides, isoprenaline and fluoride (Flawiai et al., 1983). In addition, Neurospora adenylate cyclase may interact with calmodulin, which induces a Ca2+-dependent stimulation of this enzyme activity in the presence of Mg2+ (Reig et al., 1984). As a general hypothesis, it may be postulated that some adenylate cyclases from lower eukaryotic organisms consist of catalytic entities similar to those found in avian or mammalian membranes, but lacking most of the regulatory factors which are characteristic of adenylate cyclases in these membranes. In fact, some functional and physical properties of Neurospora cyclase are strikingly similar to the enzyme system found in the variant of S49 lymphoma cells deficient in the guanine-nucleotide-binding protein N. Reig et al., 1982;Ross et al., 1978).
Immunological studies might extend evidence on the similarities between adenylate cyclases from lower and higher eukaryotic organisms. A previous paper from our laboratory reported the preparation of a monoclonal antibody against Neurospora adenylate cyclase (Molina y Vedia et al., 1983). The antibody inhibits Neurospora cyclase activity as well as other cyclases from lower eukaryotic organisms, but not adenylate cyclases from avian or mammalian tissues. This may indicate that some epitopic areas in the enzyme molecule were not conserved during evolution.
The present paper extends these studies to Trypanosoma cruzi adenylate cyclase activity, i.e. purification, characterization and interaction of the enzyme with a monoclonal antibody.

EXPERIMENTAL Materials
The sources of materials used in this work have been given in previous papers Reig et al., 1982;Kornblihtt et al., 1981). Immunoglobulins were obtainedfromCooper Biomedical, Malvern, Philadelphia, PA, U.S.A., or from Zymed Laboratories, South San Francisco, CA, U.S.A. Trypanosoma cruzi membranes Epimastigotes of the Tulahuen 2 strain were obtained from axenic biphasic cultures (Jerez de Burgos et al., 1976). Cells were collected by centrifugation at 10000 g for 20 min, and homogenized in 0.25 M-sucrose containing 5 mM-KCl (10 ml/g ofmaterial) with a Sorvall-Ribi press operated at 34.5 MPa (5000 lb/in2) under a N2 atmosphere. Cell debris was discarded by centrifugation at 10000g for 10min and the supernatant fluid was further centrifuged at 105 000 g for 60 min. Pellets ('crude membranes') thus obtained were stored at -20 'C.

Hexylamino-Sepharose chromatography
The column (1 cm x 12 cm) equilibrated with buffer A was loaded with 10 ml of the 'solubilized extract' (4 mg of protein/ml). After washing with 20 ml of this buffer solution, the column was subjected to a stepwise elution with 5 ml portions ofNaCl solutions (0.1, 0.2,0.4,0.5 and 1.0 M respectively) in buffer A. Fractions (1 ml) were collected at a rate of 1 ml/min. Adenylate cyclase activity was eluted at 0.3 M-NaCl. Fractions containing the activity were pooled and dialysed against buffer A (giving 'hexylamino-Sepharose fraction').

Immunoaffinity chromatography
The 'hexylamino-Sepharose fraction' was also purified by chromatography on a column containing a monoclonal antibody to T. cruzi adenylate cyclase, bound to Sepharose. The immunoglobulin, prepared and purified by the procedure described below, was bound to Sepharose 4B by the method of Cuatrecasas (1970). The immunoaffinity support contained about 4 mg of bound protein/ml of packed bed volume. The column (7 cm x 0.75 cm), equilibrated with 50 mM-Tris/HCl containing 1 mM-,f-mercaptoethanol (buffer B), was loaded with 2.5 ml of the enzyme fraction previously concentrated in a BR15 Minicon ultrafilter (Amicon Corp.) to a protein concentration of about 0.9 mg/ml. The non-adsorbed protein fraction was re-loaded on the column; this step was repeated six times. Then the column was washed with 50 ml of buffer B containing 0.5 M-NaCl and 0.5°% Lubrol PX and eluted with 50 mM-triethylamine (pH 11.8); 0.5 ml fractions were collected. The eluted fractions were immediately neutralized by addition of appropriate volumes (about 100 41) of 1 M-Mops buffer, pH 7.1.

Molecular and hydrodynamic parameters
Determinations were carried out by the procedures described by Reig et al. (1982). Sucrose gradients were made in buffer A (with H20 or 2H20) containing 0.15 M-NaCl. Gel-filtration studies were performed on Bio-Gel A-Sm columns (75 cm x 0.9 cm) equilibrated with buffer A containing 0.2 M-NaCl.

Preparation of monoclonal antibodies
The procedure followed for this has been described elsewhere (Molina y Vedia et al., 1983). Clones were selected for the capacity of culture media to inhibit cyclase activity in a 'Bio-Gel fraction' of Neurospora (Reig et al., 1982) or in T. cruzi 'crude membranes'. Purification of immunoglobulins from culture media was as described by Fundenberg (1967). Characterization of the secreted immunoglobulins was performed by immunoelectrophoresis as described elsewhere (Williams, 1971). Adenylate cyclase assay Enzyme activity was assayed by the method of Flawiai et al. (1983) in the presence of 2.5 mM-MnCl2.
Other analytical procedures Details of the procedure for determination of protein concentration, activity of enzyme markers and polyacrylamide-gel electrophoresis in the presence of SDS were given in previous papers (Flawiai et al., 1983;Reig et al., 1982;Kornblihtt et al., 1981). Polyacrylamide gels were stained with Coomassie Brilliant Blue R-250 (Reig et al., 1982) or with AgNO3 by the procedure of Oakley et al. (1980). Electrophoretic transfer of polypeptides from polyacrylamide gels to nitrocellulose membranes was performed by the method ofTowbin et al. (1979) with some modifications. Two samples of the protein were subjected to electrophoresis in a polyacrylamide gel slab in the presence of SDS (Laemmli, 1970). After electrophoresis, the gel was cut with a razor blade and the lanes corresponding to Mr standards and to one of the samples were stained with silver by the procedure of Oakley et al. (1980). The other portion of the gel was subjected to electrophoretic transfer of polypeptides to a nitrocellulose sheet by the procedure of Towbin et al. (1979). A current of 1 A was applied for 3 h in a chamber containing 50 mM-sodium phosphate buffer, pH 6.5. The temperature ofthe electrophoretic system was maintained at about 25 'C.
After transfer, the nitrocellulose sheet was subjected to successive washings, with 50 mM-Tris/HCl buffer, pH 7.4, containing 0.14 M-NaCl (TBS buffer) (5 min at room temperature), with 300 bovine serum albumin in TBS buffer (30 min at 37 'C), and four times with TBS buffer (10 min each at room temperature). After that, the nitrocellulose membrane was incubated in a polyethylene bag for 14 h at room temperature in the presence of the purified monoclonal antibody (about 1 mg of protein) in 10 ml ofTBS buffer containing 3% bovine serum albumin and 10% mouse serum. The sheet was then washed three times with TBS buffer (10 min each, room temperature) and incubated for 3 h at room temperature with a 1: 900 dilution ofgoat anti-(mouse IgG) (heavyand light-chainspecific) conjugated to peroxidase in TBS buffer containing 3 % bovine serum albumin. After washing five times with TBS buffer (5 min each, room temperature) the membrane was treated for 30 min at room temperature with a solution containing 250,ug of o-dianisidine/ml and 0.075% H202 in 50 mM-Tris/HCl buffer, pH 7.4. RESULTS AND DISCUSSION Adenylate cyclase solubilization Several procedures were used for enzyme extraction from T. cruzi particulate fractions. They involved the use of several detergents (ionic or non-ionic), urea or guanidinium chloride, with or without sonication or freeze-thawing. However, the only method giving a reasonable solubilization of adenylate cyclase activity (about 50 % ) was extraction in the presence ofa non-ionic detergent (0.1 % Lubrol PX) in the presence ofa relatively high (NH4)2S04 concentration (0.5 M). The salt requirement may be attributable to the chaotropic effect of (NH4)2S04 .

Enzyme purification
Adenylate cyclase activity in the solubilized extracts was purified to provide an adequate antigen to generate monoclonal antibodies. As occurs with Neurospora (Reig et al., 1982) and testicular (Kornblihtt et al., 1981) adenylate cyclase, the loss of activity was the major problem in purification of the T. cruzi enzyme. Recovery was not improved by the addition of glycerol, by increase in ,J-mercaptoethanol concentration, by pH change of the elution buffers, or by addition of polyamines etc. After two purification steps such as hexylamino-Sepharose chromatography and isoelectric focusing, 99% of the enzyme activity was lost (Table 1). However, as shown in Fig. 1, isoelectric focusing in a liquid column gave a reasonable resolution of most of the protein from the enzyme activity.

Molecular parameters
Enzyme samples from hexylamino-Sepharose fractions were centrifuged in sucrose gradients made in H20 or 2H20, or gel-filtered through Bio-Gel A-5m columns as previously described (Reig et al., 1982). Results from an average of at least five different experiments are shown in Table 2. The calculated Mr for the enzyme-detergent complex was 244000. Assuming that the partial specific volume of the protein is 0.735 and that of the detergent 0.958 (Haga et al., 1977), the Mr for the enzyme should be about 124000. This value is twice that reported for testicular soluble adenylate cyclase and two-thirds that of the Neurospora enzyme. These two enzymes are considered to be lipid-free soluble proteins (Kornblihtt et al., 1981;Reig et al., 1982). Monoclonal antibodies A N. crassa soluble adenylate cyclase preparation was used to test the inhibitory action on enzyme activity of culture media from different clones. The reason for using the Neurospora enzyme was the simplicity of obtaining it   et al., 1983). Table 3 shows the effect of preincubation with control medium or media from several clones on Neurospora adenylate cyclase activity. Clone 36 was selected for these studies and subjected to further subcloning by the limiting-dilution procedure. After subcloning, the immunoglobulin secreted by the culture medium from the selected subclone (designated subclone 36) was identified by immunoelectrophoresis as a mouse immunoglobulin of the type G 2b. Culture medium from the subclone 36 was subjected to a purification protocol (Fundenberg, 1967) which involved precipitation with (NH4)2S04 followed by chromatography on DEAEcellulose and dialysis. A similar protocol was used for the purification of a control medium from a myeloma-cell culture. A sample of the immunoglobulin preparation from subclone 36 was treated with rabbit anti-(mouse IgG). The effect of these preparations on Trypanosoma Other conditions were given in the Experimental section. adenylate cyclase was tested as shown in Fig. 2. Inhibition by the purified fraction from subclone 36 was roughly proportional to the protein concentration. No effect was detected when this fraction was pretreated with anti-(mouse IgG) or when a protein fraction from a control medium was used (Fig. 2a). On the other hand, the inhibitory effect required a preincubation step of the adenylate cyclase preparation with the hybridoma immunoglobulin (Fig. 2b). Reactivity of the subclone 36 antibody with adenylate cyclase from different sources was also studied. As shown in Table 4, the antibody elicited variable extents of inhibition of Mn2+-dependent adenylate cyclases from lower eukaryotic organisms, but not on enzyme Table 4. Effect of control and subclone-36 media on T. cruzi adenylate cyclase activity References: 'the present paper; 2Reig et al. 3Flawia' et al. (1983);4Paveto et al. (1975); 'Harwood & Peterkofsky (1975);6Kornblihtt et al. (1981);7Neville (1968 preparations from E. coli, rat testes, rat liver membranes or turkey erythrocyte membranes assayed with Mn2+. Purification by immunoaffinity chromatography Active fractions from hexylamino-Sepharose columns were purified by affinity chromatography on Sepharose 4B columns having the purified monoclonal antibody to T. cruzi cyclase bound to this matrix. After absorption, the enzyme activity was eluted by a drastic increase of the pH. Control experiments revealed that the extent of enzyme irreversibly inactivated after such a pH change was about 90 %. On the other hand, it was difficult to determine the very low protein concentration in the eluate. For this reason protein was assayed on a pool of fraction from four different immunoaffinity chroma, tographies. Correcting for enzyme inactivation, the extent of adenylate cyclase purification after this step was over 5000-fold (Table 1). On SDS/polyacrylamide-gel electrophoresis the fraction purified by immunoaffinity chromatography showed polypeptidebandswithapparent Mr values of about 56500, 49000 and 34000. After transfer to nitrocellulose membranes, the main polypeptide band (Mr 56 500) was the only one reacting with the monoclonal antibody secreted by subclone 36 (Fig. 3).
Evidence reported in this paper indicates that a polypeptide of Mr about 56000 may be an integral part of the T. cruzi adenylate cyclase molecule. Other Mn2+-dependent adenylate cyclases, from rat testicular tissue and Neurospora, have been characterized in this laboratory. Purified preparations of these cyclases showed the existence of polypeptide components with apparent Mr values slightly higher than that found in the T. cruzi enzyme: 69000 and 66000 respectively (Kornblihtt et al., 1981;Reig et al., 1982). Another interesting fact is that the monoclonal antibody to T. cruzi adenylate cyclase, which inhibits this enzyme activity, also inhibits Neurospora adenylate cyclase activity. As reported elsewhere, the reverse is also true: a monoclonal antibody to Neurospora adenylate cyclase inhibits both Neurospora and T. cruzi adenylate cyclase activities (Molina y Vedia et al., 1983). This evidence strongly indicates that the same epitope exists in the enzyme from these two lower eukaryotic organisms and that this epitope is in some way involved in the catalytic activity of these cyclases. Another characteristic of these monoclonal antibodies is that neither inhibits some adenylate cyclase activities showing the immunoprecipitation 'in situ' of the monoclonal antibody to T. cruzi adenylate cyclase. Positions of the Mr markers (phosphorylase b, bovine serum albumin, ovalbumin, carbonic anhydrase and trypsin inhibitor) are also given. Conditions were described in the Experimental section.
from higher eukaryotic organisms, indicating that some important epitopic areas of this enzyme are not stnrctly conserved during evolution.
Closely related to this latter point is the stimulatory Vol. 234 effect of the diterpene forskolin on the catalytic subunit of adenylate cyclases from higher eukaryotic organisms (Seamon et al., 1981). This property was used by Pfeuffer & Metzger (1982) to purify the catalytic entity from cardiac muscle adenylate cyclase. With the Neurospora enzyme (Flawiai et al., 1983) and the T. cruzi adenylate cyclase (results not shown), it was not possible to see any effect of the diterpene on these enzyme activities. This observation reinforces the view that some structural and functional characteristics of adenylate cyclase may not be maintained through the evolutionary scale of eukaryotic organisms.