Toxoplasma gondii expresses two distinct lactate dehydrogenase homologous genes during its life cycle in intermediate hosts

Gene 184 (1997) 1–12

Toxoplasma gondii expresses two distinct lactate dehydrogenase homologous genes during its life cycle in intermediate hosts Shumin Yang 1,a,b, Stephen F. Parmley b,* a Department of Medicine, Division of Infectious Diseases and Geographic Medicine, Stanford University School of Medicine, Palo Alto, CA 94301, USA b Department of Immunology and Infectious Diseases, Research Institute, Palo Alto Medical Foundation, 860 Bryant Street, Palo Alto, CA 94301, USA Received 15 February 1996; revised 24 June 1996; accepted 25 June 1996; Received by A. Bernardi

Abstract Two Toxoplasma gondii genes were characterized that are differentially expressed during the parasite’s life cycle. The genes named LDH1 and LDH2, respectively, encode polypeptides similar to the enzyme lactate dehydrogenase (LDH; -lactate:NAD+ oxidoreductase, EC 1.1.1.27) from a variety of organisms. They show 64.0% nucleotide identity in the coding region and both have an intron at the same relative position. The deduced amino acid sequences of LDH1 and LDH2 share 71.1% identity. LDH1 and LDH2 are most similar to an LDH of Plasmodium falciparum (46.5% and 48.5% amino acid identities, respectively). The mRNA of LDH2 was only detected in the bradyzoite stage, while the mRNA of LDH1 was detected in both the bradyzoite and tachyzoite stages. However, by isoelectric focusing and immunoblot analysis, only one LDH isoform was found to be expressed in each stage. Furthermore, the expression of a reporter gene carrying chloramphenicol acetyltransferase (CAT ) coding sequence and the putative LDH2 promoter sequence was significantly up-regulated by growing parasites in tissue culture in media with alkaline pH (pH 8.2, a condition known to induce the expression of bradyzoite-specific antigens), while the expression of a CAT reporter construct carrying the putative LDH1 promoter sequence was down-regulated by similar treatment. These results indicate that LDH expression is developmentally regulated in T. gondii and suggest a possible correlation between stage conversion and alteration in carbohydrate or energy metabolism in this parasite. Keywords: Apicomplexan parasite; Bradyzoite; Tachyzoite; cDNA cloning; Developmental gene expression

1. Introduction The parasitic protozoan Toxoplasma gondii causes toxoplasmosis in a wide range of animals including humans (Dubey, 1993). Although it remains asymptomatic in individuals with normal immune systems, toxoplasmosis can cause severe diseases in fetuses of susceptible pregnant women and in individuals with compromised immune functions, such as patients undergoing immune suppressive treatments during organ * Corresponding author. Tel. +1 415 8534775; Fax +1 415 3299853; e-mail: [email protected] 1 Present address: Heska, 1825 Sharp Point Drive, Fort Collins, CO 80525, USA. Abbreviations: CAT, chloramphenicol acetyltransferase; LDH, lactate dehydrogenase; RACE, rapid amplification of cDNA ends; RT-PCR, reverse transcription and polymerase chain reaction; SAG1, surface antigen 1; TUB2, b-tubulin. 0378-1119/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved PII S 03 7 8 -1 1 1 9 ( 9 6 ) 0 0 5 66 - 5

transplantation and patients with immune suppressive diseases. Furthermore, toxoplasmosis has emerged as one of the leading opportunistic pathogens causing mortality in patients with AIDS (Luft and Remington, 1992). In the definitive feline host, the sexual stage of T. gondii occurs resulting in the development of the oocyst form in the intestine (Dubey, 1993). While in intermediate hosts such as humans and domestic animals, the parasite cycles between the tachyzoite form and the bradyzoite form. The rapidly dividing tachyzoite form predominates during acute infection, while the bradyzoite form predominates during latent (chronic) infection. Recrudescence of a latent infection resulting from conversion of bradyzoites to tachyzoites can cause toxoplasmic encephalitis and other life-threatening diseases in humans (Luft and Remington, 1992). T. gondii modifies its morphology and probably also its metabolism to adapt to the varying dwelling environments during its complex life cycle. The parasite

2

S. Yang, S.F. Parmley/Gene 184 (1997) 1–12

S. Yang, S.F. Parmley/Gene 184 (1997) 1–12

expresses stage-specific proteins, which are probably needed for these adaptations ( Kasper, 1989; Tomavo et al., 1991; Weiss et al., 1992). Although several genes encoding stage-specific antigens have been cloned, the nucleotide and deduced amino acid sequences of these genes have provided little insight into their functions or the mechanism of their differential gene expression during development. The mechanisms of stage conversion between T. gondii tachyzoites and bradyzoites are not known at the present time. Likewise, the factors regulating stage conversion are poorly defined. However, stage-specific gene products of the organism have been used as markers for studies of the kinetics of stage conversion using in vitro or in vivo models (Burg et al., 1988; Prince et al., 1990; Tomavo et al., 1991; Weiss et al., 1992, 1994; Gazzinelli et al., 1993; Bohne et al., 1993a,1994; Soete et al., 1993,Soete et al., 1994). Stage transformation is common to many parasitic protozoa including T. gondii, and is considered a complicated developmental process. Little is known about the factors driving stage conversion, but it most likely involves multiple factors of both parasite and host origin. Recent studies have shown that, in vitro, a small population of T. gondii organisms undergo spontaneous stage conversion from tachyzoites to bradyzoites ( Tomavo et al., 1991; Bohne et al., 1993b; Soete et al., 1993). The percentage of bradyzoites in this population can be increased if the conditions of the cultivation are modified, such as by altering the pH of the media, addition to the media of cytokines such as IFN-c, or treatment with mitochondria inhibitors such as antimycin A and oligomycin (Bohne et al., 1993a,1994; Soete et al., 1993,1994). The fact that mitochondrial inhibitors can induce the expression of bradyzoite-

3

specific antigens suggests that alterations in parasite mitochondrial functions might be correlated to stage conversion and may also explain, at least partly, how host responses to parasite infection can influence stage conversion in vivo. For example, nitric oxide, a product of activated macrophages, has been suggested to induce conversion to the bradyzoite stage, probably by interfering with the function of parasite electron transport system (Bohne et al., 1993a,1994). Additionally, sera from infected rabbits have been shown to induce the formation of bradyzoites and cysts in culture (Popiel et al., 1994). The reverse effect has been observed in infected animals treated with antibodies to cytokines such as IFN-c and TNF-a, which seem to induce the expression of tachyzoite specific genes, indicating that these cytokines may be important in preventing bradyzoites from transforming to tachyzoites (Suzuki et al., 1988; Suzuki and Remington, 1989; Gazzinelli et al., 1993). Therefore, it is likely that one way that the host attempts to control parasite proliferation is by interfering with parasite mitochondrial functions such as generation of energy. The parasite responds to the host reaction by transforming to a slower-replicating and less energy-consuming stage, the bradyzoite form, thus maintaining its viability for a long time assuming the host fails to clear the parasite from the tissues. Impairments of the host immune system that eliminate cytokines or other immune factors that normally interfere with parasite proliferation are exploited by the parasite which converts back to the tachyzoite stage. Understanding the developmental biology of the organism may be of significance for the development of new strategies for the prevention or treatment of toxoplasmosis. Such strategies may include identifying mole-

Fig. 1. Organization, partial restriction maps and alignment of the coding sequences of two T. gondii LDH genes. (A) Partial restriction maps of the 4.0-kb BamHI-SacI fragment of LDH1 and the 4.3-kb Bsp106-BamHI fragment of LDH2. Transcribed but not translated sequences at the 5∞ and 3∞ ends are indicated by boxes. Coding sequences are indicated by solid bars. Sequences flanking the transcribed regions are indicated by solid lines. The position of the intron is indicated. Abbreviations of restriction endonucleases are: BamHI (B), Bsp106I (Bp), EcoRI (RI ), EcoRV (RV ), HindIII (H ), SacI (Sc),and Sal (Sl ). (B) Alignment of the coding sequences of the two LDH genes. Gaps indicated by ‘^’ are introduced to maximize homology. Nucleotide differences between the two genes are indicated by ‘-’. In addition, nucleotide differences at the third position of codons that have no other differences are indicated by ‘*’. The sequences of the introns are not shown but their positions are indicated by an arrow and their nucleotide boundaries are indicated. The boundaries of the transcribed but not translated sequences flanking the open reading frame at 5∞ (5∞UTR) and 3∞ (3∞UTR) are indicated but the sequences are not shown. The nucleotide sequences for LDH1 and LDH2 mRNA can be retrieved from GenBank with accession Nos. U35118 and U23207, respectively. Methods: A 4.3-kb Bsp106I-BamHI fragment encompassing the entire LDH2 gene ( Yang and Parmley, 1995) was isolated from RH strain and was sequenced. The corresponding genomic sequence from the ME49 strain was cloned by PCR using primers derived from the LDH2 sequence of RH strain and the cloned DNA was sequenced. A cDNA fragment containing coding sequence for amino acids 206-326 of LDH2 ( Fig. 2) was cloned in-frame with glutathione S-transferase (GST ) in pGEX1N (Smith and Johnson, 1988) and was expressed in Escherichia coli. The fusion protein called GST-LDH2 was partially purified using glutathione agarose and used to immunize Swiss Webster mice. The resulting antibodies (LDH2CR) were found to cross-react with an antigen in tachyzoite lysates (data not shown). To clone the cDNA encoding LDH1, a cDNA fragment was isolated from a lgt11 expression library constructed from tachyzoites of RH strain (Burg et al., 1988) by screening with LDH2CR antibodies that had been affinity purified against the tachyzoite antigen on immunoblots. An LDH1 cDNA fragment overlapping with the above fragment was isolated by PCR of ME49 strain tachyzoite cDNA with heterologous primers derived from LDH2. Genomic clones of LDH1 were isolated from two DNA libraries, a lgt10 genomic library of RH strain and a lZap genomic library (kindly provided by Dr. John Boothroyd and NIH AIDS Research and Reference Reagent Program, respectively). The complete cDNA and the corresponding genomic DNA sequences of LDH1 from ME49 strain were isolated by PCR. The transcription start site and polyadenylation site were determined by amplification of cDNA ends (RACE) (Frohman et al., 1988). Cloned DNA was sequenced with Sequenase Version 2 kit ( United States Biochemical, Cleveland, OH ) and the sequences were analyzed using DNA Inspector software ( Textco, West Lebanon, NH ) and the Intelligenetics Suite program provided by the Program in Molecular Genetics and Medicine at Stanford University.

4

S. Yang, S.F. Parmley/Gene 184 (1997) 1–12

cules involved in stage conversion as possible targets for chemotherapy and vaccination. Toward that end, we have sought stage-specific genes (Parmley et al., 1994). In our previous work ( Yang and Parmley, 1995), we identified a bradyzoite stage-specific cDNA that encodes a polypeptide homologous to lactate dehydrogenase (LDH; -lactate:NAD+ oxidoreductase, EC 1.1.1.27). We report here the cloning and characterization of another T. gondii gene, which is very similar to our previously characterized LDH gene. The results of the present study show that two distinct LDH genes are differentially expressed in bradyzoites or tachyzoites, suggesting a possible correlation between stage conversion and alteration in carbohydrate metabolism or energy production.

2. Results and discussion 2.1. Cloning of two T. gondii LDH genes and comparison of their nucleotide sequences The T. gondii bradyzoite-specific cDNA clone Tgb9 which was previously shown to encode a polypeptide with homology to lactate dehydrogenase (named LDH2) was used as a probe to obtain large genomic fragments encompassing the entire gene ( Yang and Parmley, 1995). Tachyzoite cDNA clones encoding another isoform of LDH (named LDH1) were isolated by screening an expression library with LDH2 antibodies that crossreact with a tachyzoite antigen (LDH2CR) and by polymerase chain reaction (PCR) cloning with heterologous LDH2 primers. Genomic fragments encompassing the entire LDH1 gene were isolated by screening genomic DNA libraries and by PCR cloning. Fig. 1A shows the organization and partial restriction maps of a 4.0-kilobase (kb) genomic fragment flanked by a Bam HI site and a SacI site of LDH1 and a 4.3-kb Bsp 106IBam HI genomic fragment of LDH2. DNA fragments covering the complete mRNA of LDH1 and LDH2 were isolated from cDNA libraries and by rapid amplification of cDNA ends (RACE) cloning. The LDH1 mRNA is 1822 nt long with 143 nt of 5∞ nontranslated sequence and 692 nt of 3∞ nontranslated sequence. The LDH2 mRNA is 2571 nt with 246 nt of 5∞ nontranslated sequence and 1347 nt of 3∞ nontranslated sequence. The sequences of the protein coding regions of each gene are shown in Fig. 1B. The first nucleotide of the predicted translation start codon ATG was assigned nt 1 in each gene. The LDH1 transcript has an open reading frame of 987 nt, while the LDH2 transcript has an open reading frame of 978 nt. Both LDH1 and LDH2 have introns in their coding regions. The introns are found at the same relative position in the two genes, i.e. at nt 130 and 127 in LDH1 and LDH2, respectively ( Fig. 1B).

The intron in LDH1 is 538 nt and the intron in LDH2 is 588 nt. Very little sequence identity is found in the sequences of the introns and regions flanking the coding sequence between the two LDH genes of T. gondii. However, there is marked sequence identity in their open reading frame regions. The LDH1 genes from RH strain and ME49 strain of T. gondii were found to have few polymorphic nucleotides in the coding region (S. Yang and S. F. Parmley, unpublished ). Likewise, LDH2 from these strains had few differences. Since strainspecific polymorphisms in these two LDH genes are limited, only the sequences from the ME49 strain are reported here. The alignment of the coding sequences of LDH1 and LDH2 is shown in Fig. 1B. The two T. gondii LDH genes shared 64.0% of nucleotide identity in the coding region. Between the two genes, 131 codons had single mutations at the third (wobble) position of the codon. Of these, all but two were silent mutations. Consequently, the two genes share a larger percentage identity at the amino acid level than at the nucleotide level (see below). Results from genomic Southern blot studies suggest that both LDHs are single copy genes and are located in distinct regions of T. gondii genome (data not shown).

2.2. Comparison of the deduced amino acid sequences of the LDH genes When the deduced amino acids sequences of the LDH1 and LDH2 genes were aligned (Fig. 2) the amino acid identity was found to be 71.4% with 7.3% conservative substitutions amongst the differences. LDH1 encodes three more amino acids than LDH2. The predicted molecular masses of LDH1 and LDH2 are similar, at 35 550 and 35 342 DA, respectively. The predicted isoelectric points (pI’s) of LDH1 and LDH2 are 5.96 and 7.08, respectively. When compared to protein sequences in PIR and Swiss protein data bases, the amino acid sequences of the T. gondii LDHs were found to share from 23% to 35% identity with LDHs from a variety of species such as bacteria, plants, fishes, and mammals (Fig. 2). Among these organisms T. gondii LDH1 and LDH2 share the highest degree of amino acid identity, 33% and 35% respectively, with the LDH from the bacterium Thermotoga maritima. However, when the T. gondii LDHs were compared to the LDH from the related Apicomplexan protozoa Plasmodium falciparum (Bzik et al., 1993), LDH1 and LDH2 were found to share 46.5% and 48.5% amino acid identity, respectively ( Yang and Parmley, 1995). Furthermore, when the amino acid sequences of the two LDHs from T. gondii were aligned with LDHs from a variety of species, a pentapeptide insertion, which was also found in LDH of P. falciparum (Bzik et al., 1993), was present in both LDH1 and LDH2 (Fig. 2). The sequence of the

S. Yang, S.F. Parmley/Gene 184 (1997) 1–12

5

Fig. 2. The predicted amino acid sequences of T. gondii LDH1 and LDH2 and alignment of T. gondii LDHs to LDH of other organisms. LDH proteins aligned were Hv, Hordeum vulgare (barley; Hondred and Hanson, 1990); Sa, Squalus acanthius (dogfish; Taylor, 1977); Mm, Mus musculus (mouse; Li et al., 1983); TgLDH1, T. gondii LDH1 (this work); TgLDH2, T. gondii LDH2 ( Yang and Parmley, 1995); Pf, Plasmodium falciparum (Bzik et al., 1993); and Tm, Thermotoga maritima (Ostendorp et al., 1993). Gaps indicated by ‘-’ are introduced to maximize homology. The pentapeptide insertions are underlined. Conservative active center residues ( Eventoff et al., 1977; Dunn et al., 1991) indicated by ‘Cons.’ are included above the sequences of the LDHs.

6

S. Yang, S.F. Parmley/Gene 184 (1997) 1–12

insertion in LDH2 ( KSDKE) is identical to that in the P. falciparum LDH, while the sequence of the insertion in LDH1 ( KPDSE ) is similar but not identical.

position (52.1%). For example, among the 131 aligned codons that have only one nucleotide difference at the third position between LDH1 and LDH2, G or C is found at the third position in 74.8% of these codons in LDH1, while A or T is found at the third position in 66.4% of these codons in LDH2. Furthermore, in LDH1 all amino acid residues N (11), Q (9), P (14), Y (12), C (10), and F (8) are encoded by codons with either C or G at the third position. As with the five T. gondii genes previously studied (Johnson, 1990), LDH1 does not make use of codons TTA (L), CTA (L), TCA (S), and GGG (G). In addition, LDH1 does not make use of AAT(N ), ATA(I ), CAA(Q), TAT( Y ), TGT(C ), and TTT( F ). LDH2 does not make use of TCC(S) and CCC(P) which are codons frequently used by LDH1 and the five previously studied T. gondii genes. In the calculations of codon bias between LDH1 and LDH2 described above, the triplets ATT (I ), ATG (M ), and TGG ( W ) were excluded because there are no alterna-

2.3. Analysis of codon usage by the two LDH genes An early analysis of the coding sequences of five T. gondii proteins indicated that codon usage by these five parasite genes has a bias (67%) toward G or C in the third position and that these genes make little use of at least six codons ( TTA, CTA, TCA, CGA, CGG, GGG) (Johnson, 1990). The amino acid distribution of the two LDHs of T. gondii are similar to each other, however the codon usage by the two LDH genes are very different ( Table 1). Although the G+C base composition in the coding sequence of LDH1 is 59.0%, LDH1 has a bias (80.8%) toward C or G in the third position. The G+C content of LDH2 coding region is 49.0%, and LDH2 does not have significant bias toward G+C in the third Table 1 Codon frequency of the two LDH genes Codon

AAA AAT ACA ACT AGA AGT ATA ATT CAA CAT CCA CCT CGA CGT CTA CTT GAA GAT GCA GCT GGA GGT GTA GTT TAA TAT TCA TCT TGA TGT TTA TTT Third position of codons A+T

AA

Lys Asn Thr Thr Arg Ser Ile Ile Gln His Pro Pro Arg Arg Leu Leu Glu Asp Ala Ala Gly Gly Val Val Ter Tyr Ser Ser Ter Cys Leu Phe

LHD1

Codon

LHD2

No.

%

No.

%

2 0 2 1 1 1 0 6 0 1 0 0 2 2 0 4 1 2 4 9 2 10 1 7 1 0 0 7 0 0 0

9.5 0.0 13.3 6.7 7.1 5.6 0.0 28.6 0.0 33.3 0.0 0.0 14.3 14.3 0.0 16.7 5.0 11.8 12.9 29.0 7.1 35.7 2.6 18.4 100.0 0.0 0.0 38.9 0.0 0.0 0.0

9 3 7 3 6 4 6 10 3 2 4 2 4 2 2 8 6 8 9 6 7 8 9 6 5 4 2 1 2 5 4

40.9 37.5 41.2 17.6 40.0 22.2 27.3 45.5 30.0 40.0 33.3 16.7 26.7 13.3 6.9 27.6 28.6 47.1 31.0 20.7 23.3 26.7 29.0 19.4 62.5 22.2 11.1 100.0 25.0 17.2 36.4

AAG AAC ACC ACG AGG AGC ATC ATG CAG CAC CCC CCG CGC CGG CTC CTG GAG GAC GCC GCG GGC GGG GTC GTG TAG TAC TCC TCG TGG TGC TTG TTC

59

19.2

145

47.9

C+G

AA

Lys Asn Thr Thr Arg Ser Ile Met Gln His Pro Pro Arg Arg Leu Leu Glu Asp Ala Ala Gly Gly Val Val Ter Tyr Ser Ser Trp Cys Leu Phe

LHD1

LHD2

No.

%

No.

%

19 11 11 1 1 2 15 14 9 2 6 8 7 1 11 7 19 15 12 6 16 0 21 9 12 7 1 1 10 2 8

90.5 100.0 73.3 6.7 7.1 11.1 71.4 100.0 100.0 66.7 42.9 57.1 50.0 7.1 45.8 29.2 95.0 88.2 38.7 19.4 57.1 0.0 55.3 23.7 100.0 38.9 5.6 100.0 100.0 8.3 100.0

13 5 3 4 2 6 6 12 7 3 0 6 1 1 2 8 5 9 4 10 5 10 3 13 3 0 2 1 6 4 7

59.1 62.5 17.6 23.5 6.7 33.3 27.3 100.0 70.0 60.0 0.0 50.0 6.7 6.7 6.9 27.6 71.4 52.9 13.8 34.5 16.7 33.3 9.7 41.9 37.5 0.0 11.1 100.0 75.0 13.8 63.6

248

80.8

158

52.1

S. Yang, S.F. Parmley/Gene 184 (1997) 1–12

7

tive codons for these amino acids. In addition, the stop codons TAA, TAG and TGA were excluded. The implication, if any, of the obvious differential codon usage by the two LDH genes is not clear at the present time. 2.4. mRNA expression of T. gondii LDH genes In Northern blot hybridization, the LDH1 probe hybridized to a 2.0 kb tachyzoite messenger RNA species (Fig. 3, LDH1). The LDH2 probe did not detect any signal on a blot containing tachyzoite total RNA. Under similar conditions, each LDH probe hybridized to the homologous gene but did not cross hybridize to the heterologous LDH genes in T. gondii genomic Southern blots (data not shown). A cDNA probe from the tachyzoite-specific gene SAG1 hybridized to bands with similar intensities on the two blots, indicating similar amounts of total RNA was present. By reverse transcription and PCR (RT-PCR), the mRNA of LDH2 was only detected in the bradyzoite stage (Fig. 4, LDH2), while the mRNA of LDH1 was detected in both stages (Fig. 4, LDH1). Quantification of the differences in the levels of LDH1 mRNA expression in tachyzoites and bradyzoites was not attempted in the present study. As expected, mRNA of the SAG1 gene was only detected in the tachyzoite stage ( Fig. 4, SAG1), while mRNA of the constitutively expressed b-tubulin gene (TUB2) was detected in both stages (Fig. 4, TUB2). The cDNA samples were free of contaminating genomic DNA, since the sizes of the PCR products amplified with primers from LDH1, LDH2, and TUB2 were consistent with cDNA and not genomic DNA. The absence of LDH2 mRNA in the tachyzoite stage suggests that transcription of LDH2 is suppressed in the developmental transition from the bradyzoite to the tachyzoite stage and suggests that LDH1 is the only LDH produced in the tachyzoite stage. Preliminary RT-PCR data indicates that the LDH2 mRNA levels increase markedly when T. gondii tachyzoites are induced to switch to the bradyzoite stage by incubation in media with alkaline pH (pH 8.2) or in media containing mitochondrial inhibitors, antimycin A and oligomycin (data not shown). Since LDH2 mRNA stability was not determined in the present study it is not known whether this is due to an increase in LDH2 mRNA stability or transcriptional activation of the LDH2 gene. 2.5. LDH antigen detection in bradyzoites and tachyzoites Since both LDH1 and LDH2 mRNAs are present in bradyzoites it is possible that both LDH1 and LDH2 polypeptides are produced in this stage. However, it is unknown whether both LDH1 and LDH2 transcripts are translated. Therefore, the developmental expression of the native polypeptides encoded by each LDH tran-

Fig. 3. Determination by Northern blot hybridization of mRNA expression of LDH genes in tachyzoites. (A) Northern blots probed with LDH1 and LDH2. (B) Northern blots probed with SAG1. Methods: Tachyzoites of the ME49 strain of T. gondii were purified (Suzuki and Remington, 1989) and total RNA was prepared from the organisms as described previously (Chomczynski and Sacchi, 1987). Total RNA was separated in a formaldehyde-denaturing agarose gel (1.1%) and transferred to nitrocellulose (Sambrook et al., 1989). Each blot contained 6.5 mg of total RNA and was hybridized with 32P-labeled cloned cDNA fragments of LDH1 (nt -2 to 1700) or LDH2 (nt -2 to 1791) ( Fig. 1). The blots were washed at 42°C in solution containing 0.1% sodium dodecyl sulfate (SDS ) and decreasing concentrations of SSC (2×, 1×, 0.4× and 0.2×), each for 20 min. The size of LDH1 mRNA was estimated in relation to 1 kilobase ladder (BRL, Gaithersburg, MD). To control for the amounts of total RNA used, the LDH1 and LDH2 probes were stripped from the above blots (A) by boiling in 0.01×SSC and 0.1% SDS three times, each for 15 min and then hybridized with a 32P-labeled DNA fragment (PstI-BamHI ) from SAG1 (B), an abundant tachyzoite-specific transcript (Burg et al., 1988).

8

S. Yang, S.F. Parmley/Gene 184 (1997) 1–12

Fig. 4. Determination by RT-PCR of mRNA expression of T. gondii LDH genes in the tachyzoite and bradyzoite stages. Dilutions of cDNA from approximately equal numbers of tachyzoites or bradyzoites were amplified by PCR with primers specific for the LDH1, LDH2, SAG1, and TUB2 genes. Similar amounts of RNA corresponding to that used for cDNA synthesis were included as negative controls. Genomic DNA was included as positive controls. The locations of the amplified cDNA and genomic DNA fragments for each gene are indicated by arrows. Methods: Bradyzoite RNA was isolated from cysts of the ME49 strain purified from brains of chronically infected CBA/Ca mice by percoll gradient centrifugation (Parmley et al., 1994). Because, the quantity of readily obtainable bradyzoite RNA was insufficient for Northern blot hybridization, the bradyzoite mRNA was analyzed by RT-PCR. After treatment with DNase I to remove potential contaminating genomic DNA, RNA samples isolated from equal numbers of bradyzoites and tachyzoites were treated with AMV reverse transcriptase (Promega, Madison, WI ) to generate cDNA. The first strand cDNAs were used as templates for amplification with AmpliTaq DNA polymerase (Perkin-Elmer, Norwalk, CT ) and primers from four different genes. The sequence (5∞ to 3∞) of the primer and the size (in bp) of cDNA fragment and genomic DNA fragment for each gene were as follows: LDH1 (LDH1a: TTTGCACGTGTACGCAAG and LDH1b: CTTGATGTTCTGACCGAT; 431; 969); LDH2 (LDH2a: AGGGCACAGAAGACAATC and LDH2b: TGTGCCGATAACAGTAGC; 784; 1372); SAG1 (Burg et al., 1988) (SAG1a: ATGTCGGTTTCGCTGCACCACTTC and SAG1b: TCACGCGACACAAGCTGCGATAGAGCC; 1010; 1010); and TUB2: (Nagel and Boothroyd, 1988) ( TUB2a: CCAGCGTCTGTGACATCC and TUB2b: CCCATCTCGCCCTCTTCC; 282; 512). Aliquots of total RNA or genomic DNA isolated from tachyzoites were used as control templates. PCR was performed for 35 cycles (94°C, 30 s, 58°C, 30 s, and 72°C, 90 s), the products were separated in 1.2% agarose gels, and analyzed by Southern blot hybridization using 32P-labeled probes that hybridize to regions wholly internal to the amplification primers. Probes used were: LDH1 (LDH1c: AAATGGCACCCGCACTTG);

script was determined. By two-dimensional (2-D) gel electrophoresis and immunoblotting of cyst or tachyzoite lysate, only one antigen band was detected in each lysate by the LDH2CR cross-reactive antibodies. The approximate pI and molecular mass of the bradyzoite antigen were 7.0 and 35 kDa, respectively. The approximate pI and molecular mass of the tachyzoite antigen were 6.0 and 33 kDa, respectively (Fig. 5). There was no overlap in the relative positions of each antigen in the bradyzoite blot versus the tachyzoite blot. The pI and size of the bradyzoite antigen is similar to that of the predicted pI of LDH2 (7.08), while the pI of the tachyzoite antigen is similar to that of the predicted pI of LDH1 (5.96). The observed higher pI of the bradyzoite LDH could not be due to tyrosine phosphorylation (Cooper et al., 1983) of the tachyzoite LDH, since such modification would make the tachyzoite LDH isoform more acidic. Post-translational modifications that make LDH more basic have not been reported. Therefore, the bradyzoite and tachyzoite antigens detected with the LDH2CR antibodies most likely represent separate polypeptides. These results suggest that different forms of LDH protein are produced in the bradyzoite and tachyzoite stages. The form exclusive to the tachyzoite stage seems to be LDH1, while the form exclusive to the bradyzoite stage seems to be LDH2. However, their identity needs further clarification, such as by amino acid sequence determination or the use of LDH1 and LDH2 isoformspecific antibodies, which are not yet available. The conclusion that the LDH1 polypeptide is the only LDH produced in tachyzoites is consistent with the tachyzoite RNA results (Fig. 3). If LDH2 is the only LDH produced in bradyzoites, then the LDH1 mRNA detected in bradyzoites (Fig. 4) must somehow be prevented from translation. Thus, the mechanisms of developmental regulation of these two genes are most likely different. Production of LDH2 seems to be under transcriptional control, while production of LDH1 seems to be under some form of translational control. Parasites isolated from sarcoma cells grown in the peritoneum of mice and human foreskin fibroblasts grown in vitro both express an LDH with a pI close to 6.0, which is most likely LDH1. Parasites isolated from human foreskin fibroblasts grown in vitro express an additional LDH with a pI close to 6.9, which is most likely LDH2 and is probably expressed by a subpopulation of parasites that have spontaneously switched to bradyzoites under the cultivation conditions (Darde et al., 1990). Although the parasites used in these studies were assumed to be tachyzoites, subsequent studies have

LDH2 (LDH2c: CCATGACGGGTACCGTTAGC ); and TUB2 ( TUB2c: TCCTCTGCGGTGGCGTCC ). The SAG1 probe was the same as that used for Northern hybridization (Fig. 3).

S. Yang, S.F. Parmley/Gene 184 (1997) 1–12

9

Fig. 5. Analysis of T. gondii LDH antigens by two-dimensional gel electrophoresis and immunoblot. The cross-reactive LDH2CR antibodies (see legend to Fig. 1) were used to identify the native protein products of the two T. gondii LDH genes in lysates from tachyzoites (‘Tachyzoites’ Panel ) or cysts (‘Cysts’ Panel ) by two-dimensional gel electrophoresis and immunoblot. The pI and molecular weight of the markers (Bio-Rad) are indicated. A reference point that corresponds to molecular weight 31.4 and pI 6.3 is indicated by ‘+’ on both blots. Methods: Lysates from 5×106 tachyzoites or 1×106 bradyzoites in cysts of the ME49 strain were separated by two-dimensional gel electrophoresis as previously described (O’Farrell, 1975). After isoelectric focusing, cyst or tachyzoite proteins were separated through polyacrylamide gel (SDS-PAGE ) and transferred to nitrocellulose membrane. Blots were incubated with LDH2CR antibodies and developed with goat anti-mouse IgG (1:1000 dilution) followed by substrate (diaminobenzidine and hydrogen peroxide).

confirmed that in vitro tachyzoite cultures contain a small population of organisms that have undergone spontaneous stage conversion to bradyzoites ( Tomavo et al., 1991; Bohne et al., 1993b; Soete et al., 1993). 2.6. Expression of reporter gene carrying putative promoter sequences of LDH genes Transient transfection was used to further investigate the stage differential expression of LDH1 and LDH2 using CAT reporter constructs (Fig. 6). Parasites transfected with the vector only (pCAT ) did not express appreciable level of CAT activity. But parasites transfected with either pL1CL1 or pL2CL2 expressed significant levels of CAT activity under appropriate conditions. The expression of each CAT construct in transfected tachyzoites was affected differently by treat-

ment with alkaline media (pH 8.2) to induce conversion to the bradyzoite stage. Data from a representative experiment is shown in Fig. 6. While CAT expression from pL1CL1 was reduced by approximately 4 fold by treatment with alkaline media, CAT expression from pL2CL2 was induced from the residual, background expression in neutral media by more than 45 fold by treatment with alkaline media. The reduced CAT expression from pL1CL1 in alkaline media (Fig. 6) and the detection of similar LDH1 mRNA levels by RT-PCR in tachyzoites and bradyzoites (Fig. 4) suggests that regulation of expression of LDH1 occurs through a post-transcriptional mechanism. Experiments to identify the DNA sequences containing minimal promoter activities of the two genes are in progress. The CAT activities from the LDH1 and LDH2 constructs during in vitro stage conversion are consistent

10

S. Yang, S.F. Parmley/Gene 184 (1997) 1–12

Fig. 6. Transient transfection of CAT reporter constructs carrying putative promoter sequences of LDH1 and LDH2. The restriction maps of the wild-type CAT plasmid (pCAT ) or CAT constructs containing flanking sequences from LDH1 (pL1CL1) or LDH2 (pL2CL2) are shown adjacent to the graph with their corresponding CAT activities following transfection and cultivation in neutral (pH 7.4) or alkaline (pH 8.2) media. Restriction endonucleases used are: BamHI (B), Bsp106I (Bp), HindIII (H ), NsiI (N ), and PacI (P). Methods: pBluescript plasmids carrying CAT coding sequence only (pCAT; Soldati and Boothroyd, 1993), CAT coding sequence carrying 1436 bp of 5∞ and 1489 bp of 3∞ sequences flanking the translation start and stop codons of LDH1 (pL1CL1), or 993 bp of 5∞ and 1750 bp of 3∞ sequences of LDH2 (pL2CL2) were constructed. Transfection was performed as previously described (Soldati and Boothroyd, 1993). Briefly, 50 mg of plasmid DNA was electroporated into tachyzoites of the ME49 strain grown in tissue culture of human foreskin fibroblasts. The transfected organisms were divided equally into two T25 flasks of human foreskin fibroblasts at allowed to infect for 6 h. Expression of bradyzoite stage-specific genes was induced by shifting the pH of the culture media from 7.4 to 8.2, as described previously (Soete et al., 1994; Weiss et al., 1994). The media in one flask was replaced with pH 7.4 media and incubated in 37°C incubator with 5% CO . The media in the second flask was replaced with pH 8.2 media and incubated in 37°C 2 incubator without CO . 48 h after switching media, the CAT activity in each flask was determined by assaying of an aliquot of the extract of 2 transfected parasites (Fig. 5B). Since treatment with alkaline media reduces parasite replication while simultaneously inducing conversion to bradyzoites, a co-transfection control b-Gal reporter construct (TUB1-b-Gal ) containing promoter sequence of the constitutively expressed a-tubulin gene of T. gondii was included in each transfection (Seeber and Boothroyd, 1995). The b-gal activities were determined in extracts from each flask. The b-gal activities were used to standardize the CAT activities of transfectants cultivated in alkaline media to those in neutral media. The wildtype CAT plasmid (pCAT ) was used as negative control.

with the two-dimensional immunoblot results ( Fig. 5) of bradyzoites and tachyzoites obtained in vivo. Together these results strongly suggest that LDH1 is replaced by LDH2 during development from tachyzoites to bradyzoites. LDH is a physiologically significant enzyme for metabolism and possibly for regulation of gene expression in a variety of organisms (Markert et al., 1975; Cooper et al., 1983). Although they all catalyze the interconversion between pyruvate and lactate, different LDH isoforms do not have the same kinetics and properties. The expression of LDH genes can be regulated both tissue-specifically and developmentally in organisms such as vertebrates (Markert et al., 1975; Li, 1990). Some fish express different forms of LDH with changing environments (Powers et al., 1991). LDH plays an indispensable role when glycolysis becomes the only pathway by which the cell can obtain energy such as when the function of mitochondria is impaired in the presence of inhibitors or under anaerobic conditions. Although early studies using isolated parasites indicated that phosphorylating glycolysis is an important pathway by which T. gondii obtains energy (Fulton and Spooner, 1960), functional mitochondria seem to exist in this parasite (Melo et al., 1992; Pfefferkorn and Borotz, 1994). It is possible that, although tachyzoites can utilize both glycolysis and

oxidative phosphorylation to obtain energy, glycolysis may become the predominant pathway for the parasite to obtain energy during the bradyzoite stage (where there is an abundant storage of polysaccharides in the form of amylopectin granules) since the bradyzoite needs less energy and the activity of the parasite mitochondria is down-regulated either by the parasite itself or by host responses to parasite infection. It is tempting, therefore, to speculate that T. gondii LDH1 and LDH2 may have different characteristics which are required for the parasite to adapt to the particular microenvironment and metabolic status of each stage of its life cycle. Therefore, characterization of the differences of enzymatic properties between the two LDH isoforms would provide further insight into the role that LDHs may play in energy metabolism of the parasite and the relationship between energy production and stage conversion. Substantial sequence differences exist between T. gondii LDHs and mammalian LDHs. In addition, the pentapeptide insertion, which is predicted to be inserted into the active site loop of the enzyme (Grau et al., 1981; Dunn et al., 1991), is unique to LDHs from T. gondii as well as from P. falciparum. These differences suggest that T. gondii LDHs may have unique properties, which could be exploited for design of therapeutic agents that act by specifically inhibiting parasite enzymes. Indeed,

S. Yang, S.F. Parmley/Gene 184 (1997) 1–12

recent studies with phosphofructokinase of T. gondii have shown that it is possible to use inhibitors targeted specifically to the enzymes involved in carbohydrate or energy metabolism of the parasite as potential drugs for the treatment of toxoplasmosis (Peng et al., 1995).

Acknowledgement We thank Drs. Jack S. Remington, John C. Boothroyd, Dennis A. Powers for critical review of this work. This work was supported by grants from the NRI Competitive Grant Program/United States Department of Agriculture grant 9102189, NIH Public Health Service grant AI04717, and the John D. and Catherine T. MacArthur Foundation. We thank the Stanford University Digestive Diseases Oligonucleotide Synthesis Core facility for preparation of primers.

References Bohne W., Heesemann, J. and Gross U. (1993a) Induction of bradyzoite-specific Toxoplasma gondii antigens in gamma interferontreated mouse macrophages. Infect. Immun. 61, 1141-1145. Bohne, W., Heesemann, J. and Gross, U. (1993b) Coexistence of heterogeneous populations of Toxoplasma gondii parasites within parasitophorous vacuoles of murine macrophages as revealed by a bradyzoite-specific monoclonal antibody. Parasitol. Res. 79, 485-487. Bohne, W., Heesemann, J. and Gross, U. (1994) Reduced replication of Toxoplasma gondii is necessary for induction of bradyzoite-specific antigens: a possible role for nitric oxide in triggering stage conversion. Infect. Immun. 62, 1761-1767. Burg, J.L., Perelman, D., Kasper, L.H., Ware, P.L. and Boothroyd J.C. (1988) Molecular analysis of the gene encoding the major surface antigen of Toxoplasma gondii. J. Immunol. 141, 3584-3591. Bzik, D.J., Fox, B.A. and Gonyer, K. (1993) Expression of Plasmodium falciparum lactate dehydrogenase in Escherichia coli. Mol. Biochem. Parasitol. 59, 155-166. Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156-159. Cooper, J.A., Reiss, N.A., Schwartz, R.J. and Hunter, T. (1983) Three glycolytic enzymes are phosphorylated at tyrosine in cells transformed by Rous sarcoma virus. Nature 302, 218-223. Darde, M.L., Bouteille, B. and Pestre-Alexandre, M. (1990) Comparison of isoenzyme profiles of Toxoplasma gondii tachyzoites produced under different culture conditions. Parasitol. Res. 76, 367-371. Dubey, J.P. (1993) Toxoplasma, Neospora, Sarcocystis, and other tissue cyst-forming coccidia of humans and animals. In: Kreier, J.P. ( Ed.), Parasitic Protozoa 6, 2nd edition. Academic Press, San Diego, CA, pp. 1-158. Dunn, C.R., Wilks, H.M., Halsall, D.J., Atkinson, T., Clarke, A.R., Muirhead, H. and Holbrook, J.J. (1991) Design and synthesis of new enzymes based on the lactate dehydrogenase framework. Phil. Trans. R. Soc. London B 332, 177-184. Eventoff, W., Rossmann, M.G., Taylor, S.S., Torff, H.-J., Meyer, H., Keil, W. and Kiltz, H.-H. (1977) Structural adaptations of lactate dehydrogenase isozymes. Proc. Natl. Acad. Sci. USA 74, 2677-2681. Frohman, M.A., Dush, M.K. and Martin, G.R. (1988) Rapid production of full-length cDNAs from rare transcripts: Amplification

11

using a single gene-specific oligonucleotide primer. Proc. Natl. Acad. Sci. USA 85, 8998-9002. Fulton, J.D. and Spooner, D.F. (1960) Metabolic studies on Toxoplasma gondii. Exp. Parasitol. 9, 293-301. Gazzinelli, R.T., Eltoum, I., Wynn, T.A. and Sher, A. (1993) Acute cerebral toxoplasmosis is induced by in vivo neutralization of TNF-a and correlates with the down-regulated expression of inducible nitric oxide synthase and other markers of macrophage activation. J. Immunol. 151, 3672-3681. Grau, U.M., Trommer, W.E. and Rossmann, M.G. (1981) Structure of the active ternary complex of pig heart lactate dehydrogenase ˚ resolution. J. Mol. Biol. 151, 289-307. with S-lac-NAD at 2.7 A Hondred, D. and Hanson, A.D. (1990) Hypoxically inducible barley lactate dehydrogenase: cDNA cloning and molecular analysis. Proc. Natl. Acad. Sci. USA 87, 7300-7304. Johnson, A.M. (1990) Comparison of dinucleotide frequency and codon usage in Toxoplasma gondii and Plasmodium falcuparum: Evolutionary implications. J. Mol. Evol. 30, 383-387. Kasper, L.H. (1989) Identification of stage-specific antigens of Toxoplasma gondii. Infect. Immun. 57, 668-672. Li, S.S., Fitch, W.M., Pan, Y.E. and Sharief, F.S. (1983) Evolutionary relationship of vertebrate lactate dehydrogenase isozymes A 4 (muscle), B (heart), and C (testis). J. Biol. Chem. 258, 7029-7033. 4 4 Li, S.S.-L (1990) Human and mouse lactate dehydrogenase genes A (muscle), B (heart), and C (testis): Protein structure, genomic organization, regulation of expression, and molecular evolution. In: Ogita, I.I. and Markert, C.L. ( Eds.), Isozymes: Structure, Function, and Use in Biology and Medicine. Wiley-Liss, New York, pp. 75-99. Luft, B.J. and Remington, J.S. (1992) Toxoplasmic encephalitis in AIDS. Clin. Inf. Dis. 15, 211-222. Markert, C.L., Shaklee, J.B. and Whitt, G.S. (1975) Evolution of a gene. Science 189, 102-114. Melo, E.J.T., Carvalho, T.U. and de Souza, W. (1992) Penetration of Toxoplasma gondii into host cells induces changes in the distribution of the mitochondria and the endoplasmic reticulum. Cell Struct. Function 17, 311-317. Nagel, S.D. and Boothroyd, J.C. (1988) The a- and b-tubulins of Toxoplasma gondii are encoded by single copy genes containing multiple introns. Mol. Biochem. Parasitol. 29, 261-273. O’Farrell, P.H. (1975) High-resolution two-dimensional gel electrophoresis of proteins. J. Biol. Chem. 250, 4007-4021 Ostendorp, R., Liebl, W., Schurig, H. and Jaenicke, R. (1993) The L-lactate dehydrogenase gene of the hyperthermophilic bacterium Thermotoga maritima cloned by complementation in Escherichia coli. Eur. J. Biochem. 216, 709-715. Parmley, S.F., Yang, S., Harth,G., Sibley, L.D., Sucharczuk, A. and Remington, J.S. (1994) Molecular characterization of a 65-kilodalton Toxoplasma gondii antigen expressed abundantly in the matrix of tissue cysts. Mol. Biochem. Parasitol. 66, 283-296. Peng, Z.-Y., Mansour, J.M., Araujo, F.G., Ju J.-Y., McKenna, C.E. and Mansour, T.E. (1995) Some phosphonic acid analogs as inhibitors of pyrophosphate-dependent phosphofructokinase, a novel target in Toxoplasma gondii. Biochem. Pharmacol. 49, 105-113. Pfefferkorn, E.R. and Borotz, S.E. (1994) Comparison of mutants of Toxoplasma gondii selected for resistance to azithromycin, spiramycin, or clindamycin. Antimicrob. Agents Chemother. 38, 31-37. Popiel, I., Gold, M. and Choromanski, L. (1994) Tissue cyst formation of Toxoplasma gondii T263 in cell culture. J. Euk. Microbiol. 41, 17S. Powers, D.A., Lauerman, T., Crawford, D. and DeMichele, L. (1991) Genetic mechanisms for adapting to a changing environment. Annu. Rev. Genet. 25, 629-659. Prince, J.B., Auer, K.L., Huskinson, J., Parmley, S.F., Araujo, F.G. and Remington, J.S. (1990) Cloning, expression, and cDNA sequence of surface antigen P22 from Toxoplasma gondii. Mol. Biochem. Parasitol. 43, 97-106. Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Clon-

12

S. Yang, S.F. Parmley/Gene 184 (1997) 1–12

ing. A Laboratory Manual, 2nd edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Seeber, F. and Boothroyd, J.C. (1995) Escherichia coli b-galactosidase as an in vitro and in vivo reporter enzyme and stable transfection marker in the intracellular protozoan parasite Toxoplasma gondii. Gene 169, 39–45. Smith, D.B. and Johnson, K.S. (1988) Single-step purification of polypeptides expressed in Escherichia coli as fusions with glutathione Stransferase. Gene 67, 31-40. Soete, M., Fortier, B., Camus, D. and Dubremetz, J.F. (1993) Toxoplasma gondii. Kinetics of bradyzoite-tachyzoite interconversion in vitro. Exp. Parasitol. 76, 259-264. Soete M., Camus, D. and Dubremetz, J.F. (1994) Experimental induction of bradyzoite-specific antigen expression and cyst formation by the RH strain of Toxoplasma gondii in vitro. Exp. Parasitol. 78, 361-370. Soldati, D. and Boothroyd, J.C. (1993) Transient transfection and expression in the obligate intracellular parasite Toxoplasma gondii. Science 260, 349-352. Suzuki, Y., Orellana, M.A., Schreiber, R.D. and Remington, J.S.

(1988) Interferon-c: the major mediator of resistance against Toxoplasma gondii. Science 240, 516-518. Suzuki, Y. and Remington, J.S. (1989) A method for obtaining large numbers of trophozoites of avirulent strains of Toxoplasma gondii using an antibody to interferon-gamma. J. Parasitol. 75, 174-176. Taylor, S.S. (1977) Amino acid sequence of dogfish muscle lactate dehydrogease. J. Biol. Chem. 252, 1799-1806. Tomavo, S., Fortier, B., Soete, M., Ansel, C., Camus D. and Dubremetz, J.F. (1991) Characterization of bradyzoite-specific antigens of Toxoplasma gondii. Infect. Immun. 59, 3750-3753. Weiss, L.M., LaPlace, D., Tanowitz, H.B. and Wittner, M. (1992) Identification of Toxoplasma gondii bradyzoite-specific monoclonal antibodies. J. Inf. Dis. 166, 213-215. Weiss, L.M., LaPlace, D., Takvorjan, P.M., Cali, A., Tanowitz, B. and Wittner, M. (1994) Development of bradyzoites of Toxoplasma gondii in vitro. J. Euk. Microbiol. 41, 18S. Yang, S. and Parmley, S.F. (1995) A bradyzoite stage-specifically expressed gene of Toxoplasma gondii encodes a polypeptide homologous to lactate dehydrogenase. Mol. Biochem. Parasitol. 73, 291-294.