Isolation, cDNA sequences, and biochemical characterization of the major cyclosporin-binding proteins of Toxoplasma gondii.

The activities of the immunosuppressive, antifungal compounds cyclosporin A (CsA), FK-506, and rapamycin are dependent upon high affinity binding proteins collectively termed immunophilins. We report the isolation, biochemical characterization, and amino acid sequences of two major CsA-binding proteins, cyclophilins, from the pathogenic protozoan, Toxoplasma gondii. The 18.5- and 20-kDa molecular mass proteins exhibit peptidylproline cis-trans-isomerase activity, which is inhibitable by 10(-8) M CsA. The amino acid sequences of these two proteins, deduced from cDNA clones, reveal up to 70% amino acid identity to previously isolated cyclophilins. The 18.5-kDa protein appears to be synthesized as a precursor with a 15 amino acid signal peptide. The amino-terminal region of the mature 20-kDa protein has significant homology to the B subunit of the calmodulin-dependent phosphatase, calcineurin. The two T. gondii cyclophilins are products of different genes and appear to have different subcellular distributions.

The T-cell immunosuppressive agents cyclosporin A (CsA),' FK-506, and rapamycin have revolutionized solid organ and bone marrow transplantation (1)(2)(3). All three compounds, however, were initially described as antifungal agents (4)(5)(6) and have subsequently been shown to have antimicrobial activity against a variety of eukaryotic pathogens (reviewed in Refs. 7 and 8). Both the immunosuppressive and antimicrobial effects of these compounds involve receptor proteins collectively termed immunophilins . Immunophilins catalyze the isomerization of peptidylproline bonds (PPIase activity) (10,(13)(14)(15), and s t e r e o s~c i~c~l y bind either CsA or rapamycin a n d FK-506. In addition to their isolation from m a m m a l i~ sources, CsA-binding PPIases or cyclophilins (CyPsf have been * This work was supported by American Cancer Society Grant CH67, National Institutes of Health Grants GM 40660, GM 49858, and DK 07476, and National Cooperative Drug Discovery Group Grant UO1 AI 31808. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. to the GenBankTM/Eit4BL Data Bank with accession number[s) U04633.
The nucleotide sequencers) reported in this paper has been submitted  identified from several species of fungi Echinococcus granulosus (191, and Schistosoma japonicum (20). Related prokaryotic PPIases have been found in Escherichia coli (21,221, Salmonella typhi (231, and Bacillus subtilis (24). FK-506-binding proteins have been isolated from fungi (9,(25)(26)(27), ~e i s s e r i u meningitidis (281, ~h l~m y d i u ~r u c~o m a t i s (29,30), and Legionella sp. (31)(32)(33). A better understanding of the key biochemical pathways involving microbial immunophilins may uncover unique sites for antibiotic drug development.
The CsA-susceptible protozoan, Toxoplasma gondii (34,35), is a leading cause of encephalitis in immunosuppressed hosts (36). I: gondii, in its infective stage, actively invades host cells and survives within a specialized vacuole (371. Because PPIases from other pathogens with a similar intracellular survival strategy have been shown to be virulence factors (31321, we hypothesized that the anti-toxoplasma effects of CsA might be mediated through a cyclosporin-sensitive PPIase. This report details the isolation, biochemical characterization, and amino acid sequences of two CsA-sensitive Gyps from T. gondii. We speculate on the role of these proteins in the intracellular survival of Z gondii. EXPERIMENTAL PROCEDURES Reagents-Cyclosporin A and 8-ornithino-CsA were gifts from Dr. J. Bore1 of Sandoz (Basel, Switzerland) and Dr. P. Durette of Merck, Inc. (Rahway, NJ), respectively. Substrate for the PPIase activity assay (succinyl-alanine-alanine-proline-phenylalanine (p-nitroanilide)) was a gift from Miles Laboratories, Inc. (West Haven, CT). AfEnity columns of 8-ornithino-CsA were prepared using Mi-Gel 10 resin (Bio-Rad, Hercules, CAI following the manufacturer's instructions. Recombinant human cyclophilin and affinity-purified anti-cyclophilin antibodies were prepared in our laboratory, as previously described (38)(39)(40). CHAPS was purchased from Boehringer M a n~e i m ; phenylmethylsulfonyl fluoride, leupeptin, and aprotinin were purchased from Sigma. The X gtll T gondii cDNA library was a gift of J. Boothroyd, Stanford University. I: gondii Isolation and Preparation of I: gondii Lysates-Rh strain I: gondii were grown either in vitro in cell cultures of human foreskin fibroblasts (American Type Culture Collection, Rockville, MD) or in viuo in female BALB/c or Swiss-Webster mice (Charles River, Kingston, MA) innoculated intraperitoneally. I: gondii were harvested from mice by peritoneal lavage, and the resulting suspension was passed twice through a 27-gauge needle, followed by filtration through a 3-pm membrane (Millipore Corp., Bedford, M A ) to exclude macrophages. Organisms harvested from mice or tissue culture were isolated by centrifugation at 1000 x g for 8 min at 4 "C, resuspended in phosphate-buffered saline, pH 7.4, and centrifuged again at 1000 x g for 8 min. The I: gondii pellet was resuspended in 500-1000 pl of bufYer (50 w TriSiIIC1, pH 7.2, 150 m M KCl, 10 nm glucose, 5 ~l l~ ~-mercaptoethanol, 1 m EDTA) plus protease inhibitors t l p~enylmethylsulfonly fluoride, 1 pgiml leupeptin, 1 p@mI aprotinin). An equal volume of 3% CHAPS (in the same buffer) was added after the organisms were adequately resuspended (CHAF'S final concentration, 1.5%). The lysate was incubated on ice for 45 min followed by centrifugation at 100,000 x g for 30 min. For some experiments, frozen 1: gondii lysates provided by Dr. Ben Luft, State University of New York Stoneybrook, were extracted with CHAPS plus protease inhibitors and processed in an identical fashion. The pellet was discarded, and the supernatant (hereafter referred to as whole I: gondii lysate) was processed further as described below.

II gondii Cyclophilins
For subcellular localization experiments, Z gondii prepared as above were resuspended in p h o s p h a~b~e r e d saline (with phenylmethylsulfonyl fluoride, leupeptin, and aprotinin as above) rather than CHAPS buffer. Organisms were frozen at 20 "C, thawed twice, and then lysed by vortexing for 20 s with one-half volume of glass beads (0.5 mm). The resulting lysates were spun at 10,000 x g. The pellet was extracted with 1.5% CHAPS for 45 min on ice (hereafter, this fraction is referred to as the 10,000 x g pellet extract). The supernatant (hereafter, this fraction is referred to as the 10,000 x g supernatant extract) and the 10,000 x g pellet extract were both spun at 100,000 x g to remove particulates and processed further as described below.
QuantitutionlPurification of CsA-binding Proteins-T. gondii lysates were screened for CsA binding activity and the concentration of protein with high affinity for CsA determined using the LH-20 assay previously described (41,42). The percentage of CsA-binding protein was then calculated by determining the total protein concentration using the Bic-Rad assay (Hercules, CA) with bovine albumin standards.
The I: gondii lysates were passed through 0.3 ml of a previously described 8-ornithino-CsA affinity column (43). The column was then washed with at least 20 volumes of buffer (20 m K,HPO,, pH 7.4,lOO m~ NaCI) and eluted with 10 pg of CsAiml in wash buffer. In later experiments, the buffer was adjusted to pH 3.5 or 3.3 and used without added CsA to elute the column; the eluate was immediately neutralized by mixing in a 1:2 ratio (buffereluate) with 200 m M K2HP04, pH 8.0, 100 m~ NaC1, 5 m P-mercaptoethanol buffer (final pH of eluate, approximately 7.5). Eluted proteins were concentrated using a Centricon-10 filter (Amicon, Beverly, MA).
Production of Anti-I: gondiz Cyclophilin Antiserum-Approximately 2 pg of each I: gondii cyclophilin isolated as above were cut out of SDS-PAGE gels, and the gel piece containing both protein bands was homogeNzed together in the minimum volume of phosphate-b~ered saline. The emulsion was then mixed with equal amounts of Freund's complete adjuvant and injected subcutaneously into male Sprague-Dawley rats (National Institutes of Health, Bethesda, MD). Rats were boosted with similar preparations mixed with incomplete Freund's adjuvant on days 14 and 35 after the initial immunization. The animals were then boosted and serum collected at periodic intervals. Immunoblotting-Proteins isolated by SDS-PAGE were transferred to nitrocellulose filters using transfer buffer (25 m~ Tris, 192 m M glycine, 20% MeOH, and 0.1% SDS) in a U. S. A. Scientific Trans-blotter (Minneapolis, M N ) at 12 V for 2 h. Filters were blocked with 5% nonfat dried skim milk (BLOTTO) in Tris-buffered saline plus 0.1% Tween-20 (TBS-T) at mom temperature for 2 h. After incubation with antisera diluted 1:50@"1:10,000 in 5% BLOTTO/TBS-T for 1 h, the filters were washed three times with 5% BLOTTOPTBS-T (5 min each) and then with TBS-T for 30 min. Filters were then incubated with horseradishperoxidase-~nked goat anti-rat IgG or goat anti-rabbit I& (Sigma) diluted 1:2000 in TBS-T for 1 h and washed three times (5 min each) with TBS-T and then for 30 min in TBS-T. Proteins were visualized using the ECL system (Amersham Corp., Chicago, IL) and detection of chemiluminescence on x-ray film (Kodak-XAR).
Glycosylation Detection-Proteins were transferred to nitrocellulose filters as above. Protein glycosylation was investigated using the glycan detection kit (Boehringer Mannheim), applying method B described by the manufacturer, and using transfenin as a positive control.
Peptide Sequence Analysis--Proteins eluted from the 8-ornithino-CsA column were fractionated by SDS-PAGE and transferred to polyvinylidene difluoride membranes (Millipore Corp., Bedford, MA) as described above for immunoblotting. Polyvinylidene difluoride membranes were stained with 0.5% Coomassie Blue in 50% MeOH, destained in 50% MeOH, and washed with H,O for 5 min. The stained protein bands were cut out, and sequencing was performed in the William Keck F o~d a t i o~a l e School of Medicine, Protein and Nucleic Acid Chemistry Facility. NH2-terminal sequencing was accomplished using an Applied Biosystems model 470Agas phase Sequenator with an on-line model 120A phenylthiohydantoin derivative analyzer. Additional sequences were obtained by cyanogen bromide elution of the protein bands from the polyvinylidene difluoride membrane and subsequent trypsin digestion. Peptides were separated on a 2.1 x 250-mm Vydac C-18 reverse phase column in a Hewlett Packard model 1090 high pressure liquid chromatography system as described previously (44). Peptide fragments were sequenced as above and matched against those recorded in the National Center for Biotechnical Information data bases (45,46).
PPIase Assay-PPIase activity of I: gondii proteins eluted from the 8-ornithino-CsA column was assessed using fractions eluted with low pH buffer to avoid CsA inhibition. Isomerization of N-succinyl-Ala-Ala-~o-Phe(p-nitroa~lide) was measured in the chymotrypsin cleavage assay described by Kofron et al. 147) at 10 "C using 30 p~ substrate and 60 UM chymotrypsin in 35 m~ Hepes pH 8. I: gondii protein concentration was determined by amino acid analysis or Bio-Rad (Hercules, CA) protein assay standardized for these particular proteins based on amino acid analysis. CsA-mediated inhibition of PPIase activity was determined by addition of increasing amounts of CsA in 40% ethanol (not to exceed a final concentration of 0.4% ethanol) to the PPIase assay. The mixture was then incubated at 10 "C for 20 min, followed by the addition of chymotrypsin and then substrate. , and chlorofo~isoamyl alcohol (24:l). The DNA was precipitated using an equal volume of isopropanol, followed by incubation at -70 "C for 20 min and centrifugation at 12,000 x g at 4 "C for 10 min. The DNA pellet was washed with 70% ethanol, respun at 12,000 x g and the ethanol removed. After drying under vacuum, the DNA was resuspended in sterile H20. Approximately 10 ng of cDNA was used as a template for the PCR reaction performed in an Omnigene thermocycler (HybaidLabnet, Woodbridge, NJ) in a total volume of 50 pl using the degenerate primers above at a final concentration of 1.6 m, 200 m~ dNTPs, 1.25 units of Taq DNA polymerase (Perkin-Elmer Corp.), and 5 pl of 10 x PCR buffer (Perkin-Elmer Corp.). The following thermal cycle protocol was used: denatured DNAftemplate and primers) at 95 "C for 4 min, cool to 72 "c, followed by addition 0.5 pl of Taq DNA polymerase, then 30 cycles of denaturing at 95 "C x 1 mi~annealing at 50 "C x 1 mi~extension at 72 "C x 1 min. PCR products were separated on 3.5% Nuseive (FMC, Rockland, ME) agarose gels and visualized with ethidium bromide staining and ultraviolet light.
Screening of a I: gondii cDNA LibrarylCloning and Sequencing-Unique PCR products generated using the degenerate primers above were cloned into Bluescript KSplasmid vectors. Eledroporation-competent DH-5a E. coli were electrotransformed using a Bio-Rad gene pulser (Hercules, CAI and transformed bacteria grown in SOC media without ampicillin at 37 "C for 1 h. Aliquots of this suspension were transferred to L-B/ampicillin-selective agar plates and incubated overnight. Colonies containing inserts were purified to homogeneity and overnight cultures of these bacteria grown in L-B broth with 50 &ml ampicillin. Plasmid DNA was isolated using standard procedures (48).
Cloned inserts were sequenced using the Sequenase kit (U. S. Bio- chemical Corp.) and BsSs-dATP (DuPont NEN) using T3 and T7 primers corresponding to plasmid DNA regions. Sequenced DNA was separated on 6% polyacrylamide gels and the gels exposed to Kodak-XAR film (Rochester, N Y ) after drying.
"P-labeled DNA probes were prepared by labeling of the specific PCR products above using the Random primer kit (Boehringer Mannheim) and ["PldCTP (DuPont NEN) (specific activity, 107-10" d p d p g DNA). Approximately 50,000 A phage were plated and plaque lifts performed using nitrocellulose membranes (Schleicher and Schuell). Filters were hybridized overnight with labeled probe a t 42 "C in 50% formamide, 5 x Denhardt's, 0.1% SDS, 6 x TEN (10 mM Tris, pH 8.0, 100 mM NaCI, 1 mM EDTA) with 50 ng/ml tRNA. Filters were washed a t 62 "C in progressively lower concentrations of SSC buffer with 0.18 SDS until background radioactive counts were minimized. The filters were then exposed to Kodak-XAR film at -80 "C for 24-36 h. Positive plaques were identified, isolated, and serially screened two more times until all plaques were positive. cDNA from positive clones was prepared as described above, the inserts cut out with EcoRI restriction endonuclease (New England Biolabs, Beverly, MA), and cloned into Bluescript KS-plasmid. The sequences of the inserts were determined using the Sequenase kit (U. S. Biochemical Corp.) as described above. Some DNA sequencing was performed in the William Keck Foundation DNA sequencing facility using an Applied Biosystems model 373A Sequencer. New primers for sequencing were synthesized as cDNA data was obtained (see Fig. 6 for base sequences used). Because the cDNA encoding the 20-kDa T gondii CsA-binding protein contained an internal EcoRI site, two plasmids containing inserts were required to completely determine the cDNA sequence. Final sequencing of the cDNA encoding both major T gondii CsA-binding proteins was accomplished from both stands.

T gondii Has n o Major
CsA-binding Proteins-We first documented CsA binding activity in T gondii lysates. As determined with the LH20 binding assay, CsA-binding proteins comprised 0.1-0.2% of all CHAPS-extractable proteins in whole T gondii lysates. A one-step aftinity purification on 8-ornithino-CsA, followed by SDS-PAGE analysis demonstrated two major protein bands at approximately 18.5 and 20 kDa molecular mass (Fig. 1, lane B ) . Three much lighter staining bands, faintly visible in Fig. 1 and perhaps representing proteins with lower CsA affinity, were consistently noted at approximately 60-, 40-, and 26-kDa molecular mass when 8-ornithino-CsA columns were not washed with at least 20-column volumes of buffer before elution. Elution of the 8-ornithino-CsA aftinity column by progressively lowering the pH of the washing buffer from 3.5 to 3.3 successfully eluted the column without the need to add CsA and resolved the two major CsA-binding proteins ( Fig. 1, lanes C and D ) . Further characterization of these two protein bands demonstrated that neither was glycosylated (data not shown). The isoelectric point of the 18.5-kDa protein was 5.7. Two isoforms were noted in the 20-kDa protein fraction eluted at pH 3.3, one with a PI of 7.05, the other at 7.80 (Fig. 2).

T gondii CsA-binding Proteins Are Not Due to Host Cell
Protein Contamination-Two experiments were performed to demonstrate that these proteins were derived from T gondii and not host cells. First, CsA-binding proteins were also isolated by 8-ornithino-CsA affinity chromatography from uninfected human foreskin fibroblast cells. The major CsA-elutable protein from human foreskin fibroblast cells migrated with a lower apparent molecular mass on SDS-PAGE (18 kDa) than the two T gondii proteins (Fig. 3A ). Second, immunoblot analysis with a 1:lOOO dilution of affinity-purified anti-serum prepared against bovine thymus CyP-18 detected only recombinant human CyP-18 (HCyP-18) not the two T gondii CsAbinding proteins. Conversely, a 1:lOOO dilution of polyclonal anti-serum prepared against the T gondii CyPs easily visualized the T gondii proteins but only very weakly detected recombinant HCyP-18 (Fig. 3B).

T gondii CsA-binding Proteins Partition into Different Cell
Fractions-The distribution of the T gondii CyPs was assessed in crude cell fractionation experiments. Parasites were lysed by two freezekhaw cycles followed by vortexing with glass beads, then fractionated as described under "Experimental Procedures." CyPs of both 18.5 and 20 kDa were present in the 10,000 x g supernatant extract. However, a 20-kDa protein that eluted at pH 3.3 predominated in the 10,000 x g pellet extract (Fig. 4).

T gondii CsA-binding Proteins are PPZases-Both T gondii
CsA-binding protein fractions demonstrated PPIase activity a t concentrations comparable with mammalian CyPs (5-10 nM). The activity of both fractions was suppressed by CsA, with an IC5o for inhibiting solutions containing 10 nM T gondii CyP of approximately 32 nM for the T gondii CyP-18.5 fraction and 5 nM for the T gondii CyP-20 fraction (Fig. 5).

Peptide Sequence Analysis of I: gondii CsA-binding
Proteins--The two T gondii CsA-binding protein fractions were subjected to amino acid sequence analysis. NH2-terminal sequencing of the 18.5 kDa/pH 3.5 eluted protein yielded a 16- amino acid peptide (Fig. 6) unrelated to any known sequence in the protein data bases (Fig. 7 A 1. Two peptides derived from this protein by trypsin digestion had 618 and 616 amino acid identity with the highly conserved core region of HCyP-18.
A search of the protein data bases for homology to the 32 amino acid NH2-terminal peptide fragment from the 20-kDa T gondii CyP revealed an interesting association. The 37 highest homology scores matched sequences from previously identified CyPs. However, the best match score from a non-CyP protein was achieved by the B subunit of the human protein phosphatase, calcineurin. The two proteins share 11/20 amino acid identity in fragments of the carboxyl-terminal region of calcineurin B matched against the amino-terminal portion of T gondii CyP-20 (Fig. 7B). As determined by the complete cDNA sequence below, this segment of T gondii CyP-20 is the only region of homology to the calcineurin B subunit.
cDNA Cloning and Deduced Amino Acid Sequences of the T gondii CsA-binding Proteins-Full amino acid sequences were derived from A gtll clones isolated from a T gondii cDNA library (Fig. 6). The central regions of highly conserved amino acid sequence in eukaryotic CyPs are apparent in the the two 3: gondii proteins. The initial amino acid identified by NH2-terminal sequencing of the T gondii CyP-18.5 is located 15 amino acids downstream of the presumed initiation methionine. These 15 amino acids are typical of a signal peptide, and if translated, the molecular weight of the precursor protein would be 19,623. Although the mature protein migrates with an apparent molecular mass of 18.5 kDa, cleavage of the signal sequence would result in a protein with a calculated molecular weight of 17,890. A search of the major protein data bases indicates T gondii CyP-18.5 is most closely related to the CyPs of fungi, chickens, and mammals but shares significant homology with other parasitic CyPs (E. granulosus and S. japonicum) (Fig. 7).  a o s a t g c c g g a g l c a g a a 8 g g c g l a c a t g g a l~t c g a c a l c g a c g g a g a a c a t g c c g g E  N  A  G  V  R  K  A  Y  M  D  I  D  I  D  G  E  H  A  G   L  F  D  K  Y  K  G  S  V  F  H  R  I  I  P  D  F  M  I  Q   2 4 1 g g~g g n g~t l l c g~g a . c C . c n . c g g C~c t g g~g g~c~C 8 g C~t c f~C g g C C g~B g~t t 1 g s c g a c g a~~a c t t t g a 1 t t g n n g c~c g a g c g a g g c g t c a t c t c t a t g g c g~~c g c c g c c g a a c n c c~s t g g c n g c c a a t l l t c . c c r c c g t g 8 a g a c a g~g t g g c l c g~c g a c a c g t t g t t t t c g g g~~g a t~a c~~c t g~g t c g t g g c~t a c c g t c c~g g c t a t t g~ t t c t t c g a c~t c~g c a 1 t g a c~~g a a g c c t g c g g g c c g c 8

A I s~t t c c a~. s t g . a g c t c g t g c l g t t t t t c c t c g c t c l l g c g g l g t c t g g c g c c g t g g c .
7 2 1~~g t t t~t g t g~~a a g g a g g t g a~t t~a~~~g~~t g a a t g g a a~l g g t g g~ 9 0 l t g c c c c t g g t t g g~c g g c n~g c~c g t t g t t g t t t t t g g g~a g g t t g t t g c t g g a c a g g~g g t t 6 1 g t g a a a a t g a t g g a g g c n g a g g g c c g t t c c a a t g g t c a~c c c a~g t g t g c t g t t g~a a t t The I: gondii CyP-20 sequence was isolated as part of a large open reading frame with cDNA encoding 174 amino acids before the identified NHZ-terminal amino acid. There are 2 methionines encoded within this NH2-terminal sequence, one immediately proximal to the NHz-terminal residue determined by amino-terminal sequencing. The long open reading frame preceding the I: gondii CyP-20 sequence does not match any known protein or DNA sequence in the NCBI data base when searched using only the non-CyP coding region as the query sequence. Translation of the two alternate frames of this cDNA yields several stop codons in the second frame but only a single stop codon in the third frame. None of the peptide sequences derived from these translated cDNA frames (or the negative strand) matches any identified protein.
The most closely matched sequences for I: gondii CyP-20 are CyPs isolated from plants, primarily due to a 7-amino acid region commonly present in plant CyPs and mammalian CyP-40 (49,50) but absent in most other eukaryotic CyPs (Fig.  7). Close homology of the T gondii CyP-20 sequence is also noted to other parasite sequences.

DISCUSSION
Cyclophilins of T gondii-We have identified two major cyclosporin-binding proteins in the CsA-sensitive protozoan T . gondii. Biochemical characteristics and sequence homology clearly identify both of these proteins as CyPs. However, dif-A.

T gondir
CyP-20 T. gondu Human CyP-I8 E. g r a n u l w CyP S. japnicum Q P

I m k l v l f f l a l s v s g a v n E N A G V R K A Y M~ I D I ( G E H 1 / I I I L I I R E I I A
V I I T I l l / l A V l G E /~I I~S l I I I I I K l G V K C I I I I~I G~I I I I I I V I A I I D I . 1

T E ( I I I R~/ I I I~I T T E S W P T V Q A I E A L G G S ( G R P S K l A K l T D l G l L E
Human CYP-18

I D I I I I I I I L I I R I I D I I N I I I K I I S V I S K S I K V I E P I T I I R I I E I I B.
HumsnCaIcm~Bnubuait Both T gondii CyPs isolated in these experiments are abundant proteins and exhibit PPIase activity, which is inhibitable by nanomolar concentrations of CsA. Based on molecular weight differences and distinct elution characteristics, there initially appeared to be only two proteins. The similar molecular weight of the two major bands identified by SDS-PAGE analysis first led us to speculate that one of the two proteins might represent a post-translational modification of the other. However, we detected no glycosylation of either protein, despite the later identification of several N-glycosylation sites in the amino acid sequence of both proteins.

. . . I N A D K D G D G R I S F E E F C
Isoelectric focusing clearly indicated there were at least variant isoforms within the 20-kDa band and perhaps different proteins. Subsequent sequence analysis firmly established that at least 2 different genes encode l! gondii CyPs. The deduced amino acid sequences of the two T gondii CyPs isolated indicate that both are highly homologous to other eukaryotic CyPs. The central core regions have up to 70% sequence identity with HCyP-18 (Fig. 7A), including a tryptophan known to be essential for CsA binding (51)  . The other two parasitic CyPs previously isolated (19,20) are homologous to both T gondii CyPs, but the overall " identity is greatest with T gondii CyP-20.
Based on the cDNA sequence, T gondii CyP-18.5 may be initially synthesized with an amino-terminal signal sequence, which is subsequently cleaved. The presence of a single gene encoding two CyPs, one with a signal sequence, has been previously noted in Neurospora crassa, in which one gene encodes both the cytosolic and mitochondrial CyPs (16). The initial 15 amino acids identified in the T gondii 18.5-kDa CyP gene, if transcribed, could increase affinity for 8-ornithino-CsA enough to alter its elution characteristics. The minor bands seen after isoelectric focusing of the 20-kDa protein fraction eluted from the 8-ornithino-CsA column by pH 3.3 buffer (Fig. 2) may represent this precursor protein, a post-translationally modified form of the cloned 20-kDa CyP or an unidentified CyP. Specific peptide sequencing or production of antibodies against the individual T gondii CyPs separated by isoelectric focusing will be required to make definitive statements regarding the composition of the 20-kDa protein fraction eluted at pH 3.3. The cDNA encoding T gondii CyP-20 is part of a much larger open reading frame with 520 bases 5' to the presumed initiation methionine that immediately precedes a proline identified as the NHz-terminal residue by peptide sequencing. This extensive 5' region has no identity with previously reported cDNA or protein sequences. While this region could represent a cloning artifact, it might well encode a precursor protein or functionally related protein. Efforts to determine the identity and relation of this sequence to the T gondii CyPs are ongoing.
The homology of the amino-terminal portion of the T gondii CyP-20 protein to the carboxyl-terminal region of the B subunit of the mammalian Ca2+-calmodulin-dependent phosphatase calcineurin is remarkable, because the human CyP-CsA complex can inhibit the phosphatase activity of calcineurin (52,531. This mechanism of action has been proposed by several authors to be critical to the CsA-mediated events that interrupt T-cell signaling (11,(53)(54)(55)(56). I: gondii CyP-20 is identical to calcineurin in 11/20 amino acids of this region; eight of these 11 amino acids are also conserved in HCyP-18 (Fig. 7B). Studies of the three dimensional interaction of the CsA-CyP complex with the A and B subunits of calcineurin, calmodulin, and Ca2+ are ongoing (57,58). While there are no currently available data indicating that the region of homology between CyP and calcineurin B contributes to this association, further studies focused on this sequence are warranted. Subcellular Location and Function of the 2: gondii CyPs-An essential clue to the function of these proteins may lie in their subcellular location. The 20-kDa I: gondii CyP forms predominate in the 10,000 x g pellet extract, which would be expected to contain the pellicle and subcellular organelles including rhoptries, dense granules, mitochondria, and nuclei (59).' This suggests an organelle association of the cloned I: gondii CyP-20 species, perhaps in a modified form, or possibly the I: gondii Cyp-18.5 precursor with the signal sequence intact. However, all disruption schemes of coccidian parasites (French press, glass beads, Nz cavitation bomb) have been shown to cause some organelle disruption (59),2 and definitive statements regarding the purity of these cell extracts are difficult. We have been unsuccessful in identifying the distribution of these proteins within the parasite andlor host cell using immunofluorescence microscopy and the polyclonal antiserum prepared against both I: gondi CyPs. Precise definition of the subcellular location of the I: gondii CyPs will require extensive analysis using reagents or approaches that are beyond the scope of the current study.
PPIase Participation in Microbial Pathogenesis-The most well defined microbial PPIase is the legionella MIP (macrophage infectivity potentiator) protein, an outer membrane virulence protein with high affinity for FK-506 that assists intracellular survival of the organism (31, 32). Experimental evidence suggests that MIP is important in an early event after legionella entry into the host cell (32). Another intracellular bacterium with a MIP-like protein, chlamydia, has recently demonstrated behavior similar to MIP(-) legionella in cell cultures treated with high doses of FK-506 (30). The intracellular survival strategy of I: gondii is similar to that of legionella and chlamydia with parasite survival within a vacuole, rather than free in the cytosol. In addition, the response noted with CsA treatment of I: gondii-infected macrophages reported by Mc-Cabe et al. (35) mirrors the MIP(-) legionella and FK506treated chlamydia experiments. In all three circumstances, an early event after parasite invasion that inhibits survival, but not replication, appears to be affected (30,32,35). These preliminary investigations suggest PPIases might be involved in biochemical pathways essential for the intracellular survival of prokaryotic and eukaryotic parasites.
We speculate that there are multiple sites in the intracellular life cycle of I: gondii where PPIase activity could increase virulence. An essential chaperone function has been established for a previously isolated CyP homolog, the nina A protein of Drosophila (61)(62)(63). Chaperoning of proteins critical in parasite invasion, establishment of the parasitophorous vaculolar membrane, or evasion of host cell defenses could account for PPIase-enhanced virulence. In addition, immunophilins have been found as components of hormone-free steroid receptor publication.
Alternatively, the activity of key regulatory enzymes could be altered by PPIases. Host cell signaling is modified consequent to intracellular parasitism (70). There is mounting evidence that the Ca2+-calmodulin-dependent phosphatase, calcineurin, may be critical in modulating both the antimicrobial (60) and immunosuppressive effects of CsA (11,(53)(54)(55). No calcineurinlike phosphatase has been described in I: gondii to date, and the amino acid homology noted between I: gondii CyP-20 and the B subunit of mammalian calcineurin raises the possibility that microbial PPIases could interact with host cell proteins.
Additional investigation into the location and function of microbial immunophilins may provide new insights into the cell biology of signal transduction, protein-protein interactions, and the pathogenesis of intracellular infection.