IMP dehydrogenase from the protozoan parasite Toxoplasma gondii

Article abstract of DOI:10.1128/AAC.49.6.2172-2179.2005

The opportunistic apicomplexan parasite Toxoplasma gondii damages fetuses in utero and threatens immunocompromised individuals. The toxicity associated with standard antitoxoplasmal therapies, which target the folate pathway, underscores the importance of examining alternative pharmacological strategies. Parasitic protozoa cannot synthesize purines de novo; consequently, targeting purine salvage enzymes is a plausible pharmacological strategy. Several enzymes critical to purine metabolism have been studied in T. gondii, but IMP dehydrogenase (IMPDH), which catalyzes the conversion of IMP to XMP, has yet to be characterized. Thus, we have cloned the gene encoding this enzyme in T. gondii. Northern blot analysis shows that two IMPDH transcripts are present in T. gondii tachyzoites. The larger transcript contains an open reading frame of 1,656 nucleotides whose deduced protein sequence consists of 551 amino acids (TgIMPDH). The shorter transcript is an alternative splice product that generates a 371-amino-acid protein lacking the active-site flap (TgIMPDH-S). When TgIMPDH is expressed as a recombinant protein fused to a FLAG tag, the fusion protein localizes to the parasite cytoplasm. Immunoprecipitation with anti-FLAG was employed to purify recombinant TgIMPDH, which converts IMP to XMP as expected. Mycophenolic acid is an uncompetitive inhibitor relative to NAD+, with a intercept inhibition constant (K_ii) of 0.03 ± 0.004 μM. Tiazofurin and its seleno analog were not inhibitory to the purified enzyme, but adenine dinucleotide analogs such as TAD and the nonhydrolyzable β-methylene derivatives of TAD or SAD were inhibitory, with K_ii values 13- to 60-fold higher than that of mycophenolic acid. Copyright

Full text of DOI:10.1128/AAC.49.6.2172-2179.2005

ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, June 2005, p. 2172–2179
0066-4804/05/$08.00ϩ0 doi:10.1128/AAC.49.6.2172–2179.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Vol. 49, No. 6
IMP Dehydrogenase from the Protozoan Parasite Toxoplasma gondii
William J. Sullivan, Jr.,1 Stacy E. Dixon,1 Catherine Li,2 Boris Striepen,2 and Sherry F. Queener1*
Department of Pharmacology and Toxicology, Indiana University School of Medicine, Indianapolis, IN 46202,1 and
Center for Tropical and Emerging Global Diseases and Department of Cellular Biology,
University of Georgia, Athens, GA 306022
Received 17 November 2004/Returned for modification 26 January 2005/Accepted 8 February 2005
The opportunistic apicomplexan parasite Toxoplasma gondii damages fetuses in utero and threatens immu-
nocompromised individuals. The toxicity associated with standard antitoxoplasmal therapies, which target the
folate pathway, underscores the importance of examining alternative pharmacological strategies. Parasitic
protozoa cannot synthesize purines de novo; consequently, targeting purine salvage enzymes is a plausible
pharmacological strategy. Several enzymes critical to purine metabolism have been studied in T. gondii, but
IMP dehydrogenase (IMPDH), which catalyzes the conversion of IMP to XMP, has yet to be characterized.
Thus, we have cloned the gene encoding this enzyme in T. gondii. Northern blot analysis shows that two IMPDH
transcripts are present in T. gondii tachyzoites. The larger transcript contains an open reading frame of 1,656
nucleotides whose deduced protein sequence consists of 551 amino acids (TgIMPDH). The shorter transcript
is an alternative splice product that generates a 371-amino-acid protein lacking the active-site flap (TgIMPDH-
S). When TgIMPDH is expressed as a recombinant protein fused to a FLAG tag, the fusion protein localizes to
the parasite cytoplasm. Immunoprecipitation with anti-FLAG was employed to purify recombinant TgIMPDH,
which converts IMP to XMP as expected. Mycophenolic acid is an uncompetitive inhibitor relative to NAD؉, with
a intercept inhibition constant (Kii) of 0.03 ؎ 0.004 ␮M. Tiazofurin and its seleno analog were not inhibitory to the
purified enzyme, but adenine dinucleotide analogs such as TAD and the nonhydrolyzable ␤-methylene derivatives
of TAD or SAD were inhibitory, with Kii values 13- to 60-fold higher than that of mycophenolic acid.
IMP dehydrogenase (IMPDH) converts IMP to XMP in the
presence of NAD, a critical rate-limiting step in the biosynthe-
sis of guanine nucleotides. Increased IMPDH activity is asso-
ciated with actively dividing cells; consequently, IMPDH has
been exploited as a target for anticancer, antiviral, immuno-
suppressive, and antimicrobial drug therapies (9, 15). Signifi-
cant differences in IMDPH enzyme kinetics and inhibitor sen-
sitivities between human and microbial homologues suggest
that IMPDH may be an exploitable target for antimicrobials
(25, 36, 40, 42). Purine metabolic enzymes hold particular
promise as targets to combat intracellular parasites while min-
imizing side effects to host cells. Parasites are inherently more
reliant on these enzymes to maintain growth and are incapable
of synthesizing purine precursors de novo (34).
Toxoplasma gondii is an obligate intracellular protozoan re-
sponsible for fetal damage when the infection is acquired in
utero (24), heart transplant complications (37), and opportu-
nistic infections in immunocompromised individuals (39). T.
gondii, as a member of the phylum Apicomplexa, is related to
other parasites such as Plasmodium spp. (malaria) and Cryp-
tosporidium spp. (26). A wide variety of molecular genetic tools
have been developed for T. gondii (27, 28). In particular, the
generation of reagents for reverse genetics has facilitated the
study of the purine salvage network in this organism. Inser-
tional mutagenesis has led to the cloning of T. gondii hypoxan-
thine-xanthine-guanine-phosphoribosyltransferase (HXGPRT),
adenosine kinase (AK), and an adenosine transporter (TgAT)
(3, 6, 32). Both biochemical studies (23, 29) and the generation
of viable parasites lacking AK or HXGPRT demonstrate func-
tional redundancy in pathways leading to GMP (2, 6, 32).
Thus, drugs designed to inhibit only one route (via either
AK and IMPDH or HXGPRT) would be ineffective. Thus,
therapeutic exploitation of HXGPRT inhibitors depends
upon simultaneous inhibition of GMP synthesis via the AK
pathway. IMPDH is the logical target in the AK pathway both
because of its regulatory significance and because of the wealth
of inhibitors available. While IMPDHs have been cloned in
other apicomplexans such as Plasmodium and Cryptosporidium
spp. (14, 31), T. gondii IMPDH has not been available for
study.
Here we describe the IMPDH gene in T. gondii, which gives
rise to two transcripts, TgIMPDH and TgIMPDH-S. The latter
is generated by an alternative splicing event that removes the
active-site flap. Recombinant TgIMPDH was expressed and
purified from T. gondii parasites for kinetic analysis. This full-
length enzyme is inhibited by mycophenolic acid and adenine
dinucleotide analogs, with intercept inhibition constant (Kii)
values in the micromolar to submicromolar range. The identi-
fication of this important purine interconversion enzyme con-
stitutes a significant contribution to understanding purine sal-
vage in T. gondii.
MATERIALS AND METHODS
Parasite culture and reagents. T. gondii tachyzoites (RH and RH⌬HX strains)
were cultured in primary human foreskin fibroblasts (HFF) in Dulbecco modi-
fied Eagle medium supplemented with 1% fetal bovine serum and gentamicin.
Cultures were maintained in a humidified CO2 (5%) incubator at 37°C. Total
RNA was purified from filter-purified parasites using Trizol reagent (Invitrogen)
and mRNA using the Poly(A)Pure system (Ambion), both followed by 1 h of
incubation at 37°C with RQ1 DNase.
* Corresponding author. Mailing address: Department of Pharma-
cology and Toxicology, Medical Sciences Building Room A-519, Indi-
ana University School of Medicine, 635 Barnhill Drive, Indianapolis,
IN 46202-5120. Phone: (317) 274-1563. Fax: (317) 274-7714. E-mail:
queenes@iupui.edu.
2172

VOL. 49, 2005
T. GONDII IMPDH
2173
FIG. 1. T. gondii IMPDH gene. A. Schematic diagram of TgIMPDH locus. Gene-specific primers used for 5Ј- and 3Ј-RACE are designated by
arrows. Black boxes denote exons, white boxes denote introns, and gray boxes denote untranslated regions. B, BamHI; Bg, BglII; H, HindIII. B.
Southern blot analysis. Genomic DNA from T. gondii was digested with the indicated restriction enzymes and probed with labeled cDNA
representing TgIMPDH. Numbers at left are sizes in kilobases. C. Northern analysis. Full-length cDNA representing the entire ϳ1.6-kb coding
region of TgIMPDH was used to probe a Northern blot containing T. gondii mRNA (Tg) and host cell (HFF) total RNA.
RACE. 5Ј- and 3Ј rapid amplification of cDNA end (RACE) reactions were
performed with total RNA using GeneRacer (Invitrogen) and the following
gene-specific primers, listed as “parent” and “nesting” oligonucleotides, respec-
tively: for 3Ј-RACE, 5Ј-AGAAAGGGAAGCTCCCGATTGTGAACG and 5Ј-
AGAGTTTCCACTCGCCTCGAAGGACTCC; for 5Ј-RACE, 5Ј-AGTCCTTC
GAGGCGAGTGGAAACTCTCG and 5Ј-TCTTTGACTCCCGCAGCAGTTC
GTTTGCC (Fig. 1A). Both 5Ј- and 3Ј-RACE PCRs utilized cDNA generated
with the oligo(dT) primer included with the kit.
Bioinformatics and DNA sequencing. Preliminary genomic sequence data
were accessed via http://ToxoDB.org (release 1.0) (21). Genomic data were
provided by The Institute for Genomic Research (supported by the National
Institutes of Health [NIH] grant no. AI05093) and by the Sanger Center (Well-
come Trust). BLAST analyses were run at http://www.ncbi.nih.gov/BLAST/, and
multiple sequence alignments were produced using CLUSTALW (33). DNA
sequencing was performed at the Biochemistry Biotechnology Facility (Indiana
University School of Medicine, Indianapolis, IN).
Expression and purification of recombinant TgIMPDH. TgIMPDH was am-
plified from T. gondii total RNA using the SuperScript One-Step reverse tran-
scription-PCR kit (Invitrogen) and the following primers, which generate restric-
tion sites NdeI and AvrII for installation into a T. gondii expression vector that
uses the TUB promoter and HXGPRT selection system (6): 5Ј-CATATGAAA
ATGGCTGACGGATGGGATGCGGAGAAAATC (sense) and 5Ј-CCTAGG
GGCGTAGAGCTTTCTCTCGAAGCTGTGC (antisense). The TgIMPDH in-
sert was fused in frame to a C-terminal FLAG (Sigma) tag, creating the vector
designated ptubTgIMPDHFLAG::HX. The nucleotide sequence of both strands
of the insert was determined to confirm the fidelity of the PCR. Following
electroporation, tachyzoites were selected in 25 ␮g/ml mycophenolic acid plus 50
␮g/ml xanthine and cloned by limiting dilution (28). Clones were examined by
immunofluorescence assay with anti-FLAG (Sigma) to select for those express-
ing recombinant tagged protein. For immunoprecipitation, ϳ109 freshly lysed
tachyzoites were filtered-purified and harvested by centrifugation. Parasites were
lysed by sonication in 1.0 ml of 50 mM Tris HCl (pH 7.4), 150 mM NaCl, 1 mM
EDTA, and 1% Triton X-100 supplemented with a protease inhibitor cocktail
(Sigma). The parasite lysate was clarified by centrifugation at 8,200 ϫ g at 4°C for
10 min prior to mixing with 50 ␮l equilibrated anti-FLAG M2-agarose affinity
gel. The affinity gel and lysate were mixed overnight at 4°C and spun at 8,200 ϫ
g. The supernatant (unbound fraction) was removed and stored for later analysis.
The pelleted resin was washed three times in the lysis buffer above without the
Triton X. 3ϫ FLAG peptide was used to elute the material bound to the
anti-FLAG resin for further analysis.
Immunofluorescence assays. Parasite clones expressing recombinant
TgIMPDHFLAG were inoculated onto HFF monolayers grown on glass cover-
slips in 12-well tissue culture plates. Infected cell monolayers were fixed in
methanol for 10 min at Ϫ20°C. After blocking in phosphate-buffered saline
(PBS) containing 1% fish gelatin, 3% bovine serum albumin, and 5% goat serum
for 1 h, mouse polyclonal anti-FLAG (Sigma) was applied at 1:250 (in PBS with
3% bovine serum albumin) for 1 h, followed by goat anti-rabbit Alexa 488 at
1:1,000 after PBS washes. 4Ј,6Ј-Diamidino-2-phenylindole (DAPI) nuclear stain
was added for 5 min prior to visualization with a Leica DMLB microscope.
Images were captured using a Spot RTSE model 12.0 monochrome camera and
Spot Diagnostic Software version 3.5.9.
Enzymatic assays. IMPDH activity was measured using a fixed-time assay
followed by thin-layer chromatography to separate radiolabeled substrate and
products (12). The 50-␮l reaction mixture contained 100 mM Tris-HCl buffer
(pH 8.0), 100 mM KCl, 1 mM dithiothreitol, 3 mM EDTA, 2.3 mM NAD, 80 ␮M
IMP, and 10 to 180 ng of purified protein. At the end of the 60-min incubation
at 37°C, the reaction was quenched by adding 25 ␮l of carrier solution containing
IMP, XMP, GMP, and hypoxanthine, each at a concentration of 4 mM. Ten-
microliter samples of the reactions were spotted onto polyethyleneimine-cellu-
lose plates that were prewashed with 100% methanol. The plates were developed
in 0.5 M LiCl for 2 h. Rf values for the standards were as follows: XMP, 0.11;
GMP, 0.32; IMP, 0.5; hypoxanthine, 0.86. Activity is expressed as concentration
of product formed per minute, calculated from specific activity of the radiola-
beled substrate. The assay is linear with protein concentration over the range
used for all assays and has low variability; standard errors of the means typically
averaged 3.2% of the means within assays and 4.7% from day to day.

2174
SULLIVAN ET AL.
ANTIMICROB. AGENTS CHEMOTHER.
FIG. 2. IMPDH amino acid sequence alignments. Protein sequence alignments from various IMPDH homologues were compiled with
CLUSTALW with manual adjustment. Tg, Toxoplasma gondii; Pf, Plasmodium falciparum; Tb, Trypanosoma brucei; Ld, Leishmania donovani; Pc,
Pneumocystis carinii; Sc, Saccharomyces cerevisiae; H1 and H2, Homo sapiens isoforms 1 and 2, respectively. Indicated are the CBS domains as well
as the active-site loop and flap. The atypical serine-rich insertion in TgIMPDH is shown in italics. Asterisks denote residues that are identical across
these species. Colons represent conserved amino acid substitutions, while periods represent less conserved residue substitutions.
Curve fitting for kinetic experiments was done with Prism 3.0, using the
standard equations for linear regressions and for the hyperbolic single-site curve.
RESULTS AND DISCUSSION
Ultimately, all experiments with various NADϩ concentrations were evaluated
with the equation entered for substrate inhibition, because NADϩ at high con-
centrations inhibited the enzyme reaction: v ϭ Vmax[NADϩ]/(Km ϩ [NADϩ] ϩ
[NADϩ]2/Ki), where Vmax is the maximal velocity of the reaction, Km is the
Michaelis constant, and Ki is the inhibition constant for NADϩ. Mycophenolic
acid inhibition followed uncompetitive inhibition, according to the following
equation: v ϭ Vmax[NADϩ]/(Kapp ϩ [NADϩ][1 ϩ [I]/Kii), where Kapp is the ap-
parent Michaelis constant for NADϩ and Kii is the intercept inhibition constant.
The equation for uncompetitive inhibition was accepted when the results of four
Identification of an IMPDH homologue in T. gondii. The
T. gondii database was mined for sequences encoding IMPDH
using the BLAST algorithm and text queries. Two contigs
(Tgg_18808 and -18809) showing a high degree of similarity to
nonoverlapping regions of human IMPDH were identified.
Based on the sequence data, primers were designed to obtain
the full-length cDNA employing 5Ј- and 3Ј-RACE. Using the
primers listed in Materials and Methods, an 822-bp band rep-
resenting the 5Ј end of the gene and a 1,532-bp band repre-
senting the 3Ј end were amplified and subcloned for sequenc-
ing. Considered with the genomic data, the predicted open
reading frame is 1,656 bp, possessing 5Ј and 3Ј untranslated
regions of 292 and 518 bp, respectively. The putative start
codon fits Kozak consensus rules (22). There is no in-frame
stop codon upstream of this proposed start site, but no other
experiments showed that there was no statistical difference in the ratio of Km
/
Vmax in the uninhibited and the inhibited state. The equation for mixed non-
competitive inhibition was used to characterize inhibition by tiazofurin and its
analogs. The equation is v ϭ Vmax [NADϩ]/[Km] (1 ϩ [I]/Kis) ϩ [NADϩ](1 ϩ
[I]/Kii), where Kis is the slope inhibition constant and Kii is the intercept inhibi-
tion constant.
Nucleotide sequence accession number. Nucleotide sequence data reported in
this paper are available in the GenBank, EMBL, and DDBJ databases under the
accession numbers AY661469 (TgIMPDH) and AY663109 (TgIMPDH-S).

VOL. 49, 2005
T. GONDII IMPDH
2175
FIG. 3. Localization of recombinant TgIMPDH. Recombinant TgIMPDH containing a C-terminal FLAG-tag fusion localizes to the parasite
cytoplasm. Parasites were costained with DAPI, which stains the nuclei of parasites (pN) and host cells (hN).
translation initiation sites exist downstream of the transcrip-
tion start site determined by 5Ј-RACE.
structural feature of the molecule, the large barrel core. The
barrel core is interrupted by three elements: a flanking region,
an active-site loop, and an active-site flap. The flanking region
has been noted in other species to interrupt the barrel core of
the protein but does not interact with either substrate or co-
factor in any crystal structures currently available. The flanking
region in all IMPDH proteins studied contains two CBS (cys-
tathionine beta-synthase) motifs; for TgIMPDH these motifs
involve amino acid residues 105 to 156 and residues 168 to 216,
making their position and length nearly identical to those of
the CBS motifs in human type II IMPDH. The active-site loop
Additional genomic sequences were procured from ToxoDB,
allowing the construction of a hypothetical locus containing
eight introns. A schematic diagram of the TgIMPDH genomic
locus is shown in Fig. 1A, along with a Southern analysis con-
sistent with a single-copy locus (Fig. 1B).
TgIMPDH cDNA hybridizes to two transcripts. Northern
blot analysis used the full-length cDNA for TgIMPDH as a
probe. Figure 1C shows hybridization to a ϳ2.5-kb band, con-
sistent with our cloning results. However, an additional band at
ϳ2.0 kb also hybridizes to the probe, suggesting another iso-
form of IMPDH is present in T. gondii. Since the ToxoDB and
Southern analyses are consistent with the TgIMPDH locus
being present as a single copy, we postulated that an alterna-
tively spliced transcript may be present. Primers designed to
amplify the IMPDH coding sequence from T. gondii total RNA
produce the expected 1.6-kb band as well as a fainter 1.2-kb
band. Each band was gel purified and sequenced. The 1.6-kb
product agrees with our original open reading frame sequence
prediction; the 1.2-kb product (referred to as TgIMPDH-S, for
short form) lacks an internal 506-nucleotide region. The length
of TgIMPDH-S including the untranslated regions is 2.0 kb, in
agreement with the Northern analysis. TgIMPDH-S encodes a
predicted 371-amino-acid form of IMPDH that lacks the ac-
tive-site flap. The question of whether this shortened version of
TgIMPDH possesses catalytic activity or perhaps serves some
other regulatory function in the organism is intriguing. How-
ever, for these characterization studies we chose to use the
longer form, which has more complete homology to other
catalytically active IMPDH enzymes.
Comparison of TgIMPDH to other species. The deduced
amino acid sequence of TgIMPDH is aligned, with salient
domains identified, for comparison to IMPDH sequences
from other species (Fig. 2). BLASTp analysis reveals that
TgIMPDH is most similar to mammalian IMPDH (e.g., 53%
identity to the human type II enzyme), followed by Xenopus
laevis and the apicomplexan Plasmodium yoelii (50% identity).
The IMPDH orthologue cloned from fellow apicomplexan
Cryptosporidium parvum is only 35% identical, consistent with
the observation that CpIMPDH was acquired through lateral
gene transfer from bacteria (31).
FIG. 4. Expression and purification of recombinant TgIMPDH.
Transgenic parasite clones stably expressing recombinant TgIMPDHFLAG
were selected in mycophenolic acid and xanthine following the elec-
troporation of plasmid into RH⌬HX parasites. TgIMPDH was puri-
fied by FLAG-based affinity chromatography and eluted from the resin
with 3ϫ FLAG peptide. Lysate (L), the unbound fraction (U), and the
eluate (E) were resolved by sodium dodecyl sulfate-polyacrylamide gel
electrophoresis, and proteins were visualized by SimplyBlue stain. Mo-
lecular mass markers are shown in kilodaltons.
The deduced secondary structure of TgIMPDH is similar to
those of other known enzymes (11). The N-terminal and C-
terminal regions are short elements at either end of the main

2176
SULLIVAN ET AL.
ANTIMICROB. AGENTS CHEMOTHER.
FIG. 5. Mycophenolic acid inhibition of recombinant TgIMPDH. The dependence of the enzymatic reaction on NADϩ was assessed for
purified recombinant TgIMPDHFLAG with and without a fixed concentration of mycophenolic acid (MPA). Results from four independent ex-
periments were combined; means with error bars representing the standard errors of the means are shown. Data were fitted to the equation describing
substrate inhibition, since high concentrations of NADϩ were inhibitory (see Materials and Methods). Nonlinear regression (Prism 3.0) was used
to calculate Vmax and Km values for both curves. The Vmax values (means Ϯ standard errors of the means) are 0.570 Ϯ 0.16 and 0.168 Ϯ 0.04
␮M/min for the top and bottom curves, respectively; the Km values are 158.1 Ϯ 86.7 and 46.7 Ϯ 25.7 ␮M for the top and bottom curves, respectively.
of the T. gondii enzyme is highly conserved with other species
but not identical, as discussed below. The most striking differ-
ence in structure is in the active-site flap, a flexible element
that closes over the active site when cofactor and substrate are
bound. TgIMPDH contains a unique 50-amino-acid insertion
(residues 410 to 460, T. gondii numbering) that is rich in serine
and proline (Fig. 2). This unique insertion is in the disordered
portion of the flap region. The physiological relevance of this
insertion in the T. gondii enzyme remains to be determined.
Data from published crystal structures of human type II,
hamster, Tritrichomonas foetus, and Streptococcus pyogenes
IMPDHs allow a more detailed comparison of the potential
critical differences between the sequence of TgIMPDH and
other forms. As expected, key catalytic residues such as Cys331
(human enzyme numbering), which carries out the nucleo-
philic attack on the C-2 position of the inosine ring, are con-
served in both T. gondii and Plasmodium falciparum IMPDH.
Other residues in the active-site loop (amino acids 325 to 340
for the human enzyme) such as Gly326, Thr333, and Ser329,
which interact with substrate or cofactor in existing crystal
structures (10, 30, 41), are also conserved in the apicomplexan
enzymes.
the human enzyme is replaced by a valine in the T. gondii
enzyme and an aspartic acid in the P. falciparum enzyme. The
impact of these substitutions needs to be explored.
The IMP binding site involves residues outside the active-
site loop, as shown by crystal structures for several enzymes
(11), and these residues are generally conserved in TgIMPDH.
In the enzymes studied by crystal structure, Gly366, Gly387,
Ser388, and Tyr411 stabilize the phosphate group of IMP into
the active site. These residues are conserved in TgIMPDH. The
ribose of IMP interacts with Ser68 and Asp364 of the human
enzyme. These residues are conserved in the apicomplexan
enzymes. Likewise, residues Met414, Gly415, and Gln441, which
interact with the inosine ring in the human enzyme, are con-
served in the apicomplexan enzymes.
The NAD binding cleft has been probed with crystal struc-
tures for hamster, human, and T. foetus IMPDH (11). Polar
interactions of the carboxamide group of nicotinamide exist
with Asn303, Arg322, and Asp274, residues that are conserved
in the apicomplexan enzymes. Mycophenolic acid interacts
with these same residues. Mycophenolic acid has additional
interactions in the human enzyme with Gly326, Thr333, and
Gln441, which are conserved in the apicomplexan forms. The
ribose phosphate portion of NAD interacts with Asp274,
Ser275, and Ser276; these three amino acids are conserved in
all species so far studied.
Two residues within the active-site loop differ between the
human and the T. gondii enzyme. The conserved isoleucine
present at position 332 of the human enzymes is replaced by a
threonine residue in both the T. gondii and P. falciparum en-
zymes. Likewise, a conserved glutamic acid at position 335 in
Residues from the adjacent monomers in the active tetramer
interact with the ribose of NAD. Gln469, which is conserved

VOL. 49, 2005
T. GONDII IMPDH
TABLE 1. Kinetics of TgIMPDH-F compared to those of the enzyme from other organismsa
2177
Protozoa
Apicomplexa
Human
Type IIj
Bacteria
Fungi
Kinetic parameter
(P. carinii)h
Other
(T. foetus)e
T. gondiib
C. parvumc
E. tenellad
Type Ii
E. colik
B. burgdorferim
kcat (sϪ1
)
0.98 Ϯ 0.05
3.3 Ϯ 0.2
1.9 Ϯ 0.2
150 Ϯ 30
0.56 Ϯ 0.04 1.82 Ϯ 0.04 0.39 Ϯ 0.01
1.4 Ϯ 0.04i
13 Ϯ 1
2.6 Ϯ 0.3
Km of NAD (␮M)
144 Ϯ 12
150 Ϯ 20
150
2.0
24.4 Ϯ 1.3
42 Ϯ 6
6.0 Ϯ 1.0
2,000 Ϯ 500 1,100 Ϯ 160
32 Ϯ 3.6i
Ki of NAD (mM)
kcat/Km (mMϪ1 ⅐ sϪ1
Kii of MPAp (␮M)
4.5 Ϯ 1.2
6.8 Ϯ 4
0.030 Ϯ 0.004
2.9 Ϯ 0.7
15 Ϯ 3
9.3 Ϯ 0.3
6.8 Ϯ 1.8
13
2.6 Ϯ 0.07
23 Ϯ 6
0.06 Ϯ 0.003
0.59 Ϯ 0.02
65 Ϯ 0.5
2.8 Ϯ 0.8
6.5
2.3 Ϯ 0.4
2.4 Ϯ 0.5
6 Ϯ 1
)
43
0.011
0.18
9.0 Ϯ 0.5f
14.0 Ϯ 2.0g
69,000 Ϯ 9,000g
0.006
20f
0.022 Ϯ 0.008f
1,300 Ϯ 100f
Kii of tiazofurin (␮M)
Kii of Se-tiazofurin (␮M)
Kii of TAD (␮M)
Kii of SAD (␮M)
Kii of CH2-TAD (␮M)
Ͼ100
Ͼ100
1.77 Ϯ 0.73
1,500 Ϯ 100
1.6 Ϯ 0.2g
2.3 Ϯ 0.4f
0.71n
0.06o
0.095
0.43n
0.03o
8.5l
0.98 Ϯ 0.45
0.38 Ϯ 0.05
0.6 Ϯ 0.04
0.06 Ϯ 0.02f
0.145i
1 Ϯ 1
Kii of CH2-SAD (␮M)
a All data within the columns come from the reference denoted in the column header, unless otherwise specified by a footnote to the value. The data for the T. gondii
enzyme are means Ϯ standard errors of the means calculated from n independent experiments. n ϭ 3 for kcat, Km of NAD, and Kii of TAD; 10 for Ki of NAD; 8 for
Kii of mycophenolic acid; and 2 for Kii of CH2-TAD and CH2-SAD.
b Assays at 37°C using the thin-layer chromatography assay described in Materials and Methods: 460 ␮M NAD, 80 ␮M IMP.
c Reference 35. Assays at 25°C. Spectrophotometric assay: 500 ␮M NAD, 250 ␮M IMP.
d Reference 17. Assays at 37°C. Thin-layer chromatography assay: 500 ␮M NAD, 30 ␮M IMP.
e Reference 5. Assays at 25°C. Spectrophotometric assay: 1,000 ␮M NAD, 100 ␮M IMP.
f Reference 4. Assays at 25°C. Spectrophotometric assay: 400 ␮M NAD, 150 ␮M IMP.
g Reference 16. Assays at 25°C. Spectrophotometric assay: 1,000 ␮M NAD, 200 ␮M IMP.
h S.F. Queener and D.J. Ye, unpublished. Assays at 25°C. Spectrophotometric assay: 250 ␮M NAD, 180 ␮M IMP.
i Reference 13. Assays at 37°C. Spectrophotometric assay: not defined.
j Reference 38. Assays at 25°C. Spectrophotometric assay: 40 to 800 ␮M NAD, 250 ␮M IMP.
k Reference 19. Assays at 25°C. Spectrophotometric assay: 2,500 to 8,000 ␮M NAD, 750 to 1,000 ␮M IMP.
l
Reference 20. Assays at 25°C. Spectrophotometric assay: 2,500 ␮M NAD, 1,000 ␮M IMP.
m Reference 42. Assays at 25°C. Spectrophotometric assay: 500 ␮M NAD, 250 ␮M IMP.
n Reference 8. Assays at 37°C. Spectrophotometric assay: 100 ␮M NAD, 100 ␮M IMP.
o Reference 18. Assays at 37°C. Thin-layer chromatography assay: 1,000 ␮M NAD, 286 ␮M IMP.
p
MPA, mycophenolic acid.
between human and apicomplexan IMPDH, is one of these
residues. Another is Ala46 in the human enzyme. Considerable
variability exists at this site; the T. gondii and Pneumocystis
carinii enzymes contain a valine, but the P. falciparum enzyme
contains a leucine.
The highest degree of species variability in the NAD binding
site occurs in the portion that interacts with the adenine ring.
In the human enzyme Phe282 forms part of this interaction,
but this residue is replaced by tyrosine in several species, in-
cluding the apicomplexan enzymes. Another interaction in the
human enzyme is with His253, which is replaced by a lysine in
the T. gondii enzyme and by an arginine in the P. falciparum
enzyme. Thr45 on an adjacent monomer contributes to binding
the adenine ring in the human enzyme, but this residue is a
glycine in TgIMPDH and an alanine in the P. falciparum en-
zyme.
Another important contributor to the enzymatic reaction is
a bound water molecule and a bound potassium ion (1, 11).
Binding involves the conserved Cys331, Gly328, and Gly326
residues in the active-site loop, as well as three residues near
the C terminus: Glu500, Gly501, and Gly502. Both Glu500 and
Gly501 are conserved between human and apicomplexan en-
zymes, but Gly502 is replaced by an aspartic acid in the T.
gondii enzyme and a lysine in the P. falciparum enzyme. Since
the water molecule and the potassium ion are known to play
key catalytic roles in other IMPDH forms, the effects of these
substitutions on activity should be explored.
Localization and enzymatic analysis of recombinant
TgIMPDH. Numerous attempts to express functional recom-
binant TgIMPDH in a variety of bacterial systems failed, most
likely due to lack of solubility of the recombinant protein.
Therefore, we attempted expression in T. gondii itself and used
a C-terminal FLAG epitope tag to purify TgIMPDH for anal-
ysis (7). The FLAG tag also allowed us to determine the sub-
cellular localization of the recombinant enzyme. As expected,
transgenic parasites transfected with ptubTgIMPDHFLAG::HX
(see Materials and Methods) exhibited a staining pattern sug-
gesting that TgIMPDH is diffuse throughout the parasite cy-
toplasm (Fig. 3). Untransfected RH⌬HX parasites do not ex-
hibit fluorescence when analyzed in the same manner with
anti-FLAG (data not shown). An anti-FLAG affinity resin was
used to purify recombinant enzyme from a stably transfected
parasite clone. Purified enzyme from the transgenic clone
(RH⌬HX::TgIMPDHFLAG) was examined by staining sodium
dodecyl sulfate-polyacrylamide gels with a Coomassie blue-
based reagent. Parasites transfected with ptubTgIMPDHFLAG::
HX displayed a single band at the expected size (ϳ60 kDa)
following elution from the affinity matrix (Fig. 4).
To test enzymatic activity of TgIMPDH, the conversion of
radiolabeled IMP to XMP was measured directly by thin-layer
chromatography. Lysates from parental RH⌬HX T. gondii
served as negative controls to ensure that no endogenous
IMPDH activity could be detected following FLAG purifica-
tion. Protein-dependent conversion of IMP to XMP could

2178
SULLIVAN ET AL.
ANTIMICROB. AGENTS CHEMOTHER.
of IMPDH are significantly less sensitive to mycophenolic acid
than is TgIMPDHFLAG
.
Compounds known to inhibit mammalian forms of IMPDH
were tested for their ability to inhibit TgIMPDHFLAG (Fig. 6).
Like other IMPDH enzymes, TgIMPDHFLAG is poorly inhib-
ited by tiazofurin or Se-tiazofurin (Table 1). The stronger
inhibition by TAD and the nonhydrolyzable beta-methylene
derivatives of TAD (CH2-TAD) and SAD (CH2-SAD) fol-
lowed a mixed noncompetitive pattern with slightly more ef-
fect on Vmax than Km. The Ki values calculated from the equa-
tion for mixed noncompetitive inhibition (Table 1) show that
TgIMPDHFLAG is most potently inhibited by CH2-SAD, fol-
lowed by CH2-TAD and TAD. SAD was previously found to
be a more potent inhibitor of both type I and type II human
IMPDH than was TAD (8).
ACKNOWLEDGMENTS
We thank Pamela Torkelson for developing and performing the
IMPDH assays, Micah M. Bhatti for the generation and immunoflu-
orescence analysis of parasites expressing TgIMPDHFLAG, and Jeramy
Spurgeon for technical assistance with graphics. Preliminary genomic
sequence data were accessed via http://ToxoDB.org (release 1.0).
Genomic data were provided by The Institute for Genomic Re-
search (supported by NIH grant AI05093) and by the Sanger Center
(Wellcome Trust). This work was funded by NIH AI-42024 (S.F.Q.)
and NIH AI48475 and AI55268 and by further support from Merck
Research Laboratories (B.S.).
FIG. 6. Analogues of tiazofurin. All compounds were obtained
from the Drug Development and Clinical Sciences Branch of the Na-
tional Institute of Allergy and Infectious Diseases (Mohamed Nasr). A
commercial preparation of tiazofurin was also tested. Tiazofurin is
2-␤-D-ribofuranosylthiazole-4-carboxamide; Se-tiazofurin is 2-␤-D-ri-
bofuranosylselenazole-4-carboxamide; TAD is thiazole-4-carboxamide
adenine dinucleotide; ␤-methylene TAD is 4-carboxamido-2-␤-D-ribo-
furanosylthiazolyl adenosine methylenediphosphonic acid; ␤-methyl-
ene SAD is 4-carboxamido-2-␤-D-ribofuranosylselenazolyl (5Ј ϩ 5Ј)
adenosine methylenediphosphonic acid.
REFERENCES
1. Boritzki, T. J., D. A. Berry, J. A. Besserer, P. D. Cook, D. W. Fry, W. R.
Leopold, and R. C. Jackson. 1985. Biochemical and antitumor activity of
tiazofurin and its selenium analog (2-beta-D-ribofuranosyl-4-selenazolecar-
boxamide). Biochem. Pharmacol. 34:1109–1114.
2. Chaudhary, K., J. A. Darling, L. M. Fohl, W. J. Sullivan, Jr., R. G. Donald,
E. R. Pfefferkorn, B. Ullman, and D. S. Roos. 2004. Purine salvage pathways
in the apicomplexan parasite Toxoplasma gondii. J. Biol. Chem. 279:31221–
31227.
3. Chiang, C. W., N. Carter, W. J. Sullivan, Jr., R. G. Donald, D. S. Roos, F. N.
Naguib, M. H. el Kouni, B. Ullman, and C. M. Wilson. 1999. The adenosine
transporter of Toxoplasma gondii. Identification by insertional mutagenesis,
cloning, and recombinant expression. J. Biol. Chem. 274:35255–35261.
4. Digits, J. A., and L. Hedstrom. 2000. Drug selectivity is determined by
coupling across the NADϩ site of IMP dehydrogenase. Biochemistry 39:
1771–1777.
5. Digits, J. A., and L. Hedstrom. 1999. Kinetic mechanism of Tritrichomonas
foetus inosine 5Ј-monophosphate dehydrogenase. Biochemistry 38:2295–
2306.
be achieved using TgIMPDHFLAG protein purified from the
RH⌬HX::TgIMPDHFLAG clone; assays using the eluted pro-
tein from untransfected RH⌬HX parasites showed Ͻ0.2% of
rates seen with recombinant protein. We therefore set out to
characterize the catalytically active recombinant protein in
terms of NADϩ kinetics and sensitivity to an array of inhibi-
tors.
6. Donald, R. G., D. Carter, B. Ullman, and D. S. Roos. 1996. Insertional
tagging, cloning, and expression of the Toxoplasma gondii hypoxanthine-
TgIMPDHFLAG is sensitive to mycophenolic acid, a known
inhibitor of other forms of IMPDH. Increasing amounts of my-
cophenolic acid in the assay produced classic sigmoidal inhibi-
tion curves (data not shown). Experiments assessing the effect
of fixed concentrations of mycophenolic acid on NADϩ de-
pendence curves showed that inhibition by mycophenolic acid
is uncompetitive (Fig. 5); as expected for uncompetitive inhi-
bition, the ratio of the uninhibited to inhibited Km value was
equal to the ratio of the uninhibited to the inhibited Vmax
value, 3.385 and 3.393, respectively. These experiments al-
lowed the calculation of a Ki value of 0.030 Ϯ 0.004 ␮M (Ta-
ble 1). TgIMPDHFLAG is therefore similar in sensitivity to
IMPDH from another apicomplexan, Eimeria tenella, which
has a Ki of 0.18 ␮M (17). Cryptosporidium IMPDH was ac-
quired through lateral transfer from eubacteria, and predict-
ably the sensitivity to mycophenolic acid differs from that of
TgIMPDHFLAG (35). Both human forms of IMPDH and full-
length IMPDH from Pneumocystis have Ki values in the sub-
micromolar range, similar to TgIMPDHFLAG. Bacterial forms
xanthine-guanine phosphoribosyltransferase gene. Use as
a selectable
marker for stable transformation. J. Biol. Chem. 271:14010–14019.
7. Donald, R. G., and P. A. Liberator. 2002. Molecular characterization of a
coccidian parasite cGMP dependent protein kinase. Mol. Biochem. Parasi-
tol. 120:165–175.
8. Franchetti, P., L. Cappellacci, P. Perlini, H. N. Jayaram, A. Butler, B. P.
Schneider, F. R. Collart, E. Huberman, and M. Grifantini. 1998. Isosteric
analogues of nicotinamide adenine dinucleotide derived from furanfurin,
thiophenfurin, and selenophenfurin as mammalian inosine monophosphate
dehydrogenase (type I and II) inhibitors. J. Med. Chem. 41:1702–1707.
9. Franchetti, P., and M. Grifantini. 1999. Nucleoside and non-nucleoside IMP
dehydrogenase inhibitors as antitumor and antiviral agents. Curr. Med.
Chem. 6:599–614.
10. Gan, L., G. A. Petsko, and L. Hedstrom. 2002. Crystal structure of a ternary
complex of Tritrichomonas foetus inosine 5Ј-monophosphate dehydrogenase:
NADϩ orients the active site loop for catalysis. Biochemistry 41:13309–
13317.
11. Goldstein, B., D. Risal, and M. Strickler. 2003. IMPDH structure and ligand
binding. ACS Symp. Ser. 839:140–168.
12. Gu, J. J., A. K. Tolin, J. Jain, H. Huang, L. Santiago, and B. S. Mitchell.
2003. Targeted disruption of the inosine 5Ј-monophosphate dehydrogenase
type I gene in mice. Mol. Cell. Biol. 23:6702–6712.
13. Hager, P. W., F. R. Collart, E. Huberman, and B. S. Mitchell. 1995. Recom-
binant human inosine monophosphate dehydrogenase type I and type II

VOL. 49, 2005
T. GONDII IMPDH
2179
proteins. Purification and characterization of inhibitor binding. Biochem.
Pharmacol. 49:1323–1329.
synthesis and salvage in mutant host cells and parasites. Exp. Parasitol. 53:
77–86.
14. Hariharan, J., R. Rane, K. Ayyanathan, Philomena, V. P. Kumar, D.
Prahlad, and S. Datta. 1999. Mechanism-based inhibitors: development of a
high throughput coupled enzyme assay to screen for novel antimalarials.
J. Biomol. Screen. 4:187–192.
15. Hedstrom, L. 1999. IMP dehydrogenase: mechanism of action and inhibi-
tion. Curr. Med. Chem. 6:545–560.
16. Hedstrom, L., and C. C. Wang. 1990. Mycophenolic acid and thiazole ade-
nine dinucleotide inhibition of Tritrichomonas foetus inosine 5Ј-monophos-
phate dehydrogenase: implications on enzyme mechanism. Biochemistry 29:
849–854.
17. Hupe, D. J., B. A. Azzolina, and N. D. Behrens. 1986. IMP dehydrogenase
from the intracellular parasitic protozoan Eimeria tenella and its inhibition by
mycophenolic acid. J. Biol. Chem. 261:8363–8369.
18. Jayaram, H. N., J. A. Yalowitz, G. Krupitza, T. Szekeres, K. Krohn, and
K. W. Pankiewicz. 2003. Studies with benzamide riboside, a recent inhibitor
of inosine 5Ј-monophosphate dehydrogenase. ACS Symp. Ser. 839:231–246.
19. Kerr, K. M., J. A. Digits, N. Kuperwasser, and L. Hedstrom. 2000. Asp338
controls hydride transfer in Escherichia coli IMP dehydrogenase. Biochem-
istry 39:9804–9810.
30. Sintchak, M. D., M. A. Fleming, O. Futer, S. A. Raybuck, S. P. Chambers,
P. R. Caron, M. A. Murcko, and K. P. Wilson. 1996. Structure and mecha-
nism of inosine monophosphate dehydrogenase in complex with the immu-
nosuppressant mycophenolic acid. Cell 85:921–930.
31. Striepen, B., M. W. White, C. Li, M. N. Guerini, S. B. Malik, J. M. Logsdon,
Jr., C. Liu, and M. S. Abrahamsen. 2002. Genetic complementation in
apicomplexan parasites. Proc. Natl. Acad. Sci. USA 99:6304–6309.
32. Sullivan, W. J., Jr., C. W. Chiang, C. M. Wilson, F. N. Naguib, M. H. el
Kouni, R. G. Donald, and D. S. Roos. 1999. Insertional tagging of at least two
loci associated with resistance to adenine arabinoside in Toxoplasma gondii,
and cloning of the adenosine kinase locus. Mol. Biochem. Parasitol. 103:
1–14.
33. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W:
improving the sensitivity of progressive multiple sequence alignment through
sequence weighting, position-specific gap penalties and weight matrix choice.
Nucleic Acids Res. 22:4673–4680.
34. Ullman, B., and D. Carter. 1995. Hypoxanthine-guanine phosphoribosyl-
transferase as a therapeutic target in protozoal infections. Infect. Agents Dis.
4:29–40.
20. Kerr, K. M., and L. Hedstrom. 1997. The roles of conserved carboxylate
residues in IMP dehydrogenase and identification of a transition state ana-
log. Biochemistry 36:13365–13373.
21. Kissinger, J. C., B. Gajria, L. Li, I. T. Paulsen, and D. S. Roos. 2003.
ToxoDB: accessing the Toxoplasma gondii genome. Nucleic Acids Res. 31:
234–236.
22. Kozak, M. 1991. Structural features in eukaryotic mRNAs that modulate the
initiation of translation. J. Biol. Chem. 263:19867–19870.
23. Krug, E. C., J. J. Marr, and R. L. Berens. 1989. Purine metabolism in
Toxoplasma gondii. J. Biol. Chem. 264:10601–10607.
24. Martin, S. 2001. Congenital toxoplasmosis. Neonatal Netw. 20:23–30.
25. O’Gara, M. J., C. H. Lee, G. A. Weinberg, J. M. Nott, and S. F. Queener.
1997. IMP dehydrogenase from Pneumocystis carinii as a potential drug
target. Antimicrob. Agents Chemother. 41:40–48.
26. Roos, D. S., M. J. Crawford, R. G. Donald, L. M. Fohl, K. M. Hager, J. C.
Kissinger, M. G. Reynolds, B. Striepen, and W. J. Sullivan, Jr. 1999. Trans-
port and trafficking: Toxoplasma as a model for Plasmodium. Novartis
Found. Symp. 226:176–195.
35. Umejiego, N. N., C. Li, T. Riera, L. Hedstrom, and B. Striepen. 2004.
Cryptosporidium parvum IMP dehydrogenase: identification of functional,
structural, and dynamic properties that can be exploited for drug design.
J. Biol. Chem. 279:40320–40327.
36. Verham, R., T. D. Meek, L. Hedstrom, and C. C. Wang. 1987. Purification,
characterization, and kinetic analysis of inosine 5Ј-monophosphate dehydro-
genase of Tritrichomonas foetus. Mol. Biochem. Parasitol. 24:1–12.
37. Wagner, F. M., H. Reichenspurner, P. Uberfuhr, M. Weiss, V. Fingerle, and
B. Reichart. 1994. Toxoplasmosis after heart transplantation: diagnosis by
endomyocardial biopsy. J. Heart Lung Transplant. 13:916–918.
38. Wang, W., and L. Hedstrom. 1997. Kinetic mechanism of human inosine
5Ј-monophosphate dehydrogenase type II: random addition of substrates
and ordered release of products. Biochemistry 36:8479–8483.
39. Wong, S. Y., and J. S. Remington. 1993. Biology of Toxoplasma gondii. AIDS
7:299–316.
40. Zhang, R., G. Evans, F. Rotella, E. Westbrook, E. Huberman, A. Joachimiak,
and F. R. Collart. 1999. Differential signatures of bacterial and mammalian
IMP dehydrogenase enzymes. Curr. Med. Chem. 6:537–543.
41. Zhang, R., G. Evans, F. J. Rotella, E. M. Westbrook, D. Beno, E. Huberman,
A. Joachimiak, and F. R. Collart. 1999. Characteristics and crystal structure
of bacterial inosine-5Ј-monophosphate dehydrogenase. Biochemistry 38:
4691–4700.
27. Roos, D. S., R. G. Donald, N. S. Morrissette, and A. L. Moulton. 1994.
Molecular tools for genetic dissection of the protozoan parasite Toxoplasma
gondii. Methods Cell Biol. 45:27–63.
28. Roos, D. S., W. J. Sullivan, B. Striepen, W. Bohne, and R. G. Donald. 1997.
Tagging genes and trapping promoters in Toxoplasma gondii by insertional
mutagenesis. Methods 13:112–122.
42. Zhou, X., M. Cahoon, P. Rosa, and L. Hedstrom. 1997. Expression, purifi-
cation, and characterization of inosine 5Ј-monophosphate dehydrogenase
from Borrelia burgdorferi. J. Biol. Chem. 272:21977–21981.
29. Schwartzman, J. D., and E. R. Pfefferkorn. 1982. Toxoplasma gondii: purine